CN101623191A - Device and method for noninvasively detecting property of stomach tissue - Google Patents
Device and method for noninvasively detecting property of stomach tissue Download PDFInfo
- Publication number
- CN101623191A CN101623191A CN200910090920A CN200910090920A CN101623191A CN 101623191 A CN101623191 A CN 101623191A CN 200910090920 A CN200910090920 A CN 200910090920A CN 200910090920 A CN200910090920 A CN 200910090920A CN 101623191 A CN101623191 A CN 101623191A
- Authority
- CN
- China
- Prior art keywords
- tissue
- image
- stomach
- stomach tissue
- infrared
- 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.)
- Granted
Links
Images
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention relates to a device and a method for noninvasively detecting the property of the stomach tissue. The device comprises a light source part, a flexible part extended into a human body, a light spectrum collection device, a computer, and the like. Stomach images are acquired by a micro image acquisition device and sent to the computer; after incident infrared light passes through a secondary attenuated total reflection (ART) probe, the absorption information of the stomach tissue is transmitted to the in-vitro light spectrum collection device, converted into an electrical signal and then sent to the computer; and after a mode identification module distinguishes and analyzes the images and the light spectrum information of the stomach tissue, the mode identification module outputs a detected tissue property. The method comprises the following steps: before measuring and after establishing a distinguishing and analyzing model by utilizing the images and light spectrums of the in-vivo stomach tissue, acquiring the images and the light spectrums of a part to be detected and inputting the images and the light spectrums to the established distinguishing and analyzing model to obtain an analyzing result of the stomach tissue of a patient to be detected. The invention can realize in-vivo, noninvasive, quick and multi-point measurement, avoid missed detection, reduce the pain of the patients, establish the distinguishing and analyzing model by combining the images and full light spectrum information and has high accuracy.
Description
Technical field
The present invention relates to the biological spectrum analysis technical field, relate in particular to a kind of stomach tissue character noninvasive detection device and method.
Background technology
Cancer is a class disease of serious threat human health even life.Gastric cancer is one of modal malignant tumor, and its mortality rate is in the first place that China occupies malignant tumor, and influencing it, to treat topmost factor be stadium, and 5 years survival rates of early gastric cancer can reach more than 90%.At present, Gastric Diseases by Spraying is checked that effective method is a fibergastroscope, gastroscope can be understood the position and the scope of pathological changes down, and can draw materials to suspicious lesions and carry out pathological diagnosis.In clinical gastroscope testing result, the general very case of small scale is a cancer, and the patient of the overwhelming majority then suffers from various chronic inflammatory diseases, and if inflammatory disorders is treated control timely and made regular check on monitoring, inflammation then may further worsen, and dysplasia takes place, and finally develops into cancer.Correct realize apace that the gastroscope sample is normal, the clinical diagnosis in superficial gastritis, atrophic gastritis or the gastric cancer, and take different therapeutic schemes accordingly, significant for the survival rate and the cure rate that improve sufferer.
At present, though gastroscopy is widely adopted clinically, as patent CN 1004530250 described a kind of endoscopic systems.But gastroscopy still has certain limitation, as needs gripping stomach tissue, thereby wound, easy infection are arranged, easily causes diffusion, and is same, because need gripping tissue, is difficult for realizing so the gripping inspection is carried out at all doubtful positions, so may causes omission.The main examinate's of gastroscopic diagnostic result operating experience influence and subjective judgment, thereby degree of accuracy is lower.Organize formalin fixed after the gripping, paraffin embedding, HE dyeing carries out routine pathology at last and learn and check, thereby need the long period just can obtain diagnostic result, and cost is higher.
Fourier transform infrared spectroscopy (Fourier Transform Infrared Spectroscopy is called for short FTIR) is a kind of new method from molecular level postgraduate fabric texture that development in recent years is got up.Infrared spectrum is the sensitive probe of molecule structure change, can detect the variation of composition, structure and the conformation of intracellular matter from the variation of the corresponding group molecular vibration of some material level in the cell.Therefore, the early stage variation of pathological tissues shows on the infrared spectrum when not being embodied in morphologic variation therein to some extent.Though and some present academic researchs also relate to this field, after needing mostly to gather tissue samples, be positioned over the detection of exsomatizing on the spectrogrph, promptly, hope replaces pathologic finding with the spectrum inspection, needs the gripping tissue so equally, thereby also just has its drawback.The laboratory academic research that minority can be differentiated tissue property at body also all is at the human body superficial tissue, promptly reflects near the pathological changes situation of the tissue skin.And present spectral investigation also still rests on the characteristic spectrum peak comparison stage, wasted the bulk information of full spectrum, and too relied on subjective judgment, thereby precision is not high.
In sum, a kind of stomach tissue character noninvasive detection device and method combine spectroscopic analysis methods with image analysis method, are expected to develop into a kind of noinvasive, the new technique of stomach tissue character detection fast, and can be used for the earlier detection of cancer.
Summary of the invention
Technology of the present invention is dealt with problems: overcome the deficiencies in the prior art, a kind of stomach tissue character noninvasive detection device and method are provided, be implemented in body, multimetering, can avoid omission, do not need gripping patient's stomach tissue to carry out pathological analysis, reduced the patient suffering, avoided cross infection and diffusion, image has been combined with the full gloss spectrum information, programmed by mathematical algorithm, automatically differentiate doubtful tissue property, have higher accuracy.
Technical solution of the present invention: a kind of stomach tissue character noninvasive detection device comprises with the lower part: the Lights section, stretch into the intravital soft component of patient, spectra collection device, computer; The Lights section comprises two parts: infrared incident illumination light source in the used light source of the illumination of visible waveband being provided and providing; Stretching into the intravital soft component of patient comprises: visible light incident optical, miniature image harvester, picture signal outlet line; In infrared incident optical, secondary extinction total reflection (ATR) probe, in infrared outgoing optical fiber, wherein secondary ATR probe is positioned at the miniature image harvester and stretches into the intravital soft component of patient bottom, in infrared incident optical with in infrared emergent ray form Y shape fibre bundle and pop one's head in secondary ATR and be connected; It is flexible to stretch into the intravital soft component of patient, and control by operating flexible part, can make its left and right sides all directions bending up and down, helps its tract that passes through the gastric bending by the direction of regulating bent angle, improves overview function;
The visible light that visible light source sends enters the visible light incident optical and transfers in patient's body after Lens Coupling, illuminate stomach tissue, so that the miniature image harvester is gathered the stomach image, and image information is converted into the signal of telecommunication is sent to external computer by the picture signal outlet line;
Infrared incident optical and transferring to after the inherent secondary ATR probe of patient's body inner surface carries out twice attenuated total reflectance during the mid-infrared light that send in the mid-infrared light source enters after Lens Coupling, the emergent light load Biochemical Information of stomach tissue after outgoing optical fiber is sent to the computer part after reaching external spectra collection device and being converted into spectral signal;
Flexible by controlling, the intravital soft component of patient is stretched in rotation, the miniature image harvester is gathered the image of stomach various piece tissue in real time, after finding the suspected lesion tissue, start the spectra collection device and gather the spectral signal at this place, herein image information and spectral information are saved simultaneously, be transferred to the computer part as input, computer can and show in real time with image and spectrographic storage, simultaneously, the various character stomach tissue spectra databases that pattern recognition module utilization in the computer has been set up in advance, image information and spectral information to this place's tissue carry out discriminant analysis, and it is normal that output that can be real-time detects the organization discrimination tissue, superficial gastritis, the differentiation of atrophic gastritis or gastric cancer.
Described visible part light source is selected halogen tungsten lamp or white light LEDs for use, and described mid-infrared light source is a siliconit; Be equipped with convex lens respectively in two light source exits, the light that respectively two light sources sent focuses on and is coupled into incident optical separately, through after the Lens Coupling, visible light source and mid-infrared light source advance respectively visible incident optical and in infrared incident optical.
Described visible light incident optical is selected common quartzy material for use.
In described infrared incident optical with in infrared outgoing fiber selection optical fiber material be germanium salt glass, logical optical band is 700-4000cm
-1
It is 45 degree taper ZnSe or Ge crystal that described secondary extinction is launched (ATR) probe entirely, and its section is an isosceles right triangle.
Described miniature image harvester adopts CCD or CMOS sensitization mode.
Described spectra collection device adopts back beam split mode, load the stomach tissue property information in infrared emergent light, extremely external by fiber-optic transfer, after collimated, by being accepted by detector after the light splitting part beam split in the spectra collection device, and being converted into spectral signal, light splitting part can adopt the AOTF crystal, variable filter or grating.
Described pattern recognition module, this module is carried out discriminant analysis by the spectrum and the image information of input to stomach tissue character.Discriminant analysis is a class identification and classification method in the multivariate statistical method, and it is the observation sample result according to two or more parents, according to certain criterion and corresponding discriminant function, differentiates arbitrary individual which kind of parent of ownership of waiting to declare.
Utilize tissue spectrum that has obtained and the tissue property of having differentiated to set up the discriminant analysis model, discriminant analysis is a comparatively sophisticated class identification and classification method in the multivariate statistical method, it is the observation sample result according to two or more parents, according to certain criterion and corresponding discriminant function, differentiate arbitrary individual which kind of parent of ownership of waiting to declare.Its method comprises: the k-nearest neighbor method, and the artificial neural network method, support vector machine methods etc. are multiple, and are not limited to above discriminant analysis method.
K nearest-neighbors method adopts vector space model to classify, and notion is the stomach tissue spectrum similarity height each other of identical category, and can assess the possible classification of unknown classification case by the similarity of calculating with the known class case.As the case may be, choose the nearest k of sample to be discriminated training set sample spectrum, establish in this k sample the sample N that this does not comprise normal, superficial gastritis, atrophic gastritis and gastric cancer
1, N
2, N
3, and N
1+ N
2+ N
3=k.Computational discrimination sample spectrum is to the score that is subordinate to of four class training set samples respectively, formula such as following represented:
D in the formula
iRepresent that i training set sample spectrum is to the distance between the sample spectrum to be discriminated in this classification.D
iIn order to weigh the similarity degree between sample light, can wait with Euclidean distance, mahalanobis distance, city block distance and calculate, and be not limited to above several distance calculating method.
Calculate sample spectra to be discriminated to all kinds of scores after, relatively the score height is judged to be the highest classification of score with sample to be discriminated.
(Artificial Neural Network is to the abstract of human brain or the some fundamental characteristics of natural neutral net (NaturalNeural Network) and simulation ANN) to artificial neural network.Artificial neural network is based on to the physiological Study achievement of brain, and its purpose is to simulate some mechanism and mechanism of brain, realizes the function of certain aspect.ANN is by having the extensive in parallel network that adaptive simple unit is formed, and its tissue can be simulated the cross reaction that biological nervous system has been done real world objects.Use the identification of Artificial Neural Network implementation pattern.It is very complicated to handle some environmental informations, and background knowledge is unclear, and the indefinite problem of rule of inference allows sample spectra that bigger distortion is arranged, and its speed of service is fast, and adaptive performance is good, has higher resolution.
The method that a kind of stomach tissue character noinvasive detects may further comprise the steps:
A. before measurement, gather enough different pathological analysis results at body stomach tissue image and spectrum, set up the discriminant analysis model by chemometrics method;
B. utilize the miniature image harvester to obtain the image information of training set patient stomach tissue;
C. according to the doubtful part of spectral discrimination, gather and write down the image and the spectral information at doubtful position simultaneously;
D. in the discriminant analysis model that the input step a that the image information at doubtful position combined with spectral information sets up, the analysis result that obtains waiting diagnosing patient's stomach tissue is a kind of for normal, superficial gastritis, atrophic gastritis or gastric cancer.
Described steps A specifically comprises:
A. recruit a large amount of trial volunteers, require stomach among the experimenter close for the ratio of normal, superficial gastritis, atrophic gastritis or gastric cancer;
B. soft component is stretched in experimenter's stomach, gather the image of a part, utilize this position tissue spectrum of sampling collecting fiber simultaneously by the miniature image harvester;
C. spectrum position tissue is gathered in gripping, carries out pathological analysis, determines that its character is normal, a kind of in superficial gastritis, atrophic gastritis or the gastric cancer;
D. utilize the tissue spectrum that obtains through device described in the present invention with image information with utilize pathological analysis to obtain the result, by chemometrics method to spectrum set up normally, the disaggregated model of superficial gastritis, atrophic gastritis or gastric cancer four classes,
Step C is according to the doubtful part of spectral discrimination, gather and write down doubtful station diagram picture and spectral information simultaneously, it is characterized in that: utilize the soft component of corresponding device of the present invention to go deep into patient's stomach to be detected, by controlling its pitching or rotation, gather the image of each position of stomach in real time, when finding that pathological tissues can show as ulcer last, be that basilar part has special performance, uneven, hard sense, expose sense, perilesional mucosa has the performance of concentrating picture and fusion phenomenon or the chap of pestle shape, then begins to write down by the miniature image harvester image of this position, notes the image and the spectroscopic data at this position simultaneously.
The present invention compared with prior art has following advantage:
(1) traditional checkout gear and method need utilize the gripping pincers to gather stomach tissue to external, can bring misery to the patient, may cause diffusion simultaneously and infect, and too much collection organize everywhere also and can not realize, thereby cause omission easily.And the present invention adopts soft component to stretch in patient's body, spectrum and the image of gathering patient tissue on body, noinvasive ground carry out discriminant analysis, thereby need not gripping patient's stomach tissue, only need to gather tissue image and spectrum, can not cause suffering and injure the patient, reduce the patient suffering, and be difficult for cross infection, caused diffusion.Simultaneously, also can realize multimetering, can detect easily, can avoid omission position that might pathological changes.
(2) traditional checkout gear and method need the gripping stomach tissue to external, organize formalin fixed after the gripping, paraffin embedding, and HE dyeing carries out routine pathology at last and learn and check, thereby need the long period just can obtain diagnostic result, and cost is higher.And the middle infrared spectrum that intravital soft component gos deep into gathering in patient's body stomach tissue is goed deep in utilization of the present invention, middle infrared spectrum is the character of characterising biological tissue delicately, therefore can be implemented in body, in site measurement, do not need to carry out pathological analysis, shorten the diagnosis required time, can obtain testing result in real time.
(3) traditional checkout gear and method rely on image to determine tissue property, and if cancerous issue can be observed on iconography then, ten million cancerous tumor cell has at least been arranged.And application of spectral of the present invention is diagnosed, infrared spectrum can be in cell the variation of the corresponding group molecular vibration of some material level detect the spatial arrangements of existence, composition, content and molecule of some material the cell and the variation of structure, thereby the present invention can detect the pathological changes of tissue as early as possible, can be used for the early stage inspection of gastric cancer.
(4) traditional checkout gear and method mainly rely on image detection, in the present research then are to rely on the tissue spectrum that exsomatizes to analyze.And the present invention utilizes image and spectral information as diagnosis basis simultaneously, and quantity of information is bigger, thereby more may obtain right judgement.
(5) in the research at present, be primarily aimed in vitro tissue spectrum characteristics peak and compare, analyze, and the pattern recognition module among the present invention is at setting up the discriminant analysis model at the full spectral coverage spectrum of soma, detect more accurate sensitivity, specificity is higher.
(6) common spectra collection mode is: mid-infrared light enters the mode that skin enters outgoing optical fiber more after reflection by incident optical, and the mid-infrared light penetration depth is very shallow, thereby it is few to carry quantity of information, and organizational information is easy to be submerged in instrument and the background noise.And the present invention adopts secondary extinction total reflection (ATR) probe, has improved spectral signal intensity and signal to noise ratio, for subsequent analysis provides possibility.
Description of drawings
Fig. 1 is a kind of stomach tissue character of the present invention noninvasive detection device composition frame chart;
Fig. 2 is the present invention and the bonded composition frame chart of stomach tissue;
Fig. 3 is a secondary ATR probe profile in apparatus of the present invention;
Fig. 4 is the structure chart of secondary ATR probe in apparatus of the present invention;
Fig. 5 is method flow diagram among the present invention;
Fig. 6 is used AOTF spectrometer architecture figure in the embodiment of the invention;
The stomach tissue image of Fig. 7 for collecting in the embodiment of the invention;
The stomach tissue spectrum of Fig. 8 for collecting in the embodiment of the invention.
Among the figure: 1 the Lights section, 2 visible light sources, 3 mid-infrared light sources, 4 visible light incident opticals, 5 mid-infrared light incident opticals, 6 secondary ATR probe, 7 stomach tissue, 8 mid-infrared light outgoing optical fiber, 9 spectra collection equipment, 10 miniature image collecting devices, 11 image signal transmission circuits, 12 spectral signal transmission lines, 13 computers, 14 pattern recognition modules, 15 stretch into the intravital soft component of patient, the device of 16 fixed fibers and atr crystal, 17 secondary atr crystals.
The specific embodiment
Below in conjunction with the drawings and specific embodiments technical scheme of the present invention is done further and to be elaborated:
As shown in Figures 1 to 4, stomach tissue character noninvasive detection device of the present invention comprises as the lower part:
The visible light that visible light source 2 sends enters visible light incident optical 4 and transfers in patient's body after Lens Coupling, illuminate stomach tissue, so that miniature image 10 harvesters are gathered the stomach image, and image information is converted into the signal of telecommunication is sent to external computer 13 by picture signal 11 outlet lines;
Infrared incident optical 5 and transferring to after the inherent secondary atr crystal of patient's body 17 inner surfacies carry out twice attenuated total reflectance during the mid-infrared light that send in mid-infrared light source 3 enters after Lens Coupling, the emergent light load Biochemical Information of stomach tissue after after outgoing optical fiber 8 reaches external spectra collection device 9 and be converted into spectral signal, be sent to computer part 13 through spectral signal transmission line 12;
Flexible by controlling, the intravital soft component 15 of patient is stretched in rotation, miniature image harvester 10 is gathered the image of stomach various piece tissue in real time, after finding the suspected lesion tissue, start spectra collection device 9 and gather the spectral signal at this place, herein image information and spectral information are saved simultaneously, be transferred to computer part 13 as input, computer can and show in real time with image and spectrographic storage, simultaneously, pattern recognition module 14 in the computer utilizes the various character stomach tissue spectra databases of having set up in advance, image information and spectral information to this place's tissue carry out discriminant analysis, and it is normal that output that can be real-time detects the organization discrimination tissue, superficial gastritis, the differentiation of atrophic gastritis or gastric cancer character.
Visible part light source 2 is selected halogen tungsten lamp for use; Siliconit is selected in mid-infrared light source 3 for use; Be equipped with convex lens respectively at visible part light source 2 and 3 exits, mid-infrared light source, the light that respectively two light sources is sent focuses on and is coupled into incident optical separately, through after the Lens Coupling, visible light source 2 and mid-infrared light source 3 advance respectively visible incident optical 4 with in infrared incident optical 5.
Visible light incident optical 4 is selected common quartzy material for use.
In infrared incident optical 5 with in infrared outgoing optical fiber 8 to select the optical fiber material for use be germanium salt glass, logical optical band is 700-4000cm
-1
It is 45 degree taper ZnSe crystal that secondary extinction is launched (ATR) crystal 17 entirely, and its section is an isosceles right triangle.
It is the back beam splitting type spectrogrph of light splitting part that the spectra collection device adopts AOTF.The population structure block diagram of used AOTF mid-infrared light spectrometer system as shown in Figure 6.Load the stomach tissue property information in infrared emergent light, to external, behind the optically focused colimated light system, enter the AOTF crystal by fiber-optic transfer.Under the effect of ultrasonic frequency, diffraction takes place in AOTF, and zero level transillumination and-1 order diffraction light are radiated at respectively on two middle Infrared Detectorss through focusing on.The optical signal that detector receives changes into direct current signal through interlock circuit, by the single-chip microcomputer computing in the spectrogrph, with the diffraction light signal divided by optical signal transmissive.Because two paths of signals passed through light path much at one, thereby the result of being divided by compensated various interference equal errors in the fluctuating, light path of the intensity of light source, improved certainty of measurement.Digital frequency synthesizer is used for the diffraction light wavelength is carried out continuously or single step scanning, finally generates spectral signal and transfers to computer.
Computer 13 adopts general purpose computer, minimalist configuration P41.7GHz, 512M ram.
Embodiments of the invention adopt support vector machine as Discrimination Analysis Algorithm.Set up by gathering a large amount of experimenter's stomach tissue information before using apparatus of the present invention, module support vector machine sorting algorithm is that the optimal classification face (Optimal Hyperplane) under the linear separability situation proposes.Require classifying face can both correctly classify, require it to satisfy exactly all samples
y
i(w
Tx
i+b)-1≥0，i＝1，2，...，n
Those samples that equal sign is set up are called support vector (Support Vectors).The gap size in the classification space (Margin) of two class samples:
Therefore, optimal classification face problem can be expressed as the minima of following constrained optimization problems.For this reason, the Lagrange function that can be defined as follows:
Wherein, a
i〉=0 is the Lagrange coefficient, and problem is that w and b are asked Lagrange minimum of a function value.By finding the solution this optimization problem, obtain the optimal classification function and be:
Wherein, sgn () is a sign function.
At first the input space is transformed to a higher dimensional space, in this new space, ask for the optimum linearity classifying face then, and this nonlinear transformation is to realize by defining suitable kernel function (inner product function), order by nonlinear transformation Φ:
K(x
i，x
j)＝<Φ(x
i)·Φ(x
j)>
With kernel function K (x
i, x
j) replace the dot product x in the optimal classification plane
i Tx
j, just being equivalent to former feature space has been transformed to a certain new feature space, corresponding discriminant function formula then is:
X wherein
iBe support vector, x is a unknown vector,
It is as follows to adopt the described checkout gear of claim 1 to carry out the specific embodiment of the method that stomach tissue character noinvasive detects:
Recruit trial volunteer totally 118 people, adopt the soft component of device of the present invention to stretch in experimenter's stomach, gather the image of a part, utilize this position tissue spectrum of sampling collecting fiber simultaneously by miniature image harvester wherein;
Spectrum position tissue is gathered in gripping, carries out pathological analysis, verifies gastric cancer 35 examples through the HE of Pathology Deparment, superficial gastritis 31 examples, atrophic gastritis 33 examples, normal gastric mucosa 19 examples;
Set up aforesaid support vector machine discriminant analysis model, extract a patient to be diagnosed again in addition immediately, after checking 8 o'clock evenings before that day, do not take food thing and beverage, no smoking.The previous day, supper was eaten the digestible food of few slag.The probe that will have the optical fiber of sampling stretches in patient's stomach, open the probe light source of being with and illuminate stomach tissue, gather the image of each position of stomach in real time, find that place's tissue shows as ulcer, this place's tissue image assert that then this place is doubtful position shown in Figure of description 7, note the spectrum picture of this place's tissue, gather and note the image and the spectroscopic data at this position simultaneously, this place organizes spectrogram as shown in Figure 8.
With the support vector machine discriminant analysis module that the spectral information and the image information at this place are brought foundation into, judge that this place is gastric cancer.Gripping then should be organized to external at the place, carries out conventional analysis of cases, and the result is consistent with testing result of the present invention.
The above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.
Claims (10)
1, a kind of stomach tissue character noninvasive detection device is characterized in that comprising with the lower part: the Lights section, stretch into the intravital soft component of patient, spectra collection device, computer.The Lights section comprises two parts: the used visible light source of illumination is provided and provides spectra collection required mid-infrared light source; Stretching into the intravital soft component of patient comprises: visible light incident optical, miniature image harvester, picture signal outlet line; In infrared incident optical, secondary extinction total reflection (ATR) probe, in infrared outgoing optical fiber, wherein secondary ATR probe is positioned at the miniature image harvester and stretches into the intravital soft component of patient bottom, in infrared incident optical with in infrared emergent ray form Y shape fibre bundle and pop one's head in secondary ATR and be connected; It is flexible to stretch into the intravital soft component of patient, and control by operating flexible part, can make its left and right sides all directions bending up and down, helps its tract that passes through the gastric bending by the direction of regulating bent angle, improves overview function;
The visible light that visible light source sends enters the visible light incident optical and transfers in patient's body after Lens Coupling, illuminate stomach tissue, so that the miniature image harvester is gathered the stomach image, and image information is converted into the signal of telecommunication is sent to external computer by the picture signal outlet line;
Infrared incident optical and transferring to after the inherent secondary ATR probe of patient's body inner surface carries out twice attenuated total reflectance during the mid-infrared light that send in the mid-infrared light source enters after Lens Coupling, the emergent light load Biochemical Information of stomach tissue after outgoing optical fiber is sent to the computer part after reaching external spectra collection device and being converted into spectral signal;
Flexible by controlling, the intravital soft component of patient is stretched in rotation, the miniature image harvester is gathered the image of stomach various piece tissue in real time, after finding the suspected lesion tissue, start the spectra collection device and gather the spectral signal at this place, herein image information and spectral information are saved simultaneously, be transferred to the computer part as input, computer can and show in real time with image and spectrographic storage, simultaneously, the various character stomach tissue spectra databases that pattern recognition module utilization in the computer has been set up in advance, image information and spectral information to this place's tissue carry out discriminant analysis, and it is normal that real-time output detects the organization discrimination tissue, superficial gastritis, the differentiation of atrophic gastritis or gastric cancer character.
2, stomach tissue character noninvasive detection device according to claim 1, it is characterized in that: described visible part light source is selected halogen tungsten lamp or white light LEDs for use, and described mid-infrared light source is a siliconit; Be equipped with convex lens respectively in two light source exits, the light that respectively two light sources sent focuses on and is coupled into incident optical separately, through after the Lens Coupling, visible light source and mid-infrared light source advance respectively visible incident optical and in infrared incident optical.
3, stomach tissue character noninvasive detection device according to claim 1, it is characterized in that: described visible light incident optical is selected common quartzy material for use.
4, stomach tissue character noninvasive detection device according to claim 1 is characterized in that: in described infrared incident optical with in infrared outgoing fiber selection optical fiber material be germanium salt glass, logical optical band is 700-4000cm-1.
5, stomach tissue character noninvasive detection device according to claim 1 is characterized in that: it is 45 degree taper ZnSe or Ge crystal that described secondary extinction is launched (ATR) probe entirely, and its section is an isosceles right triangle.
6, stomach tissue character noninvasive detection device according to claim 1 is characterized in that: described miniature image harvester adopts CCD or CMOS sensitization mode.
7, stomach tissue character noninvasive detection device according to claim 1, it is characterized in that: described spectra collection device adopts back beam split mode, load the stomach tissue property information in infrared emergent light, extremely external by fiber-optic transfer, after collimated, by being accepted by photoelectric detector after the light splitting part beam split in the spectra collection device, and be converted into the signal of telecommunication, light splitting part can adopt the AOTF crystal, variable filter or grating.
8, adopt the described checkout gear of claim 1 to carry out the method that stomach tissue character noinvasive detects, it is characterized in that, may further comprise the steps:
(1) before measurement, gather enough different pathological analysis results at body stomach tissue image and spectrum, set up the discriminant analysis model by chemometrics method;
(2) utilize the miniature image harvester to obtain the image information of training set patient stomach tissue;
(3), gather and write down the image and the spectral information at doubtful position simultaneously according to the doubtful part of spectral discrimination;
(4) image information at doubtful position is combined with spectral information input step (1) is set up in the good discriminant analysis model, obtains waiting to diagnose the analysis result of patient's stomach tissue, a kind of in normal, superficial gastritis, atrophic gastritis or the gastric cancer.
9, the method that detects of the noinvasive described in the claim 8 is characterized in that: described step (1) before measurement, gather enough different pathological analysis results at body stomach tissue image and spectrum, the method for setting up the discriminant analysis model is as follows:
A. recruit a large amount of trial volunteers, require stomach among the experimenter close for the ratio of normal, superficial gastritis, atrophic gastritis or gastric cancer;
B. the soft component with the described device of claim 1 stretches in experimenter's stomach, gathers the image of a part by miniature image harvester wherein, utilizes this position tissue spectrum of sampling collecting fiber simultaneously;
C. spectrum position tissue is gathered in gripping, carries out pathological analysis, determines a kind of in normal, superficial gastritis, atrophic gastritis or the cancer of its character;
D. utilize tissue spectrum that the described device of claim 1 obtains with image information with utilize pathological analysis to obtain the result, to spectrum set up normally, the disaggregated model of superficial gastritis, atrophic gastritis and cancer four classes.
10, the method that noinvasive described in the claim 8 detects, it is characterized in that: described step (3) is according to the doubtful part of spectral discrimination, gathering and write down the image at doubtful position and the method for spectral information simultaneously is: utilize the soft component of the described device of claim 1 to go deep into patient's stomach to be detected, by controlling its pitching or rotation, gather the image of each position of stomach in real time, when finding that pathological tissues can show as ulcer last, be that basilar part has special performance, uneven, hard sense, expose sense, perilesional mucosa has the performance of concentrating picture and fusion phenomenon or the chap of pestle shape, then begin to write down the image of this position, note the image and the spectroscopic data at this position simultaneously by the miniature image harvester.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100909206A CN101623191B (en) | 2009-08-14 | 2009-08-14 | Device for noninvasively detecting property of stomach tissue |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100909206A CN101623191B (en) | 2009-08-14 | 2009-08-14 | Device for noninvasively detecting property of stomach tissue |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101623191A true CN101623191A (en) | 2010-01-13 |
| CN101623191B CN101623191B (en) | 2011-02-16 |
Family
ID=41519294
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009100909206A Active CN101623191B (en) | 2009-08-14 | 2009-08-14 | Device for noninvasively detecting property of stomach tissue |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101623191B (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012075702A1 (en) * | 2010-12-10 | 2012-06-14 | 广州宝胆医疗器械科技有限公司 | Electronic gastroscope system with infrared thermal scanning function |
| CN102512134A (en) * | 2011-11-17 | 2012-06-27 | 蔡信东 | Intelligent diagnostic enteroscope |
| CN103398948A (en) * | 2013-08-14 | 2013-11-20 | 武汉大学 | ATR (attenuated total reflectance) probe for Fourier transform infrared spectrometer |
| CN103750807A (en) * | 2011-12-30 | 2014-04-30 | 广州宝胆医疗器械科技有限公司 | Electronic gastroscope system with night vision function |
| CN105520719A (en) * | 2016-01-25 | 2016-04-27 | 上海电力学院 | Early cancer cell detection system based on infrared spectrum |
| CN107967946A (en) * | 2017-12-21 | 2018-04-27 | 武汉大学 | Operating gastroscope real-time auxiliary system and method based on deep learning |
| CN108140240A (en) * | 2015-08-12 | 2018-06-08 | 分子装置有限公司 | For automatically analyzing the system and method for the phenotypic response of cell |
| CN108185974A (en) * | 2018-02-08 | 2018-06-22 | 北京数字精准医疗科技有限公司 | A kind of endoscopic fluorescence ultrasound merges radiography navigation system |
| CN108245121A (en) * | 2016-12-29 | 2018-07-06 | 刘小华 | Endoscope and its imaging method and intelligent diagnosis system |
| CN108272437A (en) * | 2017-12-27 | 2018-07-13 | 中国科学院西安光学精密机械研究所 | Spectrum detection system for skin disease diagnosis and classifier model construction method |
| CN108670203A (en) * | 2018-06-01 | 2018-10-19 | 深圳开立生物医疗科技股份有限公司 | A kind of imaging device |
| CN108742490A (en) * | 2018-05-30 | 2018-11-06 | 广州众健医疗科技有限公司 | A kind of spectral detection system of diagnosing gastric cancer gastroscope |
| CN108784642A (en) * | 2018-05-30 | 2018-11-13 | 广州众健医疗科技有限公司 | A kind of spectrum biopsy instrument for early carcinoma of stomach diagnosis |
| CN109330550A (en) * | 2018-10-17 | 2019-02-15 | 孙静 | A kind of ultra-fine scope intelligence check device of Respiratory Medicine spectrum-type |
| CN111358573A (en) * | 2020-02-26 | 2020-07-03 | 西安交通大学医学院第二附属医院 | Hyperspectral imaging surgical area rapid detection and marking device and detection method |
| CN111568348A (en) * | 2020-06-04 | 2020-08-25 | 中国医学科学院生物医学工程研究所 | Optical diagnosis and treatment capsule endoscope with external power supply and control |
| CN112823396A (en) * | 2018-10-02 | 2021-05-18 | 翰林大学产学合作团 | Endoscope device and method for diagnosing gastric lesion based on gastric endoscope image obtained in real time |
| CN113324944A (en) * | 2021-06-08 | 2021-08-31 | 晶格码（青岛）智能科技有限公司 | In-situ on-line infrared spectrum monitoring equipment, system and method for crystallization process control |
| CN114305260A (en) * | 2021-12-20 | 2022-04-12 | 珠海格力电器股份有限公司 | Control method and device of cleaning robot, medium and cleaning robot |
-
2009
- 2009-08-14 CN CN2009100909206A patent/CN101623191B/en active Active
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012075702A1 (en) * | 2010-12-10 | 2012-06-14 | 广州宝胆医疗器械科技有限公司 | Electronic gastroscope system with infrared thermal scanning function |
| CN102512134A (en) * | 2011-11-17 | 2012-06-27 | 蔡信东 | Intelligent diagnostic enteroscope |
| CN103750807A (en) * | 2011-12-30 | 2014-04-30 | 广州宝胆医疗器械科技有限公司 | Electronic gastroscope system with night vision function |
| CN103398948A (en) * | 2013-08-14 | 2013-11-20 | 武汉大学 | ATR (attenuated total reflectance) probe for Fourier transform infrared spectrometer |
| CN103398948B (en) * | 2013-08-14 | 2015-09-16 | 武汉大学 | A kind of ATR for Fourier transform infrared spectrometer pops one's head in |
| CN108140240A (en) * | 2015-08-12 | 2018-06-08 | 分子装置有限公司 | For automatically analyzing the system and method for the phenotypic response of cell |
| CN105520719A (en) * | 2016-01-25 | 2016-04-27 | 上海电力学院 | Early cancer cell detection system based on infrared spectrum |
| CN108245121A (en) * | 2016-12-29 | 2018-07-06 | 刘小华 | Endoscope and its imaging method and intelligent diagnosis system |
| CN107967946B (en) * | 2017-12-21 | 2021-05-11 | 武汉楚精灵医疗科技有限公司 | Gastroscope operation real-time auxiliary system and method based on deep learning |
| CN107967946A (en) * | 2017-12-21 | 2018-04-27 | 武汉大学 | Operating gastroscope real-time auxiliary system and method based on deep learning |
| CN108272437A (en) * | 2017-12-27 | 2018-07-13 | 中国科学院西安光学精密机械研究所 | Spectrum detection system for skin disease diagnosis and classifier model construction method |
| CN108185974A (en) * | 2018-02-08 | 2018-06-22 | 北京数字精准医疗科技有限公司 | A kind of endoscopic fluorescence ultrasound merges radiography navigation system |
| CN108742490A (en) * | 2018-05-30 | 2018-11-06 | 广州众健医疗科技有限公司 | A kind of spectral detection system of diagnosing gastric cancer gastroscope |
| CN108784642A (en) * | 2018-05-30 | 2018-11-13 | 广州众健医疗科技有限公司 | A kind of spectrum biopsy instrument for early carcinoma of stomach diagnosis |
| CN108670203A (en) * | 2018-06-01 | 2018-10-19 | 深圳开立生物医疗科技股份有限公司 | A kind of imaging device |
| CN112823396A (en) * | 2018-10-02 | 2021-05-18 | 翰林大学产学合作团 | Endoscope device and method for diagnosing gastric lesion based on gastric endoscope image obtained in real time |
| CN109330550A (en) * | 2018-10-17 | 2019-02-15 | 孙静 | A kind of ultra-fine scope intelligence check device of Respiratory Medicine spectrum-type |
| CN109330550B (en) * | 2018-10-17 | 2021-01-15 | 孙静 | Breathe superfine scope intelligent inspection device of spectrum formula for internal medicine |
| CN111358573A (en) * | 2020-02-26 | 2020-07-03 | 西安交通大学医学院第二附属医院 | Hyperspectral imaging surgical area rapid detection and marking device and detection method |
| CN111358573B (en) * | 2020-02-26 | 2023-06-20 | 西安交通大学医学院第二附属医院 | Device and method for rapidly detecting and marking hyperspectral imaging operation area |
| CN111568348A (en) * | 2020-06-04 | 2020-08-25 | 中国医学科学院生物医学工程研究所 | Optical diagnosis and treatment capsule endoscope with external power supply and control |
| CN113324944A (en) * | 2021-06-08 | 2021-08-31 | 晶格码（青岛）智能科技有限公司 | In-situ on-line infrared spectrum monitoring equipment, system and method for crystallization process control |
| CN114305260A (en) * | 2021-12-20 | 2022-04-12 | 珠海格力电器股份有限公司 | Control method and device of cleaning robot, medium and cleaning robot |
| CN114305260B (en) * | 2021-12-20 | 2022-11-01 | 珠海格力电器股份有限公司 | Control method and device of cleaning robot, medium and cleaning robot |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101623191B (en) | 2011-02-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101623191B (en) | Device for noninvasively detecting property of stomach tissue | |
| AU2019232890B2 (en) | System and method for serum based cancer detection | |
| CN102088906A (en) | Improved method and apparatus for determination of a measure of a glycation end-product or disease state using tissue fluorescence | |
| CN105190262B (en) | Diagnostic instrments and method | |
| US10560643B2 (en) | Systems and methods for hyperspectral imaging | |
| CN104541153B (en) | Relate to the use of the method that the endoscope of fiber Raman spectrum carries out real-time cancer diagnosis | |
| CN1097445C (en) | Method and apparatus for tissue type recognition | |
| CN103645170B (en) | A kind of device utilizing Raman spectrum to detect tumour character | |
| US20030026762A1 (en) | Bio-spectral imaging system and methods for diagnosing cell disease state | |
| US20110028808A1 (en) | Method and apparatus for examination of cancer, systemic lupus erythematosus (sle), or antiphospholipid antibody syndrome using near-infrared light | |
| US20150003699A1 (en) | Methods and arrangements for identifying dermatological diagnoses with clinically negligible probabilties | |
| US20110059016A1 (en) | Optical assay system with a multi-probe imaging array | |
| CN102762978A (en) | Method and apparatus to detect coronary artery calcification or disease | |
| JPS59207131A (en) | Method and apparatus for examination and analysis of spectrum permeability | |
| JP2005521531A (en) | Spectroscopic analysis of tissues to detect diabetes | |
| CA2429045A1 (en) | A diagnostic system for the ear | |
| JP2006138860A (en) | Optical microprobe and spectral analysis method of material | |
| Shaikh et al. | A comparative evaluation of diffuse reflectance and Raman spectroscopy in the detection of cervical cancer | |
| Fabelo et al. | Surgical aid visualization system for glioblastoma tumor identification based on deep learning and in-vivo hyperspectral images of human patients | |
| Ding et al. | Diverse spectral band-based deep residual network for tongue squamous cell carcinoma classification using fiber optic Raman spectroscopy | |
| CN107941782A (en) | Can endoscopic fiber Raman microprobe and detection device | |
| Chang et al. | Deep learning methods for oral cancer detection using Raman spectroscopy | |
| Marenzi et al. | FPGA High Level Synthesis for the classification of skin tumors with hyperspectral images | |
| JP2021502574A (en) | Systems, constructs and methods for detecting spectral biomarkers and patterns of disease from stool samples | |
| CN107303174A (en) | A kind of internet+spectrum clinical tumor medical diagnostic method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |