CN117936101A - Method and system for intelligently detecting thyroid-related eye diseases - Google Patents
Method and system for intelligently detecting thyroid-related eye diseases Download PDFInfo
- Publication number
- CN117936101A CN117936101A CN202410324037.3A CN202410324037A CN117936101A CN 117936101 A CN117936101 A CN 117936101A CN 202410324037 A CN202410324037 A CN 202410324037A CN 117936101 A CN117936101 A CN 117936101A
- Authority
- CN
- China
- Prior art keywords
- patient
- score
- eye
- value
- lacrimal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 210000001685 thyroid gland Anatomy 0.000 title claims abstract description 83
- 238000000034 method Methods 0.000 title claims abstract description 80
- 208000030533 eye disease Diseases 0.000 title claims abstract description 75
- 238000001514 detection method Methods 0.000 claims abstract description 143
- 238000004364 calculation method Methods 0.000 claims abstract description 78
- 230000000694 effects Effects 0.000 claims abstract description 45
- 206010020850 Hyperthyroidism Diseases 0.000 claims abstract description 26
- 239000002131 composite material Substances 0.000 claims abstract description 14
- 210000001508 eye Anatomy 0.000 claims description 304
- 210000004561 lacrimal apparatus Anatomy 0.000 claims description 130
- 210000003205 muscle Anatomy 0.000 claims description 119
- 230000006870 function Effects 0.000 claims description 40
- 238000002310 reflectometry Methods 0.000 claims description 38
- 210000005252 bulbus oculi Anatomy 0.000 claims description 33
- 102000016289 Cell Adhesion Molecules Human genes 0.000 claims description 25
- 108010067225 Cell Adhesion Molecules Proteins 0.000 claims description 25
- 230000011218 segmentation Effects 0.000 claims description 25
- 210000004087 cornea Anatomy 0.000 claims description 23
- 230000004957 immunoregulator effect Effects 0.000 claims description 22
- 102000003777 Interleukin-1 beta Human genes 0.000 claims description 21
- 108090000193 Interleukin-1 beta Proteins 0.000 claims description 21
- 102000001776 Matrix metalloproteinase-9 Human genes 0.000 claims description 21
- 108010015302 Matrix metalloproteinase-9 Proteins 0.000 claims description 21
- 108060008682 Tumor Necrosis Factor Proteins 0.000 claims description 21
- 102000000852 Tumor Necrosis Factor-alpha Human genes 0.000 claims description 21
- 210000004279 orbit Anatomy 0.000 claims description 21
- LOGFVTREOLYCPF-KXNHARMFSA-N (2s,3r)-2-[[(2r)-1-[(2s)-2,6-diaminohexanoyl]pyrrolidine-2-carbonyl]amino]-3-hydroxybutanoic acid Chemical compound C[C@@H](O)[C@@H](C(O)=O)NC(=O)[C@H]1CCCN1C(=O)[C@@H](N)CCCCN LOGFVTREOLYCPF-KXNHARMFSA-N 0.000 claims description 20
- KZMAWJRXKGLWGS-UHFFFAOYSA-N 2-chloro-n-[4-(4-methoxyphenyl)-1,3-thiazol-2-yl]-n-(3-methoxypropyl)acetamide Chemical compound S1C(N(C(=O)CCl)CCCOC)=NC(C=2C=CC(OC)=CC=2)=C1 KZMAWJRXKGLWGS-UHFFFAOYSA-N 0.000 claims description 20
- 239000008280 blood Substances 0.000 claims description 14
- 210000004369 blood Anatomy 0.000 claims description 14
- 238000011156 evaluation Methods 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 14
- 102000004127 Cytokines Human genes 0.000 claims description 12
- 108090000695 Cytokines Proteins 0.000 claims description 12
- 238000003062 neural network model Methods 0.000 claims description 12
- 230000000770 proinflammatory effect Effects 0.000 claims description 10
- 230000004927 fusion Effects 0.000 claims description 9
- 239000012530 fluid Substances 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 238000002591 computed tomography Methods 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 claims description 4
- 238000002603 single-photon emission computed tomography Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 3
- 238000003745 diagnosis Methods 0.000 abstract description 23
- 239000003814 drug Substances 0.000 abstract description 4
- 102000011923 Thyrotropin Human genes 0.000 description 24
- 108010061174 Thyrotropin Proteins 0.000 description 24
- 238000005259 measurement Methods 0.000 description 15
- 210000001519 tissue Anatomy 0.000 description 14
- 238000011282 treatment Methods 0.000 description 13
- 238000002595 magnetic resonance imaging Methods 0.000 description 11
- 238000004458 analytical method Methods 0.000 description 9
- 238000004590 computer program Methods 0.000 description 9
- 208000022873 Ocular disease Diseases 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 7
- 230000004054 inflammatory process Effects 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 206010061218 Inflammation Diseases 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 230000036541 health Effects 0.000 description 6
- 238000012544 monitoring process Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 201000010099 disease Diseases 0.000 description 5
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 5
- 238000007781 pre-processing Methods 0.000 description 5
- 239000000284 extract Substances 0.000 description 4
- 208000003532 hypothyroidism Diseases 0.000 description 4
- 230000002989 hypothyroidism Effects 0.000 description 4
- 230000003902 lesion Effects 0.000 description 4
- 238000010606 normalization Methods 0.000 description 4
- 230000036285 pathological change Effects 0.000 description 4
- 231100000915 pathological change Toxicity 0.000 description 4
- 238000011269 treatment regimen Methods 0.000 description 4
- XUIIKFGFIJCVMT-GFCCVEGCSA-N D-thyroxine Chemical compound IC1=CC(C[C@@H](N)C(O)=O)=CC(I)=C1OC1=CC(I)=C(O)C(I)=C1 XUIIKFGFIJCVMT-GFCCVEGCSA-N 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 3
- 238000013136 deep learning model Methods 0.000 description 3
- 230000003862 health status Effects 0.000 description 3
- 229940034208 thyroxine Drugs 0.000 description 3
- XUIIKFGFIJCVMT-UHFFFAOYSA-N thyroxine-binding globulin Natural products IC1=CC(CC([NH3+])C([O-])=O)=CC(I)=C1OC1=CC(I)=C(O)C(I)=C1 XUIIKFGFIJCVMT-UHFFFAOYSA-N 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000002965 ELISA Methods 0.000 description 2
- 206010015943 Eye inflammation Diseases 0.000 description 2
- 238000012307 MRI technique Methods 0.000 description 2
- AUYYCJSJGJYCDS-LBPRGKRZSA-N Thyrolar Chemical compound IC1=CC(C[C@H](N)C(O)=O)=CC(I)=C1OC1=CC=C(O)C(I)=C1 AUYYCJSJGJYCDS-LBPRGKRZSA-N 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 210000000744 eyelid Anatomy 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 210000002865 immune cell Anatomy 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000004220 muscle function Effects 0.000 description 2
- 238000012805 post-processing Methods 0.000 description 2
- 238000011158 quantitative evaluation Methods 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 208000024891 symptom Diseases 0.000 description 2
- 230000006016 thyroid dysfunction Effects 0.000 description 2
- 229940035722 triiodothyronine Drugs 0.000 description 2
- MZOFCQQQCNRIBI-VMXHOPILSA-N (3s)-4-[[(2s)-1-[[(2s)-1-[[(1s)-1-carboxy-2-hydroxyethyl]amino]-4-methyl-1-oxopentan-2-yl]amino]-5-(diaminomethylideneamino)-1-oxopentan-2-yl]amino]-3-[[2-[[(2s)-2,6-diaminohexanoyl]amino]acetyl]amino]-4-oxobutanoic acid Chemical compound OC[C@@H](C(O)=O)NC(=O)[C@H](CC(C)C)NC(=O)[C@H](CCCN=C(N)N)NC(=O)[C@H](CC(O)=O)NC(=O)CNC(=O)[C@@H](N)CCCCN MZOFCQQQCNRIBI-VMXHOPILSA-N 0.000 description 1
- 208000023275 Autoimmune disease Diseases 0.000 description 1
- 201000004569 Blindness Diseases 0.000 description 1
- 208000003164 Diplopia Diseases 0.000 description 1
- 102000004190 Enzymes Human genes 0.000 description 1
- 108090000790 Enzymes Proteins 0.000 description 1
- 102000010834 Extracellular Matrix Proteins Human genes 0.000 description 1
- 108010037362 Extracellular Matrix Proteins Proteins 0.000 description 1
- 208000019749 Eye movement disease Diseases 0.000 description 1
- 238000005481 NMR spectroscopy Methods 0.000 description 1
- 206010030113 Oedema Diseases 0.000 description 1
- 206010061323 Optic neuropathy Diseases 0.000 description 1
- 208000018087 Orbital disease Diseases 0.000 description 1
- 102000003911 Thyrotropin Receptors Human genes 0.000 description 1
- 108090000253 Thyrotropin Receptors Proteins 0.000 description 1
- 238000011256 aggressive treatment Methods 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006472 autoimmune response Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000000593 degrading effect Effects 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 208000029444 double vision Diseases 0.000 description 1
- 238000002592 echocardiography Methods 0.000 description 1
- 210000002744 extracellular matrix Anatomy 0.000 description 1
- 230000004438 eyesight Effects 0.000 description 1
- 230000005714 functional activity Effects 0.000 description 1
- 230000013632 homeostatic process Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000028993 immune response Effects 0.000 description 1
- 210000000987 immune system Anatomy 0.000 description 1
- 238000003018 immunoassay Methods 0.000 description 1
- 238000010166 immunofluorescence Methods 0.000 description 1
- 230000008595 infiltration Effects 0.000 description 1
- 238000001764 infiltration Methods 0.000 description 1
- 230000002757 inflammatory effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000009533 lab test Methods 0.000 description 1
- 230000005415 magnetization Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000002503 metabolic effect Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013421 nuclear magnetic resonance imaging Methods 0.000 description 1
- 208000028780 ocular motility disease Diseases 0.000 description 1
- 208000020911 optic nerve disease Diseases 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008506 pathogenesis Effects 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 230000011514 reflex Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001225 therapeutic effect Effects 0.000 description 1
- 230000008719 thickening Effects 0.000 description 1
- 230000000451 tissue damage Effects 0.000 description 1
- 231100000827 tissue damage Toxicity 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 230000004382 visual function Effects 0.000 description 1
Landscapes
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The embodiment of the invention relates to the technical field of medicine, and discloses a method and a system for intelligently detecting thyroid-related eye diseases, wherein the method comprises the following steps: in the embodiment of the invention, the eye data information of the patient is substituted into an eye score calculation formula model to calculate the eye data score of the patient; adding the scoring data in the eye data scoring of the patient according to a certain proportion relation to calculate the initial TAO scoring value of the current patient; accumulating the initial TAO scoring value of each current patient in a specified time period, and dividing the accumulated initial TAO scoring value by the specified time period to calculate a target TAO scoring value of the patient; after combining the patient target TAO score value, hyperthyroidism eye disease clinical activity score value, and thyroid function and antibody detection report to obtain a TAO composite score, the patient's current TAO activity status is determined. By implementing the embodiment of the invention, the diagnosis efficiency of the thyroid-related eye diseases can be improved.
Description
Technical Field
The invention belongs to the technical field of medicine, and particularly relates to a method and a system for intelligently detecting thyroid-related eye diseases.
Background
Thyroid-related eye disease (TAO) is an autoimmune disease, the onset of which is closely related to thyroid dysfunction, most of which are binocular, the most common orbital disease in adults. The common clinical manifestations of TAO patients are protrusion of eyeballs, oedema of orbital tissues, eyelid withdrawal and the like, and also can cause eye movement disorder and double vision caused by the involvement of extraocular muscles, and the serious patients can cause compressive optic neuropathy, vision decline, blindness and other visual function damages.
However, in practice, conventional diagnosis of thyroid-related eye diseases has been found to rely primarily on manual examination by experienced ophthalmologists, including inquiry history, visual examination, basic ophthalmic examination, and the like. Although this approach can diagnose the typing, staging and classification of TAOs, it has some problems. First, manual diagnosis is time consuming and labor intensive, especially when dealing with large numbers of patients, and the workload of the doctor is increased. Secondly, the accuracy of the manual diagnosis is affected by the experience and skill level of doctors and other factors, and certain errors exist.
Disclosure of Invention
The embodiment of the invention discloses a method and a system for intelligently detecting thyroid-related eye diseases, which can improve the diagnosis efficiency of the thyroid-related eye diseases.
The embodiment of the invention discloses a method for intelligently detecting thyroid-related eye diseases, which comprises the following steps:
substituting the eye data information of the patient into an eye score calculation formula model to calculate the eye data score of the patient;
Adding the scoring data in the eye data scoring of the patient according to a certain proportion relation to calculate the initial TAO scoring value of the current patient; wherein the eye data score of the patient at least comprises a patient binocular eyeball salience score, a patient binocular extraocular rectus muscle thickness score, a patient binocular ocular muscle reflectivity score, a patient binocular lacrimal gland area score, a patient binocular lacrimal gland T2 value score, a patient binocular lacrimal fluid score and a patient blood immunoregulatory molecule score;
accumulating the initial TAO scoring value of each current patient in a specified time period, and dividing the accumulated initial TAO scoring value by the specified time period to calculate a target TAO scoring value of the patient;
After combining the patient target TAO score value, hyperthyroidism eye disease clinical activity score value, and thyroid function and antibody detection report to obtain a TAO composite score, the patient's current TAO activity status is determined.
In a second aspect, an embodiment of the present invention discloses a detection system, including:
The first calculation unit is used for substituting the eye data information of the patient into an eye score calculation formula model so as to calculate the eye data score of the patient;
The second calculation unit is used for respectively adding the scoring data in the eye data scoring of the patient according to a certain proportion relation so as to calculate the initial TAO scoring value of the current patient; wherein the eye data score of the patient at least comprises a patient binocular eyeball salience score, a patient binocular extraocular rectus muscle thickness score, a patient binocular ocular muscle reflectivity score, a patient binocular lacrimal gland area score, a patient binocular lacrimal gland T2 value score, a patient binocular lacrimal fluid score and a patient blood immunoregulatory molecule score;
A third calculation unit, configured to accumulate each of the current patient initial TAO score values in a specified time period, and divide the accumulated value by the specified time period, so as to calculate a patient target TAO score value;
And a fourth calculation unit, configured to determine a current TAO activity condition of the patient after combining the target TAO score value, the hyperthyroidism eye disease clinical activity score value, and the thyroid function and antibody detection report to obtain a TAO comprehensive score.
A third aspect of an embodiment of the present invention discloses a detection system, including:
A memory storing executable program code;
A processor coupled to the memory;
The processor invokes the executable program code stored in the memory to execute the method for intelligently detecting thyroid-related eye diseases disclosed in the first aspect of the embodiment of the invention.
A fourth aspect of the embodiment of the present invention discloses a computer-readable storage medium storing a computer program, where the computer program causes a computer to execute a method for intelligently detecting thyroid-related eye diseases disclosed in the first aspect of the embodiment of the present invention.
A fifth aspect of an embodiment of the invention discloses a computer program product which, when run on a computer, causes the computer to perform part or all of the steps of any one of the methods of intelligently detecting thyroid-related eye disorders of the first aspect.
A sixth aspect of the embodiments of the present invention discloses an application publishing platform for publishing a computer program product, wherein the computer program product, when run on a computer, causes the computer to perform part or all of the steps of any one of the methods for intelligently detecting thyroid-related eye diseases of the first aspect.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
In the embodiment of the invention, the eye data information of the patient is substituted into an eye score calculation formula model to calculate the eye data score of the patient; adding the scoring data in the eye data scoring of the patient according to a certain proportion relation to calculate the initial TAO scoring value of the current patient; wherein the eye data score of the patient at least comprises a patient binocular eyeball salience score, a patient binocular extraocular rectus muscle thickness score, a patient binocular ocular muscle reflectivity score, a patient binocular lacrimal gland area score, a patient binocular lacrimal gland T2 value score, a patient binocular lacrimal fluid score and a patient blood immunoregulatory molecule score; accumulating the initial TAO scoring value of each current patient in a specified time period, and dividing the accumulated initial TAO scoring value by the specified time period to calculate a target TAO scoring value of the patient; after combining the patient target TAO score value, hyperthyroidism eye disease clinical activity score value, and thyroid function and antibody detection report to obtain a TAO composite score, the patient's current TAO activity status is determined. Therefore, the embodiment of the invention can improve the diagnosis efficiency of the thyroid-related eye diseases.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a method for intelligently detecting thyroid-related eye diseases according to an embodiment of the invention;
FIG. 2 is a schematic flow chart of another method for intelligently detecting thyroid-related eye diseases according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a detection system according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of another detection system according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of another detection system according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that the terms "first," "second," "third," "fourth," and the like in the description and in the claims of the present invention are used for distinguishing between different objects and not necessarily for describing a particular sequential or chronological order. The terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
The embodiment of the invention discloses a method and a system for intelligently detecting thyroid-related eye diseases, which can improve the diagnosis efficiency of the thyroid-related eye diseases.
The following detailed description refers to the accompanying drawings.
Example 1
Referring to fig. 1, fig. 1 is a schematic flow chart of a method for intelligently detecting thyroid-related eye diseases according to an embodiment of the invention. As shown in fig. 1, the method of intelligently detecting thyroid-related eye diseases may include the following steps.
101. The detection system substitutes the patient eye data information into an eye score calculation formula model to calculate the eye data score of the patient.
As an alternative implementation mode, in the embodiment of the application, the eye data information of the patient comprises the protrusion degree of eyeballs of the two eyes of the patient, the thickness of rectus muscle outside the eyes of the two eyes of the patient, the reflectivity of muscle of the two eyes of the patient, the area of lacrimal glands of the two eyes of the patient, the T2 value of lacrimal glands of the two eyes of the patient, the content of proinflammatory factors in tears of the two eyes of the patient and the content of immunoregulatory molecules in blood of the patient.
As an optional implementation manner, in the embodiment of the present application, the eye score calculation formula model is formulated by a relevant medical expert or researcher according to the research purpose evaluation standard of thyroid-related eye diseases, wherein the eye score calculation formula model not only contains one calculation formula, but also can be provided with a plurality of calculation formulas, and each calculation formula corresponds to one score in the eye data scores of the patient, and the present application is not limited in any way.
As another alternative, in the embodiment of the present application, the calculation of each score in the eye data score of the patient according to the present application may be performed simultaneously or sequentially, and the present application is not limited in any way.
102. The detection system respectively adds the scoring data in the eye data scoring of the patient according to a certain proportion relation to calculate the initial TAO scoring value of the current patient; the patient eye data score at least comprises a patient binocular eyeball salience score, a patient binocular extraocular rectus muscle thickness score, a patient binocular ocular muscle reflectivity score, a patient binocular lacrimal gland area score, a patient binocular lacrimal gland T2 value score, a patient binocular lacrimal fluid score and a patient blood immunoregulatory molecule score.
As an alternative implementation, in the embodiment of the present invention, each scoring data in the patient's eye data score has a weight, and each weight is determined according to medical study and expert opinion.
As an alternative embodiment, in the present embodiment, the present application may use the initial TAO score value= (the score of score item 1 x the weight of score item 1 in the patient's eye data score) + (the score of score item 2 x the weight of score item 2 in the patient's eye data score) +.
As an alternative implementation, in an embodiment of the present invention, the detection system may first obtain a score for the patient's eye data, which may include scores for various aspects of eye herniation, eye muscle function, eye inflammation, and the like. Each score reflects the severity of a patient on a particular ocular symptom or sign. The detection system may then weight the scoring data according to a predetermined scaling relationship. These scaling relationships are typically derived based on clinical experience and study data to ensure that the weights of the scores in the final calculation match their importance to TAO condition assessment. After the weighting process is completed, the detection system adds the scoring data according to the proportion relation, so as to obtain a comprehensive initial TAO scoring value. The grading value can comprehensively reflect the current eye disease condition of the patient and provide objective and quantitative evaluation basis for doctors.
As an optional implementation manner, in the embodiment of the invention, the detection system can calculate the initial TAO (total internal heat input) score value by adding the scoring data in the eye data scores of the patients according to a certain proportion relation, so as to provide objective and quantitative disease evaluation basis for doctors.
103. The detection system accumulates and divides each current patient initial TAO score value over a specified period of time to calculate a patient target TAO score value.
In the embodiment of the application, the formula model is finally calculated by multiple test analysis according to 100 people data and multiple fitting of equation optimization parameter model. The present application may employ a fractional (ISA) algorithm formulation model: where M IS the specified time period and IS IS the initial TAO score for each patient.
As an alternative embodiment, in an embodiment of the present application, the present application may calculate the current patient initial TAO score value once every certain width of the time window (e.g., 1 day). According to the eye area state of the monitored patient, variability data of an eye area in one period (7 days) are obtained, whether the TAO score value of the user in the period is maintained within a certain range is obtained through analysis, and then a target TAO score value can be obtained through calculation so as to ensure the diagnosis accuracy of thyroid-related eye diseases.
As an alternative embodiment, in the present example, the initial TAO score is only a fraction of the patient's condition assessment, and the physician may need to incorporate various information about the patient's medical history, clinical manifestations, laboratory tests, etc. in diagnosing TAO. Meanwhile, the TAO score value of a patient may also change with the change of the condition and the effect of treatment, so that regular monitoring and evaluation are required to obtain the target TAO score value of the patient.
104. The detection system determines the current TAO activity status of the patient after combining the patient target TAO score value, hyperthyroidism eye disease clinical activity score value, and thyroid function and antibody detection report to obtain a TAO composite score.
As an alternative implementation, in the embodiment of the present invention, the eye health condition of the patient is explained with reference to a preset scoring range or standard according to the calculated TAO comprehensive score, and different scoring ranges may correspond to different disease severity or require different treatment strategies.
As an optional implementation mode, in the embodiment of the invention, a plurality of indexes can be fused to accurately evaluate the current TAO activity condition of the patient, so that the diagnosis efficiency of the thyroid-related eye diseases is improved, the misdiagnosis condition can be reduced, and meanwhile, the method is suitable for various crowds, has the characteristics of convenience, rapidness and long-time continuous monitoring, and enables a main doctor to more quickly and accurately know the eye region condition of the patient and timely discover the eye disease problem.
As an optional implementation manner, in the embodiment of the application, a multiple scoring mode is adopted to detect and score most lesion types presented in the active period of the thyroid-related eye diseases, and then the hyperthyroidism eye disease clinical activity scoring value (CAS) and the thyroid function and antibody detection report are combined to improve the detection accuracy.
As an alternative embodiment, in embodiments of the application, in assessing thyroid-related eye disease (TAO) activity of a patient, the application may integrate information in a number of aspects, including TAO score, hyperthyroidism eye disease clinical activity score, and thyroid function and antibody detection reports. Together, these indices constitute a comprehensive score for TAO, which helps to fully understand the patient's condition and its activity. Firstly, the diagnosis method commonly used in the current medicine can be adopted, namely, the probability of the active period of the disease condition of a patient is firstly estimated through the clinical activity grading value of the hyperthyroidism eye disease, and then the severity of TAO can be estimated through a series of clinical and laboratory examinations by combining the TAO grading value obtained by the application. While a higher (lower) score generally means a more severe patient condition. Finally, thyroid function and antibody detection reports can provide information about the status of thyroid function in patients, such as levels of Thyroid Stimulating Hormone (TSH), free thyroxine (FT 4), free triiodothyronine (FT 3), and the presence or absence of antibodies (e.g., thyroid stimulating hormone receptor antibody TRAb). This information is critical to judging the activity of TAO, as TAO is often closely related to thyroid dysfunction and autoimmune response.
As an alternative implementation, in the embodiment of the present application, after integrating the information of the above three aspects, the present application may obtain a TAO comprehensive score. This score will integrate the patient's severity, clinical manifestations and thyroid function status, thus providing us with a comprehensive assessment of the patient's current TAO activity status. Based on this composite score, the system can determine the current TAO activity status of the patient. If the score is higher (lower), meaning that the patient's condition is more active, more aggressive treatment and management may be required. Conversely, if the score is low (high), it may mean that the patient's condition is relatively stable, but the treatment regimen still needs to be monitored and adjusted periodically.
As an alternative embodiment, in the present example, the present application may assign a weight to the target TAO score value, the hyperthyroidism eye disease clinical activity score value, and the thyroid function and antibody detection report, respectively, which may be determined according to medical study and expert opinion. Subsequently, the present application may calculate TAO composite score using TAO composite score = (patient target TAO score x weight of patient target TAO score) + (hyperthyroidism eye disease clinical activity score x weight of hyperthyroidism eye disease clinical activity score) + (thyroid function and antibody detection report score x weight of thyroid function and antibody detection report score), and then explain the eye health condition of the patient with reference to a preset score range or standard by means of the TAO composite score, and different score ranges may correspond to different disease severity or require different treatment strategies. Wherein, since there is information about the thyroid function status of the patient, such as the levels of Thyroid Stimulating Hormone (TSH), free thyroxine (FT 4), free triiodothyronine (FT 3), etc., in the thyroid function and antibody detection report, the system may preset a scoring standard of the thyroid function and antibody detection report, i.e. a score determined according to the information about the thyroid function status of each patient, for example, hyperthyroidism or hypothyroidism patient, generally, the change of TSH is opposite to FT3 and FT4, such as typical hyperthyroidism patient, TSH is lower than the normal range, and FT3 and FT4 are higher than the normal range. Whereas typical hypothyroidism patients, TSH is above the normal range, FT3, FT4 are below the normal range. For hyperthyroidism patients, for example, if a positive score is assigned when Thyroid Stimulating Hormone (TSH) is below the normal range of values, then a positive score should be assigned when the other first work index is above the normal range of values, so that the score is not offset. I.e. when Thyroid Stimulating Hormone (TSH) is above the conventional value range, a positive score is conferred; negative scores are assigned when Thyroid Stimulating Hormone (TSH) is below the normal value range; and carrying out other data of the thyroid function state of the patient in a similar way, then accumulating all scores, and calculating to obtain thyroid function and antibody detection report scores.
As an optional implementation manner, in the embodiment of the application, the present application can more accurately understand the current TAO activity condition of the patient by comprehensively evaluating the TAO score value, the hyperthyroidism eye disease clinical activity score value and the thyroid function and antibody detection report of the patient, so that powerful support is provided for clinical decision.
In the method of intelligently detecting thyroid-related eye disease of fig. 1, a detection system is described as an execution subject. It should be noted that, the implementation subject of the method for intelligently detecting thyroid-related eye diseases in fig. 1 may also be a stand-alone device associated with the detection system, which is not limited by the embodiment of the present invention.
Therefore, implementing the method for intelligently detecting thyroid-related eye diseases described in fig. 1 can improve the diagnosis efficiency of thyroid-related eye diseases.
In addition, by implementing the method for intelligently detecting the thyroid-related eye diseases, which is described in fig. 1, the activity condition of the current TAO of the patient can be known more accurately, so that powerful support is provided for clinical decision.
Example two
Referring to fig. 2, fig. 2 is a flow chart of another method for intelligently detecting thyroid-related eye diseases according to an embodiment of the invention. As shown in fig. 2, the method for intelligently detecting thyroid-related eye diseases may include the following steps:
201. The detection system trains the initial model of the U-net neural network by using the medical image data set with the labels so as to obtain the model of the U-net neural network; wherein the medical image dataset contains accurate segmentation annotations for the horizontal eye region and the coronal plane eye region.
202. The detection system inputs the preprocessed fusion scanning image into the U-net neural network model to generate a segmentation mask to mark a target region in the fusion scanning image.
203. The detection system quantifies the accuracy of the segmentation mask using the specified evaluation index to determine the horizontal bit eye region and the coronal bit plane eye region.
As an alternative implementation, in an embodiment of the present invention, fusing Single Photon Emission Computed Tomography (SPECT) images and electronic Computed Tomography (CT) images may provide richer, more accurate anatomical and functional information in medical image processing. These two imaging techniques each have advantages: SPECT is capable of displaying functional information of tissue, whereas CT provides accurate anatomical information. By fusing them, a fused scan image can be created that contains both structural and functional information. Once the fused scan image is obtained, the next step is to segment the specific region in the image using a deep learning model, particularly a U-Net neural network model. The U-Net neural network model is a powerful segmentation tool, particularly suited for medical image segmentation tasks, because it can capture the contextual information of the image and precisely map out the region of interest. When applying the U-Net neural network model to the fused scan image, the goal is to segment out the level eye region and the coronal eye region. The selection of these two regions is based on their importance for diagnosis and treatment of ocular diseases. The level eye region provides a cross-sectional view of the eyeball, helping to assess the shape, size and internal structure of the eyeball. Whereas the coronal plane eye region provides a side view of the eyeball, helping to observe the relationship of the eyeball to surrounding tissues.
As an alternative implementation, in the embodiment of the present application, the process of splitting using the U-Net neural network model is approximately as follows: the fused scan image can be first subjected to the necessary pre-processing, such as normalization, denoising, etc., to improve the segmentation performance of the model, and then the U-Net model can be trained using the labeled medical image dataset. The data sets should contain accurate segmentation labels of the horizontal eye region and the coronal eye region, the preprocessed fusion scan image is input into a trained U-Net model, the model generates a segmentation mask, a target region in the image is marked, and the segmentation result generated by the model is subjected to post-processing, such as smoothing edges, removing small regions and the like, so as to improve the segmentation accuracy, and finally, the application can use proper evaluation indexes (such as a Dice coefficient, ioU and the like) to quantify the segmentation result accuracy.
204. The detection system identifies vertex pixel coordinates of the anterior surface of the cornea of both eyes from the horizontal eye region and most prominent point pixel coordinates of the lateral edge of the orbit from the coronal eye region.
205. The detection system calculates the linear distance from the vertex of the anterior surface of the cornea to the most prominent point of the lateral edge of the eye socket by using a distance calculation formula after converting the pixel coordinates of the vertex of the anterior surface of the cornea and the most prominent point of the lateral edge of the eye socket into space coordinates.
206. The detection system utilizes a trigonometric function to substitute the vertex space coordinates of the front surface of the cornea of the eyes, the space coordinates of the most protruding point of the outer edge of the eye socket and the linear distance so as to calculate the protrusion value of the eyeballs of the eyes of the patient.
As an alternative implementation, in the embodiment of the present invention, in order to calculate the protrusion value of the eyes of the patient, the detection system needs to first extract the coordinate information of the key points from the horizontal eye region and the coronal eye region. These key points include the vertex coordinates of the anterior surface of the cornea of the two eyes and the most prominent point coordinates of the lateral edge of the orbit. In the horizontal eye region, the system can detect the vertex of the anterior surface of the cornea of both eyes. This may be achieved by means of brightness in the image, edge detection or shape recognition, etc. Whereas in the coronal plane eye region, the system may identify the most prominent points of the lateral orbital rims. This is typically located at the bony structures outside the orbit. After the system detects these keypoints, the system can then accurately extract their coordinate values, which can be done by converting the pixel coordinates into actual spatial coordinates (e.g., millimeters), which typically depend on the resolution of the scanned image and the calibration information. The system may then measure the straight line distance from the corneal vertex to the most prominent point of the lateral orbital edge by Euclidean distance formula or other distance calculation method. The system can then calculate the eye prominence of both eyes separately and make a normalization or adjustment relative to the orbit size to obtain a more accurate measurement. Finally, the system may output the calculated eye prominence value in a suitable format, such as a numerical value, chart, or report. Wherein the accuracy and reliability of the calculation may need to be compared with standard values or previous measurement results.
As an alternative implementation, in the embodiment of the present application, the accurate calculation of the eye prominence of the present application may be affected by a variety of factors, including the resolution of the image, the accuracy of the scanning technique, the degree of patient coordination, and any external factors that may affect the eye position (e.g., eyelid position, eye muscle movement, etc.). Therefore, when the eyeball saliency is measured, whether the measurement is performed under the condition of controlling external influence factors or not can be determined, and if so, the operation is performed.
As an alternative implementation, in the present embodiment, assuming that the spatial coordinates of the anterior surface vertex of the left eye cornea are a (x 1, y1, z 1), the spatial coordinates of the anterior surface vertex of the right eye cornea are B (x 2, y2, z 2), the spatial coordinates of the most protruding point of the lateral edge of the left eye orbit C (x 3, y3, z 3), the spatial coordinates of the most protruding point of the lateral edge of the right eye orbit D (x 4, y4, z 4), then the system can calculate the straight line distance from the anterior surface vertex of the cornea to the most protruding point of the lateral edge of the eye orbit by a distance formula between two points in three-dimensional space, i.e. for the left eye,For the right eye,/>We can then calculate the eye prominence value using a trigonometric function. Assuming that the angle θ between the line of the lateral orbital edge to the anterior surface of the cornea and a certain reference plane (e.g., horizontal or sagittal plane), the left eye prominence value/>For/>Right eye protrusion value/>For/>The angle theta of the application can be calculated by the angle in the three-dimensional coordinates. In particular, the present application may use an arctan function (arctan) or an arctan binary function (atan 2) to calculate vector/>And/>Respectively, and the reference plane. The calculations herein assume that the eye prominence is measured in a particular direction (defined by angle θ). In practical applications, this direction may need to be determined according to specific clinical requirements and measurement criteria, and the present application is not limited in any way. In addition, in order to obtain accurate measurement results, it is also necessary to ensure accurate acquisition and calibration of spatial coordinates. Finally, by substituting specific spatial coordinates and the calculated straight line distance, the system can calculate the protrusion value of the eyes of the patient by using the formula. These values can provide the doctor with important information about the position of the patient's eye, facilitating diagnosis and treatment of ocular diseases such as TAO.
207. The detection system selects the image level with the highest radionuclide uptake from the coronal plane eye region as the target eye image level.
As an alternative embodiment, in the present example, how much radionuclide is taken may reflect the metabolic activity or functional state of the eye tissue, so the present application chooses the image plane that the radionuclide is taken most to help more accurately assess the eye condition.
As an alternative implementation, in the embodiment of the present invention, first, the system may perform necessary preprocessing on the image of the coronal bit-plane eye region, including denoising, contrast enhancement, and the like, so as to improve the image quality and the accuracy of subsequent analysis. The system may then calculate the radionuclide uptake for each image slice using image processing techniques. This typically involves statistics and analysis of pixel values in the image to quantify the distribution and concentration of radionuclides. Next, the system may compare radionuclide uptake at different image levels. This step may be accomplished automatically by an algorithm that sorts the intake of each level to identify the level with the highest intake. Finally, the system can select the image layer with the most radionuclides taken as the target eye image layer according to the comparison result. This aspect generally represents the most active or most visible region of the eye's functional activity, is of great importance for subsequent medical analysis and diagnosis, and can provide powerful support for subsequent medical analysis and diagnosis.
As an alternative implementation, in the embodiment of the present application, when selecting the target eye image level, other factors, such as image quality, sharpness of the eye structure, etc., should also be considered. Furthermore, for different patients or different scanning conditions, it may be necessary to adjust selection criteria or algorithm parameters to ensure accuracy and reliability of the results.
208. After the external rectus muscle of both eyes of the patient is divided from the target eye image plane, the pixel distance on the image perpendicular to the long axis of the muscle at the midpoint position of the external rectus muscle of both eyes of the patient is measured, and the pixel distance is subjected to spatial dimension conversion to obtain the thickness value of the external rectus muscle of both eyes of the patient.
As an alternative implementation, in an embodiment of the present invention, first, the system may enhance and pre-process the target eye image level to improve the visibility and contrast of the extraocular muscles. This may include adjusting the brightness, contrast, sharpened edges, etc. of the image to better identify the boundaries and morphology of the extraocular muscles. Second, the system may utilize image processing algorithms, such as edge detection, region segmentation, or deep learning models, to identify and locate extraocular muscles in the image plane of the target eye. These algorithms can identify specific texture, shape and location features of the extraocular muscle, distinguishing it from surrounding tissue. Next, once the extraocular muscles are identified, they can be classified and marked by the detection system. This can be achieved by drawing border lines on the image, filling in different colors or using labels. For the upper rectus muscle, the lower rectus muscle, the inner rectus muscle and the outer rectus muscle of the eyes, the detection system needs to be divided respectively, and the clear and accurate boundary of each muscle is ensured. After the division of the extraocular muscles is completed, the detection system may need to perform verification and correction steps. This can be achieved by comparing with the results of the manual demarcation by the medical expert to ensure accuracy and reliability of the demarcation. If there is a discrepancy or error, the system may automatically or manually make the correction. Finally, the detection system may output the segmented extraocular muscle images and associated measurement and analysis results as a report or image file. These results can be used for medical diagnosis, treatment planning and condition monitoring.
As an alternative implementation, in an embodiment of the present invention, first, the detection system can accurately identify and locate the external rectus muscles of both eyes of the patient on the target eye image level. This typically involves using image processing algorithms or deep learning models to identify specific texture, shape and location features of the external rectus muscle. Subsequently, once the external rectus muscle is identified and located, the system may effect a determination of the midpoint of the rectus muscle by calculating the geometric center of the external rectus muscle or a midpoint along the length of the muscle. This enables an accurate determination of the rectus midpoint, which is a key step in measuring thickness values. And after determining the midpoint of the rectus muscle, the detection system may evaluate the thickness of the external rectus muscle by measuring the thickness of the muscle at that point. This may be achieved by measuring the pixel distance on an image perpendicular to the long axis of the muscle and may need to take into account the resolution of the image and calibration information to translate into a real spatial dimension (e.g. millimeters). The detection system may then perform the same measurement procedure on the external rectus muscles of both eyes of the patient, ensuring accurate thickness assessment on each side. The measurement results for each eye should be recorded separately for subsequent analysis and comparison. Finally, the detection system may output the measured value of the thickness of the midpoint of the rectus muscle of the patient's extraocular rectus muscle in a suitable format, such as a numerical value, a chart, or a report. These data can be compared to normal reference values or previous measurements to assess extraocular muscle abnormalities, such as thickening or thinning, to assist the physician in diagnosis and treatment decisions.
As an alternative implementation, in an embodiment of the invention, the measurement of the thickness of the external rectus muscle may be affected by a number of factors, including image quality, individual differences in ocular anatomy, and possible pathological changes. Therefore, in making the measurements, the detection system should employ as standardized and reliable a method as possible to ensure accuracy and reliability of the measurement results.
209. The detection system divides the lacrimal gland boundary of the patient from the fusion scanning image according to different slice directions and spatial position information in the fusion scanning image; wherein, the lacrimal boundary at least comprises a patient coronal lacrimal boundary and a patient cross-sectional lacrimal boundary.
210. The number of pixels within the patient's coronal lacrimal gland boundary is multiplied by the actual area represented by each pixel, and the number of pixels within the patient's cross-sectional lacrimal gland boundary is multiplied by the actual area represented by each pixel to calculate the patient's coronal lacrimal gland area and the patient's cross-sectional lacrimal gland area, respectively.
As an alternative embodiment, in an embodiment of the present invention, the detection system may first confirm that the lacrimal gland boundary has been accurately delineated. This typically involves checking the integrity, continuity and accuracy of the boundary, ensuring that no missing or redundant regions are incorporated. The system may then segment the lacrimal boundary of the coronal and cross-sectional planes based on the different slice directions and spatial location information of the fused scan images. In the coronal image, the detection system may derive the coronal lacrimal area by calculating the number of pixels (or area units) within the lacrimal boundary and multiplying the actual area represented by each pixel (depending on the resolution of the image and the calibration information). Similarly, in the cross-sectional image, the detection system may also calculate the number of pixels within the lacrimal boundary using the same method and convert to the actual area to yield the cross-sectional lacrimal area. The system may then convert the calculated area value from pixel units to units commonly used in medicine (e.g., square millimeters or square centimeters). In addition, normalization may be required to compare lacrimal gland areas in different individuals or under different scanning conditions. After the calculation is completed, the detection system can verify the result, so that the accuracy of the calculation is ensured. If there is a large discrepancy from the expected result, it may be necessary to check the accuracy or recalculation of the boundary loop. Finally, the detection system outputs the calculated coronal lacrimal area and cross-sectional lacrimal area in a suitable format, such as a numerical value, chart, or report. These results can provide the physician with important information about lacrimal gland size and morphology, aiding in the diagnosis and treatment of ocular disorders.
211. The detection system acquires the T2 weighting value of each pixel in the lacrimal boundary by adopting a short T reversal sequence MRI technology, and forms a T2 weighted lacrimal boundary image.
As an alternative implementation, in an embodiment of the present invention, first, the detection system ensures that the MRI (magnetic resonance imaging) scanner is properly set up and employs a short T-reversal MRI (magnetic resonance imaging) technique. This technique is particularly useful for obtaining T2 weighted information of tissue because it is very sensitive to the motion state of water molecules. Subsequently, the detection system may initiate an MRI (magnetic resonance imaging) scan to acquire image data of the lacrimal gland region. After the scanning is completed, the detection system may process the acquired image data. This includes preprocessing steps to remove noise, enhance contrast, etc., to improve image quality. Next, the detection system may perform a calculation of a T2 weighting value for each pixel in the lacrimal boundary. This involves analyzing the MRI (magnetic resonance imaging) signal intensity of the pixels and extracting information about the T2 relaxation time according to the principles of the short T-reversal sequential MRI technique. The T2 relaxation time describes the state of movement of water molecules in a tissue in a magnetic field, reflecting the structure and properties of the tissue. And after calculating the T2 weighting value for each pixel, the detection system may map these values onto the corresponding pixel to generate a T2 weighted lacrimal boundary image. Such images can highlight specific characteristics of lacrimal gland tissue, such as moisture content, tissue structure, etc. The generated T2 weighted lacrimal boundary image is subjected to quality inspection to ensure the definition and accuracy of the image. If any problems are found, the detection system may re-scan or adjust the process parameters. Finally, the detection system outputs the T2 weighted lacrimal boundary image in an appropriate format, such as a digital image file. From these images, the physician can assess the health of the lacrimal gland and detect possible abnormalities or lesions.
As an optional implementation manner, in the embodiment of the application, the T2 weighted value of the lacrimal gland boundary is obtained by adopting the short T reversal sequence MRI technology, and the T2 weighted lacrimal gland boundary image is formed, so that the detection system can provide more detailed and accurate lacrimal gland structure information for doctors, and is helpful for early discovery and diagnosis of ocular diseases.
212. After the process of registering and aligning the T2 weighted lacrimal boundary image with the T2 weighted image on the coronal plane eye region, the detection system superimposes the T2 weighted lacrimal boundary image on the T2 weighted image on the coronal plane eye region to obtain a patient's current lacrimal maximum T2 value and a patient's current lacrimal average T2 value.
In the present embodiment, the T2 value of the present application is a term commonly used in the process of nuclear magnetic resonance examination of a patient, T2 represents the time taken for the transverse magnetization vector to decay to 37% of the maximum value, and the MRI of the present application is nuclear magnetic resonance imaging.
As an alternative implementation, in an embodiment of the present invention, first, the detection system may register and align the T2 weighted lacrimal boundary image with the T2 weighted image of the coronal bit plane eye region. This ensures spatial consistency of the two images so that the lacrimal boundary image can be accurately superimposed on the eye region image. Upon completion of registration, the detection system may superimpose the T2-weighted lacrimal boundary image onto the T2-weighted eye region image. This typically involves applying the lacrimal boundary image as a layer or mask to the ocular region image so that the lacrimal boundary is clearly visible in the original ocular image. In the superimposed image, the detection system may extract a T2 weight for each pixel within the lacrimal boundary. These values reflect the nature of the movement of water molecules in the lacrimal tissue and provide information about the health status of the lacrimal gland. The detection system traverses all the extracted T2 values to find the maximum value thereof, i.e. the maximum T2 value of the current lacrimal gland of the patient. This value represents the region of the lacrimal gland where the T2 relaxation time is longest, and may be associated with a specific pathological change or tissue property. In addition to the maximum T2 value, the detection system may also calculate an average of all pixel T2 values within the lacrimal boundary, i.e., the average T2 value of the patient's current lacrimal gland. This value reflects the T2 relaxation time characteristics of the lacrimal gland as a whole, helping to assess the overall health status of the lacrimal gland. After calculating the maximum T2 value and the average T2 value, the detection system can verify the result, and the accuracy of calculation is ensured. If outliers or unexpected results are found, the system may re-perform image processing or computation. Finally, the detection system outputs the maximum T2 value and the average T2 value of the current lacrimal gland of the patient in an appropriate format, such as a numerical value, a chart, or a report. These results can provide the physician with objective information about the health status of the lacrimal gland, aiding in diagnosis and treatment decisions of ocular diseases.
As an optional implementation manner, in the embodiment of the application, by superposing the T2 weighted lacrimal boundary image and calculating the maximum T2 value and the average T2 value, more comprehensive and accurate lacrimal evaluation information can be provided for doctors, and the application is beneficial to early discovery of eye diseases and formulation of corresponding treatment schemes.
213. The detection system obtains the average reflectivity of the eye muscles of the patient according to the ratio of the average height of all reflected waves between the front and rear muscle sheaths of the eye muscles to the front scleral peak.
As an alternative implementation, in an embodiment of the present invention, the detection system may first acquire an eye image containing the eye muscle structure. These images may come from ultrasound scanning or other medical imaging techniques. The system then performs the necessary preprocessing on the image to improve the visibility and accuracy of the reflected wave. In the preprocessed image, the detection system may identify reflected waves between the anterior and posterior myofascial of the eye muscle. These reflected waves represent echoes of sound waves in the musculature of the eye, the height and morphology of which may provide information about the musculature of the eye. The system may automatically measure the height of each reflected wave by a specific algorithm or technique. In addition to the reflected waves, the detection system may also locate and measure anterior scleral peaks. Anterior scleral peak is a specific point in the image, typically as a reference point to compare the heights of the reflected waves. The system can accurately identify the position of the anterior scleral peak through an image processing technology and measure the related parameters thereof. After obtaining the height of all reflected waves and the measurement of the anterior scleral peak, the detection system calculates the ratio of each reflected wave to the anterior scleral peak. These ratios reflect the intensity or height of the reflected wave relative to the anterior scleral peak. Finally, the detection system can calculate the average value of the ratio of all reflected waves to anterior scleral peak to obtain the average reflectivity of the eye muscles of the patient. This average represents a quantitative indicator of the overall reflex properties of the eye muscles and can be used to assess the health of the eye muscles or to compare differences between different individuals.
As an optional implementation manner, in the embodiment of the application, by obtaining the average reflectivity of the eye muscles of the two eyes of the patient, a doctor can more objectively evaluate the health condition of the eye muscles, discover potential eye diseases or abnormalities in time and formulate a corresponding treatment scheme. This helps to improve the diagnostic accuracy and therapeutic effect of ocular diseases.
214. The detection system acquires the content of hypertonic tear pro-inflammatory cytokines in the tears of the eyes of the patient; wherein, the content of the hypertonic tear pro-inflammatory cytokines at least comprises interleukin-1 beta content, tumor necrosis factor-alpha content and matrix metalloproteinase-9 content.
As an alternative embodiment, in the examples of the present application, the present application may be used to achieve quantitative detection of cytokines in tears by immunoassay methods, such as enzyme-linked immunosorbent assay (ELISA) or immunofluorescence. These methods utilize specific antibodies to bind to the target cytokine and indirectly determine the amount of cytokine by measuring the amount of the bound product.
As an alternative implementation mode, in the embodiment of the invention, interleukin-1 beta is an important proinflammatory cytokine, and participates in a plurality of links of ocular inflammation, and the content of the interleukin-1 beta can be measured through a specific antibody aiming at the IL-1 beta, so that the accuracy and the specificity of a result are ensured. TNF-alpha is another key pro-inflammatory factor, playing an important role in ocular inflammation, and the quantitative detection of tumor necrosis factor-alpha can be similarly performed by using high-specificity antibodies. Matrix metalloproteinase-9 is an enzyme capable of degrading extracellular matrix and is closely related to tissue damage caused by ocular inflammation, and the content of matrix metalloproteinase-9 can be determined by a specific method to evaluate the effect of ocular inflammation on tissue structure.
As an alternative embodiment, in embodiments of the present invention, the detection system may perform data analysis on the measured cytokine content to generate a corresponding result report. These results will be compared to normal reference values to assess whether the level of hypertonic tear pro-inflammatory cytokines in the patient's tear is abnormal. Based on these results, the physician can diagnose and prescribe a treatment for ocular inflammation in combination with the patient's clinical manifestations and other examination results.
215. The detection system acquires the content of immunoregulatory molecules in the blood of a patient; wherein the content of the immunoregulatory molecule at least comprises the content of intercellular adhesion molecules -1.
As an alternative embodiment, in the examples of the present invention, the immunoregulatory molecules play a key role in maintaining the immune homeostasis of the body and regulating the immune response. Intercellular adhesion molecules -1, as one of them, are involved in the migration, adhesion and inflammatory processes of immune cells. In the pathogenesis of TAO, abnormal expression of intercellular adhesion molecules -1 is associated with infiltration of immune cells and inflammatory lesions of orbital tissues. Therefore, the detection system can indirectly reflect the activity state of the TAO by acquiring the intercellular adhesion molecule -1 content in the blood of the patient. When the intercellular adhesion molecule -1 content is increased, it may mean that the mobility of TAO is increased, the inflammatory reaction is aggravated, and the orbital tissue is more damaged. The detection result can provide important basis for doctors to evaluate TAO illness state, formulate treatment scheme and predict prognosis.
As an alternative implementation manner, in the embodiment of the present invention, since the activity assessment of TAO is a comprehensive process, besides the intercellular adhesion molecule -1 content, the system needs to combine the detection results, clinical manifestations, imaging examination and other information of other immunoregulatory molecules to perform comprehensive judgment so as to ensure accurate assessment of the activity of TAO.
216. The detection system substitutes the patient eye data information into an eye score calculation formula model to calculate the eye data score of the patient.
In the embodiment of the present invention, regarding the eyeball prominence deformation score, the extraocular rectus deformation score, the coronal lacrimal gland area deformation score, the cross-sectional lacrimal gland area deformation score, the lacrimal gland maximum T2 value score, the lacrimal gland average T2 value score, the oculoplastus average reflectivity score, the interleukin-1β content score, the tumor necrosis factor- α content score, the matrix metalloproteinase-9 content score, and the intercellular adhesion molecule -1 content score, the detection system may use a calculation formula: deformation score = (initial score +.patient current measurement +.x deformation value), the system can evaluate the patient's ocular disease level based on the calculated deformation score, and this score can also help the doctor to see if the patient's ocular condition is severe and if further treatment or monitoring is needed.
For example, if the initial eye protrusion score is 10 minutes, the current patient has a 25 mm eye protrusion value and a 14 mm normal eye protrusion value, the following formula is adopted: (initial eye protrusion score ++patient eye protrusion value) = (10++25) = (25-14) = 4.4 score, then patient eye protrusion deformation score is 4.4. Finally, the formula is utilized: the patient's binocular saliency score = initial saliency score-saliency deformation score = 10-4.4 = 5.6, it is clear that the higher the saliency deformation score, the lower the patient's binocular saliency score, if the patient's saliency deformation score is high.
As an alternative embodiment, in an embodiment of the present invention, first, the detection system obtains an initial score for the patient's eye. These initial scores are based on statistics or expert criteria for normal populations and represent ideal values for eyes in a healthy state. They provide a reference basis for subsequent calculations. Next, the detection system may calculate, according to a calculation formula, an eye protrusion deformation score, an extraocular rectus deformation score, a coronal lacrimal gland area deformation score, a cross-sectional lacrimal gland area deformation score, a lacrimal gland maximum T2 value score, a lacrimal gland average T2 value score, an ocular muscle average reflectivity score, an interleukin-1 beta content score, a tumor necrosis factor-alpha content score, a matrix metalloproteinase-9 content score, and an intercellular adhesion molecule -1 content score, respectively. The scores comprehensively consider the initial state, the actual measured value and the difference value of the eyes of the patient, and can reflect the change condition of the eyes of the patient. Next, the detection system may calculate a patient binocular eye prominence score, a patient binocular extraocular rectus muscle thickness score, a coronal lacrimal gland area score, a cross-sectional lacrimal gland area score, a lacrimal gland maximum T2 value score, a lacrimal gland average T2 value score, an ocular muscle average reflectivity score, a patient binocular tear score, and an intercellular adhesion molecule -1 content score by subtracting the corresponding scores from the initial scores. For the lacrimal area score of both eyes of the patient, the detection system can sum the lacrimal area score of the coronal surface and the lacrimal area score of the cross section. For the lacrimal gland T2 value scores of the eyes of the patient, the detection system can add the lacrimal gland maximum T2 value score and the lacrimal gland average T2 value. For the patient's binocular tear score, the initial score may be subtracted from the interleukin-1 beta content score, tumor necrosis factor-alpha content score, and matrix metalloproteinase-9 content score, one by one.
217. The detection system respectively adds the scoring data in the eye data scoring of the patient according to a certain proportion relation to calculate the initial TAO scoring value of the current patient.
218. The detection system accumulates and divides each current patient initial TAO score value over a specified period of time to calculate a patient target TAO score value.
219. The detection system determines the current TAO activity status of the patient after combining the patient target TAO score value, hyperthyroidism eye disease clinical activity score value, and thyroid function and antibody detection report to obtain a TAO composite score.
As an alternative implementation manner, in the embodiment of the invention, each scoring data in the eye data score of the patient can be used as a basis for systematically and intelligently evaluating the thyroid-related eye diseases, and a more comprehensive and accurate evaluation result can be provided for a doctor, so that the doctor can know the lacrimal gland, the eye muscle, the tears, the eyeball salience and the pathological changes of the immune system of the patient at different layers, evaluate whether the abnormal or pathological changes exist in the eye functions of the patient according to the scoring, and formulate a corresponding treatment scheme.
It can be seen that implementing another method for intelligently detecting thyroid-related eye diseases described in fig. 2 can improve the diagnosis efficiency of thyroid-related eye diseases.
In addition, by implementing another method for intelligently detecting thyroid-related eye diseases described in fig. 2, the deformation condition of eyes of a patient can be comprehensively considered, and a more comprehensive and accurate evaluation index is provided.
Example III
Referring to fig. 3, fig. 3 is a schematic structural diagram of a detection system according to an embodiment of the invention. As shown in fig. 3, the detection system 300 may include a first computing unit 301, a second computing unit 302, a third computing unit 303, and a fourth computing unit 304, wherein:
The first calculating unit 301 is configured to substitute the patient eye data information into an eye score calculating formula model to calculate a patient eye data score.
A second calculating unit 302, configured to add the score data in the eye data scores of the patient according to a certain proportional relationship, so as to calculate an initial TAO score value of the current patient; the patient eye data score at least comprises a patient binocular eyeball salience score, a patient binocular extraocular rectus muscle thickness score, a patient binocular ocular muscle reflectivity score, a patient binocular lacrimal gland area score, a patient binocular lacrimal gland T2 value score, a patient binocular lacrimal fluid score and a patient blood immunoregulatory molecule score.
The third calculating unit 303 is configured to accumulate the initial TAO score value of each current patient in a specified time period, and divide the accumulated initial TAO score value by the specified time period to calculate the target TAO score value of the patient.
A fourth calculation unit 304 is configured to determine the current TAO activity status of the patient after combining the target TAO score value of the patient, the clinical activity score value of hyperthyroidism eye disease, and the thyroid function and antibody detection report to obtain a TAO composite score.
As an optional implementation manner, in the embodiment of the present application, the eye score calculation formula model of the first calculation unit 301 is formulated by a relevant medical expert or researcher according to the research purpose evaluation standard of thyroid-related eye diseases, where not only one calculation formula is contained in the eye score calculation formula model, but a plurality of calculation formulas can be set in the eye score calculation formula model, and each calculation formula corresponds to one score in the eye data scores of the patient, and the present application is not limited in any way.
As an alternative implementation, in an embodiment of the present invention, the second calculation unit 302 may calculate the current initial TAO score value using the initial TAO score value= (the score of score item 1 x the weight of score item 1 in the patient's eye data score) + (the score of score item 2 x the weight of score item 2 in the patient's eye data score) +.
As an alternative implementation, in an embodiment of the present invention, the first computing unit 301 may first obtain scores of eye data of the patient, where the data may include scores of multiple aspects of protrusion of eyeballs, eye muscle functions, eye inflammation, and the like. Each score reflects the severity of a patient on a particular ocular symptom or sign. Subsequently, the second calculation unit 302 may perform weighting processing on the scoring data according to a predetermined proportional relationship. These scaling relationships are typically derived based on clinical experience and study data to ensure that the weights of the scores in the final calculation match their importance to TAO condition assessment. After the weighting process is completed, the third computing unit 303 adds the scoring data according to the proportion relationship, so as to obtain a comprehensive initial TAO scoring value. The grading value can comprehensively reflect the current eye disease condition of the patient and provide objective and quantitative evaluation basis for doctors.
In an alternative embodiment, the second calculating unit 302 may calculate the initial TAO score value by adding the scoring data of the patient's eye data score according to a certain proportional relationship, so as to provide an objective and quantized condition evaluation basis for the doctor.
As an alternative embodiment, in an embodiment of the present application, the present application may calculate the current patient initial TAO score value once every certain width of the time window (e.g., 1 day). According to the monitoring of the eye area state of the patient, variability data of the eye area of one period (7 days) is obtained, analysis is carried out to obtain whether the TAO score value of the user is maintained within a certain range or not in the period, and then the third calculation unit 303 performs calculation to obtain a target TAO score value so as to ensure the diagnosis accuracy of the thyroid-related eye disease.
As an optional implementation mode, in the embodiment of the invention, a plurality of indexes can be fused to accurately evaluate the current TAO activity condition of the patient, so that the diagnosis efficiency of the thyroid-related eye diseases is improved, the misdiagnosis condition can be reduced, and meanwhile, the method is suitable for various crowds, has the characteristics of convenience, rapidness and long-time continuous monitoring, and enables a main doctor to more quickly and accurately know the eye region condition of the patient and timely discover the eye disease problem.
As an optional implementation manner, in the embodiment of the application, a multiple scoring mode is adopted to detect and score most lesion types presented in the active period of the thyroid-related eye diseases, and then the hyperthyroidism eye disease clinical activity scoring value (CAS) and the thyroid function and antibody detection report are combined to improve the detection accuracy.
As an alternative implementation, in an embodiment of the present invention, the fourth calculation unit 304 may assign a weight to the target TAO score value, the hyperthyroidism eye disease clinical activity score value, and the thyroid function and antibody detection report, respectively, which may be determined according to medical study and expert opinion. Subsequently, the fourth calculation unit 304 may calculate TAO composite score by using TAO composite score= (patient target TAO score value x weight of patient target TAO score value) + (hyperthyroidism eye disease clinical activity score value x weight of hyperthyroidism eye disease clinical activity score value) + (thyroid function and antibody detection report score x weight of thyroid function and antibody detection report score), and then explain the eye health condition of the patient with reference to a preset score range or standard by means of the TAO composite score, and different score ranges may correspond to different disease severity levels or require different treatment strategies. Wherein, because there is information about the thyroid function status of the patient, such as the levels of Thyroid Stimulating Hormone (TSH), free thyroxine (FT 4), etc., in the thyroid function and antibody detection report, the system may preset a scoring standard of the thyroid function and antibody detection report, that is, a score determined according to the information about the thyroid function status of each patient, for example, hyperthyroidism or hypothyroidism patient, generally, the change of TSH is opposite to FT3 and FT4, such as typical hyperthyroidism patient, TSH is lower than the normal range, and FT3 and FT4 are higher than the normal range. Whereas typical hypothyroidism patients, TSH is above the normal range, FT3, FT4 are below the normal range. For hyperthyroidism patients, for example, if a positive score is assigned when Thyroid Stimulating Hormone (TSH) is below the normal range of values, then a positive score should be assigned when the other first work index is above the normal range of values, so that the score is not offset. I.e. when Thyroid Stimulating Hormone (TSH) is above the conventional value range, a positive score is conferred; negative scores are assigned when Thyroid Stimulating Hormone (TSH) is below the normal value range; and carrying out other data of the thyroid function state of the patient in a similar way, then accumulating all scores, and calculating to obtain thyroid function and antibody detection report scores.
It can be seen that implementing the detection system depicted in fig. 3 can improve the diagnostic efficiency of thyroid-related eye diseases.
In addition, the detection system described in fig. 3 is implemented, so that the activity condition of the current TAO of the patient can be known more accurately, and powerful support is provided for clinical decision.
Example IV
Referring to fig. 4, fig. 4 is a schematic structural diagram of another detection system according to an embodiment of the present invention. Wherein the detection system of fig. 4 is optimized from the detection system of fig. 3. In comparison to the detection system of fig. 3, the detection system of fig. 4 further comprises:
The first segmentation unit 305 is configured to segment the horizontal eye region and the coronal plane eye region in the fused scan image by using the U-net neural network model before the first calculation unit 301 substitutes the patient eye data information into the eye score calculation formula model to calculate the patient eye data score and after the single photon emission computed tomography image and the electronic computed tomography image are fused to obtain the fused scan image.
And an extraction unit 306, configured to extract vertex coordinates of the anterior surface of the cornea and coordinates of the most protruding point of the lateral edge of the orbit from the horizontal eye region and the coronal eye region by using a coordinate system comparison method, so as to calculate a protrusion value of the eyeballs of the patient.
In comparison with the detection system of fig. 3, the first calculation unit 301 of fig. 4 includes:
A first calculation subunit 3011 for multiplying a quotient obtained by dividing the initial eyeball saliency score by the eyeball saliency value of both eyes of the patient by the eyeball saliency deformation value of the patient to calculate an eyeball saliency deformation score; the deformation value of the eyeball salience of the patient is the difference value between the value of the eyeball salience of the two eyes of the patient and the value of the normal eyeball salience.
The second calculating subunit 3012 is configured to subtract the eye protrusion deformation score from the eye protrusion initial score to calculate a patient binocular eye protrusion score.
In comparison with the detection system of fig. 3, the first dividing unit 305 of fig. 4 includes:
the training subunit 3051 is configured to train the initial model of the U-net neural network by using the medical image dataset with the label, so as to obtain a model of the U-net neural network; wherein the medical image dataset contains accurate segmentation annotations for the horizontal eye region and the coronal plane eye region.
An input subunit 3052 is configured to input the preprocessed fused scan image into the U-net neural network model, so as to generate a segmentation mask to mark a target region in the fused scan image.
A quantization subunit 3053, configured to quantize the accuracy of the segmentation mask using the specified evaluation index to determine a horizontal bit eye region and a coronal bit plane eye region.
As an alternative implementation, in the embodiment of the present application, the process of splitting using the U-Net neural network model is approximately as follows: the fused scan image can be first subjected to the necessary pre-processing, such as normalization, denoising, etc., to improve the segmentation performance of the model, and then the training subunit 3051 can train the U-Net model using the labeled medical image dataset. These datasets should contain accurate segmentation labels for the horizontal eye region and the coronal eye region, and the input subunit 3052 inputs the preprocessed fused scan image into the trained U-Net model, which generates a segmentation mask to mark the target region in the image, and performs post-processing on the segmentation result generated by the model, such as smoothing edges, removing small regions, etc., to improve the accuracy of segmentation, and finally the quantization subunit 3053 may use appropriate evaluation indexes (such as Dice coefficients, ioU, etc.) to quantize the accuracy of the segmentation result.
In comparison to the detection system of fig. 3, the extraction unit 306 of fig. 4 comprises:
the identification subunit 3061 is configured to identify vertex pixel coordinates of the anterior surface of the cornea of both eyes from the horizontal eye region and most prominent point pixel coordinates of the lateral edge of the orbit from the coronal eye region.
The third calculation subunit 3062 is configured to calculate, using a distance calculation formula, a linear distance from the vertex of the anterior surface of the cornea to the most protruding point of the lateral edge of the orbit after converting the pixel coordinates of the vertex of the anterior surface of the cornea and the most protruding point of the lateral edge of the orbit into spatial coordinates.
The fourth calculating subunit 3063 is configured to calculate the protrusion value of the eyes of the patient by substituting the vertex space coordinates of the front surface of the cornea of the eyes, the space coordinates of the most protruding points of the outer edge of the orbit, and the linear distance with a trigonometric function.
In comparison to the detection system of fig. 3, the detection system of fig. 4 further comprises:
A selecting unit 307, configured to, before the first calculating unit 301 substitutes the patient's eye data information into the eye score calculating formula model to calculate the patient's eye data score, and the extracting unit 306 extracts vertex coordinates of anterior surfaces of the two eyes cornea and coordinates of most protruding points of lateral edges of the orbit from the horizontal eye region and the coronal plane eye region by using a coordinate system comparison method, so as to calculate the protrusion value of the eyes of the patient, and then select, as the target eye image layer, the image layer with the most radionuclide uptake from the coronal plane eye region.
The measurement unit 308 is configured to measure a pixel distance on an image perpendicular to a long axis of a muscle at a midpoint position of the patient's external rectus muscle after dividing the patient's external rectus muscle from the target eye image plane, and perform spatial size conversion on the pixel distance to obtain a patient's external rectus muscle thickness value.
In comparison with the detection system of fig. 3, the first calculation unit 301 of fig. 4 includes:
As an alternative implementation, in an embodiment of the present invention, the first calculating subunit 3011 is further configured to multiply the quotient obtained by dividing the initial extraocular rectus thickness score by the extraocular rectus thickness value of the patient by the extraocular rectus deformation value of the patient, so as to calculate an extraocular rectus deformation score; wherein, the deformation value of the external rectus muscle of the eyes of the patient is the difference value between the thickness value of the external rectus muscle of the eyes of the patient and the thickness value of the normal external rectus muscle of the eyes;
As an alternative implementation, in an embodiment of the present invention, the second calculating subunit 3012 is further configured to subtract the extraocular rectus muscle deformability score from the extraocular rectus muscle thickness initial score to calculate an extraocular rectus muscle thickness score of both eyes of the patient.
In comparison to the detection system of fig. 3, the detection system of fig. 4 further comprises:
A second segmentation unit 309, configured to, before the first calculation unit 301 substitutes the patient's eye data information into the eye score calculation formula model to calculate the patient's eye data score, measure the pixel distance on the image perpendicular to the long axis of the muscle at the midpoint position of the extraocular rectus muscle of the patient, and perform spatial dimension conversion on the pixel distance to obtain the extraocular rectus muscle thickness value of the patient's eyes, and then segment the lacrimal gland boundary of the patient from the fused scan image according to the different slice directions and spatial position information in the fused scan image; wherein, the lacrimal boundary at least comprises a patient coronal lacrimal boundary and a patient cross-sectional lacrimal boundary.
The fifth calculating unit 310 is configured to multiply the number of pixels within the boundary of the patient's coronal lacrimal gland by the actual area represented by each pixel, and multiply the number of pixels within the boundary of the patient's cross-sectional lacrimal gland by the actual area represented by each pixel, so as to calculate the area of the patient's coronal lacrimal gland and the area of the patient's cross-sectional lacrimal gland, respectively.
In comparison with the detection system of fig. 3, the first calculation unit 301 of fig. 4 includes:
As an alternative implementation, in an embodiment of the present invention, the first calculating subunit 3011 is further configured to multiply a quotient obtained by dividing the initial coronal lacrimal gland area score by the patient's coronal lacrimal gland area by the coronal lacrimal gland area deformation value, so as to calculate the coronal lacrimal gland area deformation score; wherein, the deformation value of the area of the patient's coronal lacrimal gland is the difference between the area of the patient's coronal lacrimal gland and the area of the normal coronal lacrimal gland.
As an alternative implementation, in an embodiment of the present invention, the second calculating subunit 3012 is further configured to subtract the coronal lacrimal gland area deformation score from the coronal lacrimal gland area initial score to calculate a patient coronal lacrimal gland area score.
As an alternative implementation, in an embodiment of the present invention, the first calculating subunit 3011 is further configured to multiply the quotient obtained by dividing the cross-sectional lacrimal gland area initial score by the cross-sectional lacrimal gland area of the patient by the cross-sectional lacrimal gland area deformation value to calculate the cross-sectional lacrimal gland area deformation score; wherein, the deformation value of the lacrimal gland area of the cross section of the patient is the difference value between the lacrimal gland area of the cross section of the patient and the lacrimal gland area of the normal cross section.
As an alternative implementation, in an embodiment of the present invention, the second calculating subunit 3012 is further configured to subtract the cross-sectional lacrimal area deformation score from the cross-sectional lacrimal area initial score to calculate the patient cross-sectional lacrimal area score.
As an alternative implementation, in an embodiment of the present invention, the second calculating subunit 3012 is further configured to calculate the lacrimal gland area score of both eyes of the patient by adding the lacrimal gland area score of the coronal surface of the patient to the lacrimal gland area score of the cross section of the patient.
In comparison to the detection system of fig. 3, the detection system of fig. 4 further comprises:
A first obtaining unit 311, configured to obtain a T2 weighted value of each pixel in the lacrimal boundary and form a T2 weighted lacrimal boundary image by using a short T reverse order MRI technique after the first calculating unit 301 substitutes the patient ocular data information into the ocular score calculation formula model to calculate the patient ocular data score, and the fifth calculating unit 310 multiplies the number of pixels in the lacrimal boundary of the patient cross section by the actual area represented by each pixel to calculate the lacrimal area of the patient coronary surface and the lacrimal area of the patient cross section, respectively.
A superimposing unit 312 for superimposing the T2 weighted lacrimal boundary image on the T2 weighted image on the coronal plane eye region after the process of registering and aligning the T2 weighted image on the coronal plane eye region and the T2 weighted lacrimal boundary image to obtain a patient's current lacrimal maximum T2 value and a patient's current lacrimal average T2 value.
In comparison with the detection system of fig. 3, the first calculation unit 301 of fig. 4 includes:
As an alternative implementation, in an embodiment of the present invention, the first calculating subunit 3011 is further configured to calculate a lacrimal maximum T2 value score by dividing a quotient obtained by dividing the lacrimal maximum T2 value initial score by the current lacrimal maximum T2 value of the patient by the lacrimal maximum T2 value; wherein the difference between the maximum lacrimal gland T2 value and the maximum lacrimal gland T2 value is the difference between the current lacrimal gland T2 value and the normal lacrimal gland T2 value of the patient.
As an alternative implementation, in an embodiment of the present invention, the second calculating subunit 3012 is further configured to subtract the lacrimal gland maximum T2 value score from the lacrimal gland maximum T2 value initial score to calculate a lacrimal gland maximum T2 value score for the patient.
As an alternative implementation, in an embodiment of the present invention, the first calculating subunit 3011 is further configured to calculate a lacrimal average T2 value score by multiplying a quotient obtained by dividing the lacrimal average T2 value initial score by the current lacrimal average T2 value of the patient by a difference of the lacrimal average T2 value; wherein the difference in mean T2 values of the lacrimal glands is the difference between the mean T2 values of the lacrimal glands of the patient and the mean T2 values of the normal lacrimal glands.
As an alternative implementation, in an embodiment of the present invention, the second calculating subunit 3012 is further configured to subtract the lacrimal average T2 value score from the lacrimal average T2 value initial score to calculate a lacrimal average T2 value score for the patient.
As an alternative implementation, in an embodiment of the present invention, the second calculating subunit 3012 is further configured to calculate the T2 value score of the lacrimal gland of the patient by adding the maximum T2 value score of the lacrimal gland of the patient to the average T2 value score of the lacrimal gland of the patient.
In comparison to the detection system of fig. 3, the detection system of fig. 4 further comprises:
The second obtaining unit 313 is configured to, before the first calculating unit 301 substitutes the patient's eye data information into the eye score calculating formula model to calculate the patient's eye data score, and the superimposing unit 312 superimposes the T2 weighted lacrimal boundary image on the T2 weighted image on the coronal plane eye region to obtain a ratio of the current lacrimal maximum T2 value of the patient to the current lacrimal average T2 value of the patient, and obtain the average reflectivity of both eye muscles of the patient according to the ratio of the average height of all reflected waves between the anterior and posterior muscle sheaths of the eye muscle to the anterior scleral peak.
In comparison with the detection system of fig. 3, the first calculation unit 301 of fig. 4 includes:
As an alternative implementation, in an embodiment of the present invention, the first calculating subunit 3011 is further configured to multiply a quotient obtained by dividing the initial ocular muscle reflectivity score by the average ocular muscle reflectivity of both eyes of the patient by the average ocular muscle reflectivity difference to calculate an average ocular muscle reflectivity score; wherein, the average reflectivity difference of the eye muscle is the difference between the average reflectivity of the eye muscle of the two eyes of the patient and the average reflectivity of the normal eye muscle.
As an alternative implementation, in an embodiment of the present invention, the second calculating subunit 3012 is further configured to subtract the average reflectivity score of the eye muscle from the initial reflectivity score of the eye muscle to calculate the reflectivity score of the eye muscle of both eyes of the patient.
In comparison to the detection system of fig. 3, the detection system of fig. 4 further comprises:
A third obtaining unit 314, configured to obtain the content of the hypertonic tear pro-inflammatory cytokine in the tears of the patient after the first calculating unit 301 substitutes the eye data information of the patient into the eye score calculating formula model to calculate the eye score of the patient, and the second obtaining unit 313 obtains the average reflectivity of the eyes of the patient according to the ratio of the average height of all reflected waves between the front and back muscle sheaths of the eyes to the front scleral peak; wherein, the content of the hypertonic tear pro-inflammatory cytokines at least comprises interleukin-1 beta content, tumor necrosis factor-alpha content and matrix metalloproteinase-9 content.
As an alternative embodiment, in an embodiment of the present invention, the third obtaining unit 313 is further configured to obtain the content of immunoregulatory molecules in the blood of the patient; wherein the content of the immunoregulatory molecule at least comprises the content of intercellular adhesion molecules -1.
In comparison with the detection system of fig. 3, the first calculation unit 301 of fig. 4 includes:
As an alternative implementation, in an embodiment of the present invention, the first calculating subunit 3011 is further configured to multiply a quotient obtained by dividing the tear initial score by the interleukin-1β content difference value, so as to calculate an interleukin-1β content score; wherein the difference in interleukin-1 beta content is the difference between the interleukin-1 beta content and the normal interleukin-1 beta content.
As an alternative implementation, in an embodiment of the present invention, the first calculating subunit 3011 is further configured to multiply a quotient obtained by dividing the tear initial score by the tumor necrosis factor- α content difference value, so as to calculate a tumor necrosis factor- α content score; wherein the difference in the levels of tumor necrosis factor-alpha is the difference between the levels of tumor necrosis factor-alpha and the levels of normal tumor necrosis factor-alpha.
As an alternative embodiment, in the embodiment of the present invention, the first calculating subunit 3011 is further configured to multiply the quotient obtained by dividing the tear initial score by the content of matrix metalloproteinase-9 by the difference in the content of matrix metalloproteinase-9, so as to calculate the content score of matrix metalloproteinase-9; wherein the difference in the content of matrix metalloproteinase-9 is the difference between the content of matrix metalloproteinase-9 and the content of normal matrix metalloproteinase-9.
As an alternative embodiment, in the embodiment of the present invention, the second calculating subunit 3012 is further configured to subtract the interleukin-1 β content score, the tumor necrosis factor- α content score and the matrix metalloproteinase-9 content score from the tear initial score one by one to calculate the tear score of the eyes of the patient.
Dividing the initial score of the immunoregulatory molecule by the quotient of the intercellular adhesion molecule -1 content multiplied by the intercellular adhesion molecule -1 content difference to calculate an intercellular adhesion molecule -1 content score; wherein the difference in intercellular adhesion molecule -1 content is the difference between the intercellular adhesion molecule -1 content and the normal intercellular adhesion molecule -1 content.
In an alternative embodiment, the second calculating subunit 3012 is further configured to subtract the score of the intercellular adhesion molecule -1 from the initial score of the immunoregulatory molecule to calculate the score of the immunoregulatory molecule in the blood of the patient.
It can be seen that implementing the detection system depicted in fig. 4 can improve the diagnostic efficiency of thyroid-related eye diseases.
In addition, the detection system described in fig. 4 can be implemented to comprehensively consider the deformation condition of the eyes of the patient, and a more comprehensive and accurate evaluation index is provided.
Example five
Referring to fig. 5, fig. 5 is a schematic structural diagram of another detection system according to an embodiment of the invention. As shown in fig. 5, the detection system may include:
A memory 501 in which executable program codes are stored;
A processor 502 coupled to the memory 501;
The processor 502 invokes executable program codes stored in the memory 501 to execute any one of the methods of fig. 1-3 for intelligently detecting thyroid-related eye diseases.
The embodiment of the invention discloses a computer readable storage medium which stores a computer program, wherein the computer program enables a computer to execute any one of the intelligent detection methods of thyroid-related eye diseases shown in fig. 1-3.
The embodiments of the present invention also disclose a computer program product, wherein the computer program product, when run on a computer, causes the computer to perform some or all of the steps of the method as in the method embodiments above.
Those of ordinary skill in the art will appreciate that all or part of the steps of the various methods of the above embodiments may be implemented by hardware associated with a program that may be stored in a computer-readable storage medium, including Read-Only Memory (ROM), random-access Memory (Random Access Memory, RAM), programmable Read-Only Memory (Programmable Read-Only Memory, PROM), erasable programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), one-time programmable Read-Only Memory (OTPROM), electrically erasable programmable Read-Only Memory (EEPROM), compact disc Read-Only Memory (Compact Disc Read-Only Memory, CD-ROM), or other optical disk Memory, magnetic disk Memory, tape Memory, or any other medium that can be used to carry or store data.
The method and the system for intelligently detecting the thyroid-related eye diseases disclosed by the embodiment of the invention are described in detail, and specific examples are applied to illustrate the principle and the implementation mode of the invention, and the description of the above examples is only used for helping to understand the method and the core idea of the invention; meanwhile, as those skilled in the art will vary in the specific embodiments and application scope according to the idea of the present invention, the present disclosure should not be construed as limiting the present invention in summary.
Claims (10)
1. A method for intelligently detecting thyroid-related eye diseases, comprising:
substituting the eye data information of the patient into an eye score calculation formula model to calculate the eye data score of the patient;
Adding the scoring data in the eye data scoring of the patient according to a certain proportion relation to calculate the initial TAO scoring value of the current patient; wherein the eye data score of the patient at least comprises a patient binocular eyeball salience score, a patient binocular extraocular rectus muscle thickness score, a patient binocular ocular muscle reflectivity score, a patient binocular lacrimal gland area score, a patient binocular lacrimal gland T2 value score, a patient binocular lacrimal fluid score and a patient blood immunoregulatory molecule score;
accumulating the initial TAO scoring value of each current patient in a specified time period, and dividing the accumulated initial TAO scoring value by the specified time period to calculate a target TAO scoring value of the patient;
After combining the patient target TAO score value, hyperthyroidism eye disease clinical activity score value, and thyroid function and antibody detection report to obtain a TAO composite score, the patient's current TAO activity status is determined.
2. The method of claim 1, wherein prior to substituting the patient's ocular data information into the ocular score calculation formula model to calculate the patient's ocular data score, the method further comprises:
after fusing a single photon emission computed tomography image and an electronic computed tomography image to obtain a fused scanning image, utilizing a U-net neural network model to segment a horizontal eye region and a coronal bit plane eye region in the fused scanning image;
Extracting vertex coordinates of the front surface of the cornea and coordinates of the most protruding points of the outer edge of the eyesocket of the eyes from the horizontal eye region and the coronal plane eye region by adopting a coordinate system comparison method so as to calculate the protrusion degree value of eyeballs of the eyes of the patient;
And substituting the patient ocular data information into an ocular score calculation formula model to calculate a patient ocular data score, comprising:
Multiplying the quotient obtained by dividing the initial eyeball saliency score by the eyeball saliency value of both eyes of the patient by the eyeball saliency deformation value of the patient to calculate an eyeball saliency deformation score; wherein the patient's eye protrusion deformation value is the difference between the patient's eye protrusion value and the normal eye protrusion value;
Subtracting the eye protrusion deformation score from the eye protrusion initial score to calculate the patient's binocular eye protrusion score.
3. The method for intelligently detecting thyroid-related eye diseases according to claim 2, wherein the segmenting the horizontal eye region and the coronal eye region in the fused scan image by using the U-net neural network model comprises:
Training the initial model of the U-net neural network by using the medical image data set with the labels to obtain the U-net neural network model; wherein the medical image dataset comprises accurate segmentation annotations of the horizontal eye region and the coronal plane eye region;
Inputting the preprocessed fusion scanning image into the U-net neural network model to generate a segmentation mask to mark a target region in the fusion scanning image;
Quantizing the accuracy of the segmentation mask with a specified evaluation index to determine the horizontal bit eye region and the coronal bit plane eye region;
And extracting vertex coordinates of the front surface of the cornea and coordinates of the most protruding points of the outer edge of the orbit of the patient from the horizontal eye region and the coronal eye region by adopting a coordinate system comparison method to calculate the protrusion value of the eyeballs of the patient, wherein the method comprises the following steps:
Identifying from the horizontal eye region vertex pixel coordinates of the anterior surface of the cornea of both eyes and from the coronal eye region most prominent point pixel coordinates of the lateral edge of the orbit;
After the vertex pixel coordinates of the front surface of the cornea and the most protruding point pixel coordinates of the outer edge of the eye socket are converted into space coordinates, calculating the linear distance from the vertex of the front surface of the cornea to the most protruding point of the outer edge of the eye socket by using a distance calculation formula;
Substituting the vertex space coordinates of the front surface of the cornea of the eyes, the space coordinates of the most protruding points of the outer edge of the eyebox and the linear distance by using a trigonometric function to calculate the protrusion value of the eyeballs of the eyes of the patient.
4. The method according to claim 2, wherein before substituting the patient's eye data information into the eye score calculation formula model to calculate the patient's eye data score, and after extracting vertex coordinates of anterior surfaces of the cornea and most prominent point coordinates of lateral edges of eyeholes from the horizontal eye region and the coronal plane eye region by using the coordinate system comparison method to calculate the patient's eye protrusion value, the method further comprises:
selecting an image layer with the most radionuclide uptake from the coronal plane eye area as a target eye image layer;
After the external rectus muscle of the eyes of the patient is divided from the target eye image layer, measuring the pixel distance on the image perpendicular to the long axis of the muscle at the midpoint position of the external rectus muscle of the eyes of the patient, and performing space size conversion on the pixel distance to obtain the thickness value of the external rectus muscle of the eyes of the patient;
And substituting the patient ocular data information into an ocular score calculation formula model to calculate a patient ocular data score, comprising:
Multiplying the quotient obtained by dividing the initial extraocular rectus thickness score by the extraocular rectus thickness value of both eyes of the patient by the extraocular rectus deformation value of the patient to calculate an extraocular rectus deformation score; wherein the patient's extraocular rectus muscle deformation value is the difference between the patient's extraocular rectus muscle thickness value and the normal extraocular rectus muscle thickness value;
Subtracting the extraocular rectus muscle deformability score from the extraocular rectus muscle thickness initial score to calculate the patient extraocular rectus muscle thickness score.
5. The method according to claim 4, wherein before substituting the patient's eye data information into the eye score calculation formula model to calculate the patient's eye data score, and after measuring the pixel distance on the image perpendicular to the long axis of the muscle at the midpoint of the external rectus muscle of the patient's eyes and converting the pixel distance in space to obtain the external rectus muscle thickness value of the patient's eyes, the method further comprises:
Dividing the lacrimal gland boundary of the patient from the fusion scanning image according to different slice directions and spatial position information in the fusion scanning image; wherein the lacrimal boundary at least comprises a patient coronal lacrimal boundary and a patient cross-section lacrimal boundary;
Multiplying the number of pixels within the patient's coronal lacrimal gland boundary by the actual area represented by each pixel, and multiplying the number of pixels within the patient's cross-sectional lacrimal gland boundary by the actual area represented by each pixel to calculate a patient's coronal lacrimal gland area and a patient's cross-sectional lacrimal gland area, respectively;
And substituting the patient ocular data information into an ocular score calculation formula model to calculate a patient ocular data score, comprising:
Multiplying the quotient obtained by dividing the initial coronal lacrimal gland area score by the patient's coronal lacrimal gland area by the coronal lacrimal gland area deformation value to calculate a coronal lacrimal gland area deformation score; wherein the deformation value of the area of the patient's coronal lacrimal gland is the difference between the area of the patient's coronal lacrimal gland and the area of the normal coronal lacrimal gland;
subtracting the coronal lacrimal gland area deformation score from the coronal lacrimal gland area initial score to calculate a patient coronal lacrimal gland area score;
Multiplying the cross-sectional lacrimal gland area deformation value by a quotient obtained by dividing the cross-sectional lacrimal gland area initial score by the patient cross-sectional lacrimal gland area to calculate a cross-sectional lacrimal gland area deformation score; wherein the patient cross-sectional lacrimal gland area deformation value is the difference between the patient cross-sectional lacrimal gland area and a normal cross-sectional lacrimal gland area;
Subtracting the cross-sectional lacrimal gland area deformation score from the cross-sectional lacrimal gland area initial score to calculate a patient cross-sectional lacrimal gland area score;
And adding the area score of the patient coronary lacrimal gland with the area score of the lacrimal gland of the patient cross section to calculate the area score of the lacrimal gland of the patient eyes.
6. The method of claim 5, wherein the substituting the patient's ocular data information into an ocular score calculation formula model to calculate the patient's ocular data score is preceded by multiplying the number of pixels within the patient's cross-sectional lacrimal boundary by the actual area represented by each pixel to calculate the patient's coronal lacrimal area and the patient's cross-sectional lacrimal area, respectively, the method further comprising:
Acquiring a T2 weighting value of each pixel in the lacrimal boundary by adopting a short T reversal sequence MRI technology, and forming a T2 weighted lacrimal boundary image;
After the process of registering and aligning the T2 weighted lacrimal boundary image and the T2 weighted image on the coronal plane eye region, superimposing the T2 weighted lacrimal boundary image on the T2 weighted image on the coronal plane eye region to obtain a patient's current lacrimal maximum T2 value and a patient's current lacrimal average T2 value;
And substituting the patient ocular data information into an ocular score calculation formula model to calculate a patient ocular data score, comprising:
Dividing the initial lacrimal maximum T2 value score by the quotient of the current lacrimal maximum T2 value of the patient multiplied by the difference of the lacrimal maximum T2 value to calculate a lacrimal maximum T2 value score; wherein the difference between the maximum lacrimal gland T2 value and the maximum lacrimal gland T2 value is the difference between the current lacrimal gland T2 value and the normal lacrimal gland T2 value of the patient;
Subtracting the lacrimal gland maximum T2 value score from the lacrimal gland maximum T2 value initial score to calculate a patient lacrimal gland maximum T2 value score;
dividing the initial lacrimal average T2 value score by the current lacrimal average T2 value of the patient by the difference of the lacrimal average T2 value to calculate a lacrimal average T2 value score; wherein the difference in mean lacrimal T2 value is the difference between the mean lacrimal T2 value of the patient and the mean normal lacrimal T2 value;
subtracting the lacrimal average T2 value score from the lacrimal average T2 value initial score to calculate a patient lacrimal average T2 value score;
And adding the maximum T2 value score of the lacrimal gland of the patient with the average T2 value score of the lacrimal gland of the patient, and calculating the T2 value score of the lacrimal gland of both eyes of the patient.
7. The method of claim 6, wherein prior to substituting the patient's ocular data information into the ocular score calculation formula model to calculate the patient's ocular data score and after superimposing the T2 weighted lacrimal boundary image on the T2 weighted image on the coronal plane ocular region to obtain the patient's current lacrimal maximum T2 value and the patient's current lacrimal average T2 value, the method further comprises:
Obtaining the average reflectivity of the eye muscles of the patient according to the ratio of the average height of all reflected waves between the front and rear muscle sheaths of the eye muscles to the front scleral peak;
And substituting the patient ocular data information into an ocular score calculation formula model to calculate a patient ocular data score, comprising:
Dividing the initial ocular muscle reflectivity score by the average ocular muscle reflectivity of both eyes of the patient multiplied by the average ocular muscle reflectivity difference to calculate an average ocular muscle reflectivity score; wherein the average reflectivity difference of the eye muscles is the difference between the average reflectivity of the eye muscles of the eyes of the patient and the average reflectivity of the normal eye muscles;
subtracting the eye muscle average reflectivity score from the eye muscle reflectivity initial score to calculate the patient binocular eye muscle reflectivity score.
8. The method according to claim 7, wherein the substituting the patient's eye data information into an eye score calculation formula model to calculate the patient's eye data score, and the obtaining the average reflectivity of both eye muscles of the patient based on the ratio of the average height of all reflected waves between the anterior and posterior muscle sheaths of the eye muscles to the anterior scleral peak, further comprises:
Acquiring the content of hypertonic tear pro-inflammatory cytokines in tears of two eyes of a patient; wherein, the content of the hypertonic tear pro-inflammatory cytokines at least comprises interleukin-1 beta content, tumor necrosis factor-alpha content and matrix metalloproteinase-9 content;
obtaining the content of immunoregulatory molecules in the blood of a patient; wherein the content of the immunoregulatory molecules at least comprises the content of intercellular adhesion molecules -1;
And substituting the patient ocular data information into an ocular score calculation formula model to calculate a patient ocular data score, comprising:
Multiplying the quotient obtained by dividing the tear initial score by the interleukin-1 beta content difference to calculate an interleukin-1 beta content score; wherein the interleukin-1 beta content difference is the difference between the interleukin-1 beta content and the normal interleukin-1 beta content;
Multiplying the quotient of the tear initial score divided by the tumor necrosis factor-alpha content by a tumor necrosis factor-alpha content difference to calculate a tumor necrosis factor-alpha content score; wherein the difference in the levels of tumor necrosis factor-alpha is the difference between the levels of tumor necrosis factor-alpha and normal levels of tumor necrosis factor-alpha;
Multiplying the quotient obtained by dividing the tear initial score by the matrix metalloproteinase-9 content difference to calculate a matrix metalloproteinase-9 content score; wherein the difference in the content of matrix metalloproteinase-9 is the difference between the content of matrix metalloproteinase-9 and the content of normal matrix metalloproteinase-9;
Subtracting the interleukin-1 beta content score, the tumor necrosis factor-alpha content score and the matrix metalloproteinase-9 content score from the tear initial score one by one to calculate a binocular tear score of the patient;
Dividing the initial score of the immunoregulatory molecule by the intercellular adhesion molecule -1 content to obtain a quotient multiplied by the intercellular adhesion molecule -1 content difference to calculate an intercellular adhesion molecule -1 content score; wherein the difference in intercellular adhesion molecule -1 content is the difference between the intercellular adhesion molecule -1 content and the normal intercellular adhesion molecule -1 content;
Subtracting the intercellular adhesion molecule -1 content score from the immunoregulatory molecule initial score to calculate an immunoregulatory molecule score in the patient's blood.
9. A detection system, the detection system comprising:
The first calculation unit is used for substituting the eye data information of the patient into an eye score calculation formula model so as to calculate the eye data score of the patient;
The second calculation unit is used for respectively adding the scoring data in the eye data scoring of the patient according to a certain proportion relation so as to calculate the initial TAO scoring value of the current patient; wherein the eye data score of the patient at least comprises a patient binocular eyeball salience score, a patient binocular extraocular rectus muscle thickness score, a patient binocular ocular muscle reflectivity score, a patient binocular lacrimal gland area score, a patient binocular lacrimal gland T2 value score, a patient binocular lacrimal fluid score and a patient blood immunoregulatory molecule score;
A third calculation unit, configured to accumulate each of the current patient initial TAO score values in a specified time period, and divide the accumulated value by the specified time period, so as to calculate a patient target TAO score value;
And a fourth calculation unit, configured to determine a current TAO activity condition of the patient after combining the target TAO score value, the hyperthyroidism eye disease clinical activity score value, and the thyroid function and antibody detection report to obtain a TAO comprehensive score.
10. A detection system, the detection system comprising:
A memory storing executable program code;
A processor coupled to the memory;
the processor invokes the executable program code stored in the memory to perform the method of intelligently detecting thyroid-related eye disease of any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410324037.3A CN117936101A (en) | 2024-03-21 | 2024-03-21 | Method and system for intelligently detecting thyroid-related eye diseases |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410324037.3A CN117936101A (en) | 2024-03-21 | 2024-03-21 | Method and system for intelligently detecting thyroid-related eye diseases |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117936101A true CN117936101A (en) | 2024-04-26 |
Family
ID=90750875
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410324037.3A Pending CN117936101A (en) | 2024-03-21 | 2024-03-21 | Method and system for intelligently detecting thyroid-related eye diseases |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117936101A (en) |
-
2024
- 2024-03-21 CN CN202410324037.3A patent/CN117936101A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10076299B2 (en) | Systems and methods for determining hepatic function from liver scans | |
| EP2661734B1 (en) | Automatic quantification of asymmetry | |
| Fletcher et al. | Regional brain tissue composition in children with hydrocephalus: Relationships with cognitive development | |
| CN103249358B (en) | Medical image-processing apparatus | |
| US20120116219A1 (en) | System and method of ultrasound image processing | |
| CN111932492B (en) | Medical image processing method and device and computer readable storage medium | |
| US10729389B2 (en) | 3D assessment of conjugant eye deviation for the identificaiton of acute ischemic stroke | |
| CN108151670A (en) | A kind of back three dimensional symmetry degree evaluating apparatus and method | |
| Sanfilippo et al. | Optic disc morphology-Rethinking shape | |
| Soleimani et al. | Carotid artery wall motion estimation from consecutive ultrasonic images: Comparison between block-matching and maximum-gradient algorithms | |
| Yang et al. | Hybrid attention for automatic segmentation of whole fetal head in prenatal ultrasound volumes | |
| Koprowski et al. | Assessment of significance of features acquired from thyroid ultrasonograms in Hashimoto's disease | |
| CN111028278A (en) | Method for providing human body joint data based on tomography technology | |
| Luo et al. | The edge detectors suitable for retinal OCT image segmentation | |
| Wan et al. | A novel system for measuring pterygium's progress using deep learning | |
| Xu et al. | Automatic classification of male and female skeletal muscles using ultrasound imaging | |
| CN116452593B (en) | Method, device and system for constructing AI evaluation model of vascular cognitive disorder | |
| CN115762787B (en) | Eyelid disease operation curative effect evaluation method and system | |
| US20230157658A1 (en) | Quantification of noncalcific and calcific valve tissue from coronary ct angiography | |
| CN117936101A (en) | Method and system for intelligently detecting thyroid-related eye diseases | |
| Jin et al. | Comparison of deep learning and classical image processing for skin segmentation | |
| US11481905B2 (en) | Atlas for automatic segmentation of retina layers from OCT images | |
| Tameem et al. | Morphological atlases of knee cartilage: shape indices to analyze cartilage degradation in osteoarthritic and non-osteoarthritic population | |
| KR20200121550A (en) | Method for evaluating foot type and device evaluating foot type using the same | |
| US20220398763A1 (en) | Systems and methods of volumetrically assessing structures of skeletal cavities |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |