CN111862014A - ALVI automatic measurement method and device based on left and right ventricle segmentation - Google Patents
ALVI automatic measurement method and device based on left and right ventricle segmentation Download PDFInfo
- Publication number
- CN111862014A CN111862014A CN202010651711.0A CN202010651711A CN111862014A CN 111862014 A CN111862014 A CN 111862014A CN 202010651711 A CN202010651711 A CN 202010651711A CN 111862014 A CN111862014 A CN 111862014A
- Authority
- CN
- China
- Prior art keywords
- image
- maximum
- cranium
- alvi
- medical
- 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
- 210000005240 left ventricle Anatomy 0.000 title claims abstract description 79
- 210000005241 right ventricle Anatomy 0.000 title claims abstract description 79
- 230000011218 segmentation Effects 0.000 title claims abstract description 42
- 238000000691 measurement method Methods 0.000 title claims abstract description 28
- 238000007917 intracranial administration Methods 0.000 claims abstract description 50
- 238000005259 measurement Methods 0.000 claims abstract description 50
- 210000004556 brain Anatomy 0.000 claims abstract description 40
- 238000000034 method Methods 0.000 claims abstract description 31
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 26
- 238000003709 image segmentation Methods 0.000 claims abstract description 26
- 238000004891 communication Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 16
- 238000001514 detection method Methods 0.000 claims description 14
- 238000002059 diagnostic imaging Methods 0.000 claims description 10
- 210000003625 skull Anatomy 0.000 claims description 8
- 238000003708 edge detection Methods 0.000 claims description 7
- 210000001103 thalamus Anatomy 0.000 claims description 6
- 210000004720 cerebrum Anatomy 0.000 claims description 5
- 210000001154 skull base Anatomy 0.000 claims description 4
- 230000009466 transformation Effects 0.000 claims description 4
- 210000003140 lateral ventricle Anatomy 0.000 abstract description 4
- 230000002861 ventricular Effects 0.000 description 25
- 238000002595 magnetic resonance imaging Methods 0.000 description 14
- 230000008569 process Effects 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 238000000605 extraction Methods 0.000 description 8
- 238000012805 post-processing Methods 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000002591 computed tomography Methods 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 4
- 208000003906 hydrocephalus Diseases 0.000 description 4
- 238000003384 imaging method Methods 0.000 description 4
- 230000002490 cerebral effect Effects 0.000 description 3
- 238000013170 computed tomography imaging Methods 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000012886 linear function Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000035945 sensitivity Effects 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000001356 surgical procedure Methods 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 241000561734 Celosia cristata Species 0.000 description 1
- 238000005481 NMR spectroscopy Methods 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000001638 cerebellum Anatomy 0.000 description 1
- 210000001175 cerebrospinal fluid Anatomy 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 210000001520 comb Anatomy 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000010339 dilation Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 210000001951 dura mater Anatomy 0.000 description 1
- 230000009982 effect on human Effects 0.000 description 1
- 238000000537 electroencephalography Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000002267 hypothalamic effect Effects 0.000 description 1
- 210000003016 hypothalamus Anatomy 0.000 description 1
- 230000005415 magnetization Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 239000012528 membrane Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 210000000211 third ventricle Anatomy 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 238000002834 transmittance Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
An ALVI automatic measurement method and device based on left and right ventricle segmentation, wherein the ALVI automatic measurement method comprises the following steps: acquiring medical images of the cranium of a patient on a plurality of measuring layers; respectively carrying out image segmentation on the medical images corresponding to the measurement layers according to a preset convolutional neural network to obtain segmented images based on left and right ventricle areas in each medical image; extracting a medical image corresponding to the segmentation image with the largest area, and measuring to obtain the intracranial maximum longitudinal diameter along the sickle of the brain in the medical image; determining the maximum anteroposterior diameter of the body part of the left ventricle and the right ventricle in the segmented image with the maximum area according to the intracranial maximum longitudinal diameter; and obtaining the ALVI parameter of the cranium of the patient according to the ratio of the maximum anterior-posterior diameter of the body part to the maximum intracranial longitudinal diameter. The method adopts the ALVI parameter as the lateral ventricle anteroposterior diameter index to evaluate the ventricle size of the cranium of the patient, the index is simple and clear, and compared with the conventional Evans index, the method can evaluate the condition of ventricle expansion more effectively and has practical value in medical practice application.
Description
Technical Field
The invention relates to the technical field of medical image processing, in particular to an ALVI automatic measurement method and device based on left and right ventricle segmentation.
Background
One important field of medical imaging is the diagnosis of brain conditions, such as CT images, MRI images of the brain obtained by scanning, and the determination of the presence of hydrocephalus, which is common in neurology and surgery, in the brain of a patient based on the important imaging signs of ventricular enlargement, which is one of the diagnostic criteria for hydrocephalus.
Currently, the Evans Index (EI) is the most commonly used image index for evaluating the ventricular size in clinic, and is often defined as the ratio of the maximum distance between the anterior horn of the ventricles on both sides of the transection site to the maximum cranial cavity inner diameter of the same layer. EI was first used by Willian Evans in 1942 when measuring the size of the ventricles of a child using electroencephalography, and EI ≧ 0.3 was established as a criterion for ventricular enlargement. And then, the EI measurement method and the ventricular enlargement threshold value are directly used on CT and MRI, and the EI measurement is relatively simple, convenient, economical and time-saving, so that the EI measurement method is gradually the most common measurement index for evaluating the ventricular enlargement in neurology and surgery and is used as the ventricular size image evaluation index recommended by the hydrocephalus relevant guide.
However, the current research considers that the EI is measured in different layers and has large difference, the standard measurement method also lacks clear definition, and different scanning baselines and different measurement planes influence the shape of the ventricle on the transection position, thereby influencing the linear index measurement value. Even more, there are japanese scholars who consider EI to be more effective in expressing the enlargement of the ventricles to both sides, and not to be suitable for hydrocephalus enlarged in the long axis direction. In addition, EI ≧ 0.3 as a ventricular dilation threshold is also challenging, and studies have shown that EI thresholds do not reliably distinguish between normal and dilated ventricles.
Disclosure of Invention
The invention mainly solves the technical problem of how to overcome the defects of the Evans index in the aspect of evaluating the ventricular enlargement, and provides a new image index and a new measuring method for evaluating the ventricular enlargement phenomenon.
According to a first aspect, an embodiment provides an ALVI automatic measurement method based on left and right ventricular segmentation, comprising the steps of: acquiring medical images of the cranium of a patient on a plurality of measuring layers, wherein the medical images comprise a sickle area of the cerebrum and left and right ventricle areas; respectively carrying out image segmentation on the medical images corresponding to the measurement layers according to a preset convolutional neural network to obtain segmented images based on left and right ventricle areas in each medical image; extracting a medical image corresponding to the segmentation image with the largest area, and measuring to obtain the intracranial maximum longitudinal diameter along the sickle of the brain in the medical image; determining the maximum anteroposterior diameter of the body part of the left ventricle and the right ventricle in the segmented image with the maximum area according to the intracranial maximum longitudinal diameter; and obtaining an ALVI parameter of the cranium of the patient according to the ratio of the maximum anterior-posterior diameter of the body part to the maximum intracranial longitudinal diameter, wherein the ALVI parameter is used as a lateral ventriculo-posterior diameter index and used for evaluating the size of the ventricles in the cranium of the patient.
The acquiring of medical images of a patient's cranium on a plurality of measurement slices comprises: taking each transverse position from the skull base to the skull top in the cranium of the patient as a measuring layer, and obtaining medical images of the cranium of the patient on a plurality of measuring layers by utilizing the results of CT or MRI continuous transverse position scanning.
The image segmentation is respectively carried out on the medical images corresponding to the measurement layers according to a preset convolutional neural network to obtain a segmented image based on left and right ventricle areas in each medical image, and the image segmentation method comprises the following steps: for any measurement layer, inputting the medical image corresponding to the measurement layer into a preset convolutional neural network, wherein the convolutional neural network is based on a UNet network structure and comprises a model of one or more tasks of semantic segmentation, edge point regression and frame regression; performing region detection and segmentation on the medical image corresponding to each measuring layer according to tasks contained in the convolutional neural network, and extracting left and right ventricle regions; corresponding segmented images are formed using the extracted left and right ventricle regions.
The medical image corresponding to the segmentation image with the largest area is extracted, and the intracranial maximum longitudinal diameter along the sickle of the brain in the medical image is measured and obtained, and the method comprises the following steps: comparing the area of each segmented image, and determining the segmented image with the largest area as a complete image of the left ventricle and the right ventricle in the cranium of the patient which are not covered by the thalamus; and for the medical image corresponding to the segmentation image with the largest area, measuring according to a Hough transform straight line detection principle to obtain the intracranial maximum longitudinal diameter along the brain sickle in the medical image.
The method for measuring and obtaining the intracranial maximum longitudinal diameter along the brain sickle in the medical image according to the Hough transform straight line detection principle comprises the following steps: carrying out Canny edge detection processing on the medical image corresponding to the segmented image with the largest area to obtain a corresponding binary edge image; carrying out Hough transformation on the binary edge image to map each pixel point to Hough space, and obtaining a maximum value in the Hough space; and determining the intracranial maximum longitudinal diameter along the brain sickle in the medical image corresponding to the segmentation image with the maximum area according to the obtained maximum value.
The method for determining the maximum anteroposterior diameter of the body part of the left and right ventricles in the segmented image with the maximum area according to the intracranial maximum longitudinal diameter comprises the following steps: taking a straight line section where the intracranial maximum longitudinal diameter is located as a reference line, measuring the length of a left ventricle body part and the length of a right ventricle body part which are parallel to the reference line in a segmentation image with the largest area, and respectively obtaining a first body anteroposterior diameter and a second body anteroposterior diameter; when the absolute difference value of the anteroposterior diameter of the first body and the anteroposterior diameter of the second body is smaller than a preset threshold value, taking the larger value of the two as the maximum anteroposterior diameter of the body of the left ventricle and the right ventricle; and when the absolute difference value of the front and back diameters of the first body part and the second body part is equal to or larger than a preset threshold value, taking the average value of the front and back diameters of the first body part and the second body part as the maximum front and back diameter of the body part of the left ventricle and the right ventricle.
According to a second aspect, there is provided in one embodiment an apparatus for a post-processing workstation, comprising: the communication unit is used for being connected with the medical imaging equipment and acquiring medical images of the cranium of the patient on a plurality of measuring layers from the medical imaging equipment; a processor connected to the communication unit for obtaining ALVI parameters of the cranium of the patient according to the ALVI automatic measurement method as described in the first aspect above; and the display is connected with the processor and is used for displaying the medical image corresponding to the segmented image with the largest area and displaying the ALVI parameter.
The medical imaging device is CT equipment or MRI equipment, and in the process of transverse position scanning of the CT equipment or the MRI equipment from the basis of the cranium to the top of the cranium or from the top of the cranium to the basis of the cranium of a patient, the communication unit obtains medical images of the cranium of the patient on a plurality of measuring layers by utilizing communication, wherein the medical images comprise a sickle brain region and left and right ventricle regions.
The processor is further used for preliminarily judging the size of the ventricle in the cranium of the patient according to the ALVI parameter and controlling the display to display a judgment result.
According to a third aspect, an embodiment provides a computer readable storage medium comprising a program executable by a processor to implement the ALVI automatic measurement method as described in the first aspect above.
The beneficial effect of this application is:
according to the embodiment, the ALVI automatic measurement method and the ALVI automatic measurement device based on the left and right ventricle segmentation comprise the following steps: acquiring medical images of the cranium of a patient on a plurality of measuring layers; respectively carrying out image segmentation on the medical images corresponding to the measurement layers according to a preset convolutional neural network to obtain segmented images based on left and right ventricle areas in each medical image; extracting a medical image corresponding to the segmentation image with the largest area, and measuring to obtain the intracranial maximum longitudinal diameter along the sickle of the brain in the medical image; determining the maximum anteroposterior diameter of the body part of the left ventricle and the right ventricle in the segmented image with the maximum area according to the intracranial maximum longitudinal diameter; and obtaining the ALVI parameter of the cranium of the patient according to the ratio of the maximum anterior-posterior diameter of the body part to the maximum intracranial longitudinal diameter. On the first hand, the medical images corresponding to each measuring layer are respectively subjected to image segmentation according to the preset convolutional neural network, so that the segmentation precision of a specific region can be improved, and segmented images based on left and right ventricle regions in the medical images can be accurately obtained; in the second aspect, since the medical image corresponding to the segmented image with the largest area is extracted, the left and right ventricles can be completely reflected in the image, and convenience is provided for subsequent measurement of the maximum longitudinal diameter of the intracranial brain of the sickle and the maximum anteroposterior diameter of the body of the left and right ventricles; in the third aspect, the ALVI parameter of the cranium of the patient is obtained according to the ratio of the maximum anterior-posterior diameter of the body part to the maximum intracranial longitudinal diameter, so that the ALVI parameter can be used as the lateral ventriculo-posterior diameter index to evaluate the size of the ventricles in the cranium of the patient.
Drawings
FIG. 1 is a flowchart of an ALVI automatic measurement method based on left and right ventricular segmentation according to an embodiment of the present disclosure;
FIG. 2 is a flow chart of image segmentation to obtain segmented images based on left and right ventricular regions;
FIG. 3 is a flow chart of a measurement of the intracranial maximum longitudinal diameter along a sickle of the brain in a medical image;
FIG. 4 is a flowchart of the measurement of the maximum anteroposterior diameter of the body of the left and right ventricles;
FIG. 5 is a schematic diagram of training a learning convolutional neural network;
FIG. 6 is a schematic diagram showing a comparison of medical images corresponding to two measurement planes;
FIG. 7 is a schematic diagram of the measurement of the intracranial maximum longitudinal diameter along the brain sickle in the medical image according to the Hough transform straight line detection principle;
FIG. 8 is a schematic structural diagram of an apparatus for a post-processing workstation according to a second embodiment of the present application;
fig. 9 is a schematic structural diagram of an ALVI automatic measurement apparatus according to a third embodiment of the present application.
Detailed Description
The present invention will be described in further detail with reference to the following detailed description and accompanying drawings. Wherein like elements in different embodiments are numbered with like associated elements. In the following description, numerous details are set forth in order to provide a better understanding of the present application. However, those skilled in the art will readily recognize that some of the features may be omitted or replaced with other elements, materials, methods in different instances. In some instances, certain operations related to the present application have not been shown or described in detail in order to avoid obscuring the core of the present application from excessive description, and it is not necessary for those skilled in the art to describe these operations in detail, so that they may be fully understood from the description in the specification and the general knowledge in the art.
Furthermore, the features, operations, or characteristics described in the specification may be combined in any suitable manner to form various embodiments. Also, the various steps or actions in the method descriptions may be transposed or transposed in order, as will be apparent to one of ordinary skill in the art. Thus, the various sequences in the specification and drawings are for the purpose of describing certain embodiments only and are not intended to imply a required sequence unless otherwise indicated where such sequence must be followed.
The numbering of the components as such, e.g., "first", "second", etc., is used herein only to distinguish the objects as described, and does not have any sequential or technical meaning. The term "connected" and "coupled" when used in this application, unless otherwise indicated, includes both direct and indirect connections (couplings).
The first embodiment,
Referring to fig. 1, the present embodiment discloses an ALVI automatic measurement method based on left and right ventricular segmentation, which includes steps S110-S150, which are described below.
Step S110, medical images of the cranium of the patient on a plurality of measurement levels are acquired, wherein the medical image corresponding to each measurement level comprises a sickle brain region and left and right ventricle regions.
In one implementation, each slice from the base of the skull to the top of the skull in the patient's cranium may be used as a measurement slice, such that medical images of the patient's cranium at multiple measurement slices are obtained using the results of a CT or MRI slice scan.
It should be noted that the lateral ventricle of the cranium of the patient is located at the deep part of the cerebral hemisphere, and the left ventricle and the right ventricle are respectively in a C-shaped ventricular cavity, and the ventricular canal membrane is lined inside and the cavity is filled with cerebrospinal fluid. The left ventricle and the right ventricle are separated by a median sagittal boundary plate, and each ventricle is communicated with the third ventricle through an interventricular hole and is in an indirect communication state. The cerebral sickle is a sickle-shaped fold which is formed by the dura mater which protrudes into the middle line of the cranial vertex from the middle line to the middle between two hemispheres of the brain, the front end of the structure of the cerebral sickle starts from a cockscomb, and the rear end of the structure of the cerebral sickle is attached to the cerebellum.
It should be noted that ct (computed tomography), which is a continuous cross-sectional scan around a certain part of a human body by using a precisely collimated X-ray beam and a detector with extremely high sensitivity, has the characteristics of fast scan time, clear image and the like, and can be used for cross-sectional scanning of parts of the cranium, limbs and the like of a patient, thereby obtaining a contrast image on each cross section. The CT can detect a certain part of the human body by using an instrument with extremely high sensitivity according to the difference of the absorption and transmittance of different tissues of the human body to X-rays, i.e., attenuation coefficients, and then input the data obtained by measurement into a computer, and acquire cross-section or three-dimensional images of the part of the human body to be detected by using the computer to perform quantization processing and three-dimensional reconstruction on the data, i.e., capture medical images of different measurement layers.
Magnetic Resonance Imaging (MRI) is a method of using the principle of nuclear Magnetic Resonance, and based on the different attenuation of the released energy in different structural environments inside a substance, applying a gradient Magnetic field to detect the emitted electromagnetic waves, so as to obtain the position and the type of the nuclei constituting the object, thereby drawing a structural image inside the object. Magnetic resonance imaging examination has become a common imaging examination method, which is a new imaging examination technique that generally has no effect on human health. In medical detection, medical images of any position of a certain part of a human body can be obtained by using MRI.
And step S120, respectively carrying out image segmentation on the medical images corresponding to the measurement layers according to a preset convolutional neural network, and obtaining segmented images based on the left ventricle area and the right ventricle area in each medical image by segmentation.
The medical image segmentation is a complex and key step in the field of medical image processing and analysis, and aims to segment parts with certain special meanings in a medical image, extract relevant features, provide reliable basis for clinical diagnosis and pathological research and assist doctors in making more accurate diagnosis. The development of medical image segmentation technology not only affects the development of other related technologies in medical image processing, but also plays an extremely important role in the analysis of biomedical images; in recent years, medical image segmentation techniques have made significant progress due to the application of some emerging disciplines in medical image processing. Due to the complexity of the medical image, a series of problems such as non-uniformity, individual difference and the like need to be solved in the segmentation process, so that the general image segmentation method is difficult to be directly applied to medical image segmentation. There are many methods for image segmentation of medical images, such as threshold algorithm, edge detection algorithm, deformation model algorithm, fuzzy clustering algorithm, genetic algorithm, etc., but these image segmentation algorithms have different performances in different application fields, and the current method that performs best in the medical image segmentation field is the convolutional neural network method.
A Convolutional Neural Network (CNN) is a multi-layered neural network, each layer consisting of a plurality of two-dimensional planes, and each plane consisting of a plurality of independent neurons. The neural network is used as one of neural networks and is particularly important to be applied to the field of medical image segmentation. The convolutional neural network is a multilayer perceptron specially designed for identifying two-dimensional shapes, the network structure has high invariance to the deformation in the forms of translation, scaling, inclination or other forms, the good performances are learned by the network in a supervision mode, the network structure mainly has two characteristics of sparse connection and weight sharing, image segmentation can be performed on input medical images in the modes of feature extraction, feature mapping, subsampling and the like, a brain sickle area and left and right ventricle areas in the medical images corresponding to each measurement layer are detected, and therefore segmented images based on the left and right ventricle areas are output.
Step S130, extracting the medical image corresponding to the segmentation image with the largest area, and measuring to obtain the intracranial maximum longitudinal diameter along the brain sickle in the medical image.
It should be noted that the shape and size of the left and right ventricle regions in the medical image corresponding to each measurement slice may be different, and in order to facilitate understanding of the influence of the left and right ventricle regions on the ventricle size, it is necessary to determine the divided image having the largest area by comparing the area sizes of the respective divided images, so as to obtain the largest image region of the left and right ventricles, i.e., the image region when the central slice of the left and right ventricle body is not covered by the thalamus. For the medical image corresponding to the segmentation image with the largest area, the intracranial maximum longitudinal diameter along the brain sickle can be obtained by measuring the longitudinal diameter distance of the same layer along the brain sickle. The intracranial maximum longitudinal diameter is defined as the maximum longitudinal distance from end to end in the direction of the brain sickle at the measurement level.
Step S140, determining the maximum anteroposterior diameter of the body part of the left ventricle and the right ventricle in the segmented image with the maximum area according to the intracranial maximum longitudinal diameter. Specifically, the measurement azimuth of the maximum longitudinal intracranial diameter may be used as a reference line, and the maximum anteroposterior diameter of the body of the left and right ventricles in the segmented image having the largest area may be determined based on the reference line.
Here, the maximum anteroposterior diameter of the body of the right and left ventricles means the maximum anteroposterior distance length of the right and left ventricles parallel to the maximum longitudinal diameter of the cranium in the measurement plane.
And S150, obtaining the ALVI parameters of the cranium of the patient according to the ratio of the maximum anterior-posterior diameter of the body to the maximum intracranial longitudinal diameter. The ALVI parameter is used herein as a lateral ventriculo-antero-posterior diameter index and to evaluate the ventricular size in the cranium of the patient.
It should be noted that the lateral ventricle anterior-posterior diameter index (ALVI) is substantially the result of the ratio of the maximum anterior-posterior diameter of the body to the maximum intracranial longitudinal diameter, and the physical expression is the ratio of the maximum anterior-posterior diameter of the body of the left and right ventricles to the maximum intracranial longitudinal diameter of the same layer along the brain sickle when the central layer of the left and right ventricles is completely present for the first time in the direction from the skull base to the skull vertex at the transverse position of the cranial scan with the brain commissure line as the baseline and is not covered by the thalamus.
In this embodiment, the step S120 mainly involves a process of performing image segmentation on the medical image corresponding to each measurement slice according to a preset convolutional neural network, and then, referring to fig. 2, the step S120 may specifically include steps S121 to S123, which are respectively described as follows.
And step S121, inputting the medical image corresponding to the measurement layer into a preset convolution neural network for any measurement layer. The convolutional neural network is based on a UNet network structure and comprises a model of one or more tasks of semantic segmentation, edge point regression and frame regression.
Referring to the structure of the convolutional neural network illustrated in fig. 5, which includes a plurality of convolutional/deconvolution layers and a plurality of pooling layers, a front feature extraction section and a rear upsampling section are formed, and since the network structure is like a U-shape, it is called UNet network. In the feature extraction part, a scale is generated every time a pooling layer is passed; and in the up-sampling part, the channels corresponding to the feature extraction part are fused in the same scale as the number of the channels corresponding to the feature extraction part once per up-sampling. Therefore, the UNet network has the application characteristics of multiple scales and suitability for super-large image segmentation, and is suitable for input processing of medical images in the implementation.
It should be noted that a UNet network-based model can be used to construct a plurality of tasks, such as a semantic segmentation model, an edge point regression model, and a frame regression model, which are all related to the prior art, and therefore, a detailed description thereof is omitted here.
And S122, performing area detection and segmentation on the medical image corresponding to each measuring layer according to the UNet network structure and the task model of the convolutional neural network, and extracting left and right ventricle areas.
In this embodiment, for a medical image input to the UNet network, a full convolution neural network may be used to construct a semantic segmentation model and simply segment the medical image, an edge point regression model is used to process a ragged linear edge after segmentation, and a Bounding box regression (Bounding box regression) model is used to perform multiple transformations of translation and scale scaling on the image after edge processing.
Referring to fig. 5, in order to improve the image segmentation sensitivity based on UNet network, it is necessary to train and learn the network by some training samples. Inputting some marked medical images into a convolutional neural network based on a UNet network, obtaining a region of interest (ROI) of a specific image feature according to a candidate region after convolution processing and region extraction, accessing the ROI into a full-link layer for continuous processing, inputting processed feature information into a classifier for feature classification, and performing key point regression processing on the processed feature information to obtain key points therein; in addition, the region of interest can be subjected to convolution/deconvolution processing to obtain a mask layer with specific image characteristics. Then, the network itself can be effectively trained by means of segmentation combined with auxiliary classification, key point regression and the like, so that the main characteristic information of the medical image can be obtained by learning, and then the UNet-based network can be used for effectively detecting and segmenting the region of the input medical image after the training and learning are completed, so that the region belonging to the sickle cerebrum and the region belonging to the left and right ventricles in the medical image can be extracted.
In step S123, a corresponding divided image is formed using the extracted left and right ventricle areas. It is understood that after the left and right ventricular areas in the medical image are extracted, the areas can be formed into corresponding segmented images, and only images of the left and right ventricular areas are often found in the segmented images.
In the present embodiment, the step S130 above mainly involves measuring the intracranial maximum longitudinal diameter along the sickle of the brain in the medical image from the medical image corresponding to the segmented image with the largest area, and then, referring to fig. 3, the step S130 may specifically include steps S131 to S132, which are respectively described as follows.
Step S131, comparing the area of each segmented image, and determining the segmented image with the largest area as a complete image of the left and right ventricles of the brain of the patient which are not covered by the thalamus.
After the segmented images corresponding to the medical images are obtained, the area of the region can be easily calculated from the left and right ventricle regions displayed in the segmented images. Such as the two medical images indicated by A, B in fig. 6, a1 and a2 in the medical image a represent the sickle brain region and the left and right ventricle regions, respectively, and B1 and B2 in the medical image B represent the sickle brain region and the left and right ventricle regions, respectively; left and right ventricular regions a2 in medical image a are visible when left and right ventricular regions are not covered by the hypothalamus, and the corresponding segmented image has the largest area at this time; when left and right ventricle regions B2 in medical image B are visible when the left and right ventricle regions are masked by the hypothalamic portion (see mask position B3), the corresponding segmented image has a smaller area. Then, the segmented image with the largest area is determined by comparing the area sizes of the respective segmented images (see reference character b2 in fig. 6), and is taken as a complete image in which the left and right ventricles in the cranium of the patient are not covered by the thalamus.
And S132, measuring and obtaining the intracranial maximum longitudinal diameter along the brain sickle in the medical image according to the Hough transform straight line detection principle for the medical image corresponding to the segmentation image with the maximum area.
It should be noted that the hough transform straight line detection principle is one of the most commonly used straight line extraction methods in digital image processing, and the basic idea thereof is to transform or project points on a straight line to a straight line on a parameter plane, and convert the straight line extraction problem into the counting problem by using the relationship that the straight lines corresponding to collinear points intersect at one point in a parameter space. For detecting straight lines, the hough transform is actually a polar transform. The straight line can be represented by a simple linear function in the rectangular coordinate system X-Y
y＝ax+b；
Wherein a is the slope, bA straight line can be uniquely determined as long as a and b are determined for the intercept with the Y-axis. If using p0Represents the algebraic distance, θ, from the origin to the line0The angle between the orthogonal line representing the straight line and the x-axis is
At this time, the linear function can be converted into a polar form
ρ＝xcosθ+ysinθ。
Where (ρ, θ) is a representation in polar coordinates. If (ρ, θ) is expressed in rectangular coordinates, i.e., ρ and θ are orthogonally processed, then (ρ, θ) can be referred to as hough space. At one point in the rectangular coordinate system, a straight line corresponds to a sine curve of the Hough space, the straight line is composed of countless points, and the hough space is an infinite number of sine curves, and the sine curves intersect at one point (rho) 0,θ0) By substituting the point into a linear function, the slope and intercept of the line can be calculated, and a line can be determined. Therefore, when identifying a straight line by hough transform, a maximum value in hough space may correspond to one straight line.
In one embodiment, referring to fig. 7, the medical image corresponding to the segmented image with the largest area may be processed as follows: (1) performing Canny edge detection processing on a medical image (such as an image shown by a reference numeral C) corresponding to the segmented image with the largest area to obtain a corresponding binary edge image (such as an image shown by a reference numeral D); (2) carrying out Hough transformation on the binary edge image to map each pixel point to Hough space, and obtaining a maximum value in the Hough space; (3) and determining the intracranial maximum longitudinal diameter (such as a straight line segment shown by the reference numeral e 1) along the brain sickle in the medical image corresponding to the segmentation image with the maximum area according to the acquired maximum value.
It should be noted that Canny edge detection is a conventional image processing technique, and often includes the following detection steps: a) using a Gaussian filter to smooth the image and filter out noise; b) calculating the gradient strength and direction of each pixel point in the image; c) applying non-maximum suppression to eliminate spurious responses caused by edge detection; d) applying dual threshold detection to determine true and potential edges; e) edge detection is finally accomplished by suppressing isolated weak edges.
In the present embodiment, the above step S140 mainly involves determining the maximum anteroposterior diameter of the body of the left and right ventricles in the segmented image with the largest area according to the intracranial maximum longitudinal diameter, and then referring to fig. 4, the step S140 may specifically include steps S141 to S144, which are respectively described as follows.
Step S141, taking a straight line segment where the intracranial maximum longitudinal diameter is located as a reference line, measuring the left ventricle body length and the right ventricle body length parallel to the reference line in the segmentation image with the largest area, and respectively obtaining a first body anteroposterior diameter and a second body anteroposterior diameter.
For example, in the case of the medical image a in fig. 6, in the case of the intracranial maximum longitudinal diameter (straight line segment shown by reference numeral a 3), the left ventricular body length and the right ventricular body length can be measured by taking the intracranial maximum longitudinal diameter a3 as a reference line and measuring the anteroposterior diameter of the left and right ventricular regions a2 in parallel to the reference line, thereby obtaining the first body anteroposterior diameter a4 and the second body anteroposterior diameter a 5.
In step S142, it is determined whether the absolute difference between the front-rear diameter of the first body and the front-rear diameter of the second body is smaller than a preset threshold, if so, the process proceeds to step S143, otherwise, the process proceeds to step S144.
Note that, the preset threshold may be set to 1cm or another value, and then the process proceeds to step S143 when the absolute value obtained by subtracting the first body front-rear diameter a4 and the second body front-rear diameter a5 is smaller than 1 cm.
In step S143, when the absolute difference between the anteroposterior diameter of the first body and the anteroposterior diameter of the second body is smaller than the preset threshold, the larger value of the two is used as the maximum anteroposterior diameter of the body of the left and right ventricles.
In step S144, when the absolute difference between the anterior-posterior diameter of the first body and the anterior-posterior diameter of the second body is equal to or greater than a preset threshold, the average value of the two is used as the maximum anterior-posterior diameter of the body of the left and right ventricles.
It will be appreciated by those skilled in the art that the following technical advantages may be achieved when applying the ALVI automated measurement method disclosed in the present embodiment: (1) when the convolution neural network is used for respectively carrying out image segmentation on the medical images corresponding to the measurement layers, the segmentation precision of a specific region can be improved by combining with auxiliary tasks of classification and detection, so that segmented images based on left and right ventricle regions in the medical images can be accurately obtained; (2) after the medical image corresponding to the segmentation image with the largest area is extracted, the left ventricle and the right ventricle can be completely reflected in the image, and convenience is provided for subsequent measurement of the maximum longitudinal diameter of the intracranial brain along the sickle of the brain and the maximum front-back diameter of the body of the left ventricle and the right ventricle; (3) obtaining ALVI parameters of the cranium of the patient according to the ratio of the maximum anterior-posterior diameter of the body part to the maximum intracranial longitudinal diameter, so that the ALVI parameters can be used as lateral ventriculo-posterior diameter indexes to evaluate the size of the ventricles in the cranium of the patient; (4) the lateral ventricle anteroposterior diameter index adopted by the method is simple and clear, can more effectively evaluate the phenomenon of ventricle expansion compared with the conventional Evans Index (EI), and has practical value in medical practice application.
Example II,
Referring to fig. 8, based on the left and right ventricle segmentation based ALVI automatic measurement method disclosed in the first embodiment, the present embodiment discloses an apparatus for a post-processing workstation, the apparatus 2 is used for performing ALVI automatic measurement, and may include a communication unit 21, a processor 22 and a display 23.
It should be noted that the post-processing workstation is often used as an auxiliary and support for the image diagnosis or scientific research process, and provides an auxiliary tool for the diagnosis of the disease condition for the imaging physician. The post-processing workstation has the main functions of editing images, adjusting window positions, adjusting image gray scale, adjusting contrast, drawing and calculating pseudo colors, displaying parameters, measuring areas, marking characters, amplifying local parts and the like, and an operator can freely process required images according to requirements.
In the present embodiment, the communication unit 21 may be some conventional communication components for performing communication connection with the medical imaging device Z1 and acquiring medical images of the cranium of the patient on a plurality of measurement levels from the medical imaging device Z1. The medical imaging device Z1 may be a CT device or an MRI device, and the communication unit 21 obtains medical images of the patient's cranium on a plurality of measurement levels by using communication during the continuous transverse position scanning process from the basis of the cranium to the top of the cranium or from the top of the cranium to the bottom of the cranium of the patient, where the medical images include the area of the cerebrum, the left ventricle area, the right ventricle area, and the like. Specifically, the CT apparatus and the MRI apparatus can transmit the scanned medical image to a post-processing workstation through a PACS system (medical image archiving and transmitting system) and be received by the communication unit 21 within the post-processing workstation.
When CT equipment is adopted to scan the cranium of a patient, a conventional cranium spiral CT scanning mode can be adopted, different models used by different hospitals and cranium spiral CT scanning parameters possibly have differences, but the different CT equipment and the scanning parameters have no difference on image acquisition. For example, using GE Discovery CT750 HD spiral CT scan, scan direction from the top of the skull to the base of the skull, using the following protocol parameters: 1.25mm layer thickness, 1.25mm layer spacing, 120kv, 360 mA.
When the MRI equipment is used for scanning the cranium of a patient, a vector position 3D-T1WI scanning mode can be adopted, and different MRI equipment and scanning parameters have no difference on image acquisition. Such as with a 3.0T MRI scanner (prism, Siemens, Erlangen, Germany). The vector 3D-T1WI scanning mode adopts a rapid acquisition gradient echo sequence prepared by magnetization to cover the whole head, the scanning direction is from the top of the skull to the skull base, and the following protocol parameters are specifically used: TR/TE 2300/2.98 ms; the turning angle is equal to 9 degrees; the layer thickness is 1 mm; FOV 256 × 256 mm; a matrix 256 × 256; pixels 1 × 1 mm.
In the present embodiment, the processor 22 is connected to the communication unit 21 for obtaining the ALVI parameters of the cranium of the patient according to the ALVI automatic measurement method disclosed in the first embodiment. In this embodiment, the processor 22 acquires medical images of the patient's cranium at a plurality of measurement levels, the acquired medical images including the region of the sickle brain and the regions of the left and right ventricles; the processor 22 performs image segmentation on the medical images corresponding to the measurement layers according to a preset convolutional neural network to obtain segmented images based on the left ventricle area and the right ventricle area in each medical image; the processor 22 extracts the medical image corresponding to the segmented image with the largest area, and measures and obtains the intracranial maximum longitudinal diameter along the brain sickle in the medical image; the processor 22 determines the maximum anteroposterior diameter of the body part of the left ventricle and the right ventricle in the segmented image with the maximum area according to the intracranial maximum longitudinal diameter; the processor 22 obtains the ALVI parameter of the cranium of the patient according to the ratio of the maximum anteroposterior diameter of the body to the maximum longitudinal diameter in the cranium, and the calculated ALVI parameter is used as the lateral ventriculo-anteroposterior diameter index and evaluates the size of the ventricles in the cranium of the patient. For specific functions of the processor 22, reference may be made to steps S110 to S150 in the first embodiment, which is not described herein again.
In the present embodiment, referring to fig. 8, the display 23 is connected to the processor 22, and the display 23 is used for displaying the medical image corresponding to the segmented image with the largest area and displaying the ALVI parameters.
Further, the processor 22 may also preliminarily determine the size of the ventricle in the cranium of the patient according to the ALVI parameter, and control the display to display the determination result. For example, since the lateral ventriculo-antero-posterior diameter index (ALVI parameter) is substantially the result of the ratio of the maximum antero-posterior diameter of the body to the maximum intracranial longitudinal diameter, the processor 22 may preliminarily determine whether the ALVI parameter is greater than 0.5, and when the ALVI parameter is greater than 0.5, it is determined that the cranium of the patient has the symptom of ventricular enlargement, and this determination result may be displayed through the display 23 for the medical care personnel to use as a reference when diagnosing the condition of the patient.
Example III,
Referring to fig. 9, based on the automatic ALVI measurement method based on left and right ventricular segmentation disclosed in the first embodiment, the present embodiment discloses an automatic ALVI measurement apparatus, and the apparatus 3 includes a memory 31 and a processor 32, which are respectively described below.
In the present embodiment, the memory 31 is used for storing programs, and the processor 32 is used for implementing the ALVI automatic measurement method disclosed in the first embodiment by executing the programs stored in the memory 31.
It is noted that the memory 31 can be considered a computer-readable storage medium having a program executable by the processor 32 to implement the disclosed automatic ALVI measurement method.
In the present embodiment, the processor 32 acquires medical images of the patient's cranium at a plurality of measurement levels, the acquired medical images including the region of the sickle brain and the regions of the left and right ventricles; the processor 32 performs image segmentation on the medical images corresponding to the measurement layers according to a preset convolutional neural network to obtain segmented images based on the left ventricle area and the right ventricle area in each medical image; the processor 32 extracts the medical image corresponding to the segmented image with the largest area, and measures and obtains the intracranial maximum longitudinal diameter along the brain sickle in the medical image; the processor 32 determines the maximum anteroposterior diameter of the body part of the left ventricle and the right ventricle in the segmented image with the maximum area according to the intracranial maximum longitudinal diameter; the processor 32 obtains an ALVI parameter of the cranium of the patient according to the ratio of the maximum anteroposterior diameter of the body to the maximum longitudinal diameter of the cranium, and the calculated ALVI parameter is used as a lateral ventriculo-anteroposterior diameter index and for evaluating the size of the ventricles in the cranium of the patient. For specific functions of the processor 32, reference may be made to steps S110 to S150 in the first embodiment, which is not described herein again.
Those skilled in the art will appreciate that all or part of the functions of the various methods in the above embodiments may be implemented by hardware, or may be implemented by computer programs. When all or part of the functions of the above embodiments are implemented by a computer program, the program may be stored in a computer-readable storage medium, and the storage medium may include: a read only memory, a random access memory, a magnetic disk, an optical disk, a hard disk, etc., and the program is executed by a computer to realize the above functions. For example, the program may be stored in a memory of the device, and when the program in the memory is executed by the processor, all or part of the functions described above may be implemented. In addition, when all or part of the functions in the above embodiments are implemented by a computer program, the program may be stored in a storage medium such as a server, another computer, a magnetic disk, an optical disk, a flash disk, or a removable hard disk, and may be downloaded or copied to a memory of a local device, or may be version-updated in a system of the local device, and when the program in the memory is executed by a processor, all or part of the functions in the above embodiments may be implemented.
The present invention has been described in terms of specific examples, which are provided to aid understanding of the invention and are not intended to be limiting. For a person skilled in the art to which the invention pertains, several simple deductions, modifications or substitutions may be made according to the idea of the invention.
Claims (10)
1. An ALVI automatic measurement method based on left and right ventricle segmentation is characterized by comprising the following steps:
acquiring medical images of the cranium of a patient on a plurality of measuring layers, wherein the medical images comprise a sickle area of the cerebrum and left and right ventricle areas;
respectively carrying out image segmentation on the medical images corresponding to the measurement layers according to a preset convolutional neural network to obtain segmented images based on left and right ventricle areas in each medical image;
extracting a medical image corresponding to the segmentation image with the largest area, and measuring to obtain the intracranial maximum longitudinal diameter along the sickle of the brain in the medical image;
determining the maximum anteroposterior diameter of the body part of the left ventricle and the right ventricle in the segmented image with the maximum area according to the intracranial maximum longitudinal diameter;
and obtaining an ALVI parameter of the cranium of the patient according to the ratio of the maximum anterior-posterior diameter of the body part to the maximum intracranial longitudinal diameter, wherein the ALVI parameter is used as a lateral ventriculo-posterior diameter index and used for evaluating the ventriculo-ventricular size of the cranium of the patient.
2. The ALVI automated measurement method of claim 1, wherein the acquiring medical images of the patient's cranium at a plurality of measurement levels comprises:
taking each transversal position from the skull base to the skull top in the cranium of the patient as a measuring layer, and obtaining medical images of the cranium of the patient on a plurality of measuring layers by utilizing the result of CT or MRI transversal position scanning.
3. The ALVI automatic measurement method of claim 1, wherein the image segmentation is performed on the medical images corresponding to the measurement slices according to a preset convolutional neural network, so as to obtain a segmented image based on left and right ventricle areas in each medical image, and the method comprises the following steps:
for any measurement layer, inputting the medical image corresponding to the measurement layer into a preset convolutional neural network, wherein the convolutional neural network is based on a UNet network structure and comprises a model of one or more tasks of semantic segmentation, edge point regression and frame regression;
performing region detection and segmentation on the medical image corresponding to each measuring layer according to tasks contained in the convolutional neural network, and extracting left and right ventricle regions;
corresponding segmented images are formed using the extracted left and right ventricle regions.
4. The ALVI automatic measurement method of claim 3, wherein the step of extracting the medical image corresponding to the segmented image with the largest area and measuring the intracranial maximum longitudinal diameter along the sickle brain in the medical image comprises:
comparing the area of each segmented image, and determining the segmented image with the largest area as a complete image of the left ventricle and the right ventricle in the cranium of the patient which are not covered by the thalamus;
and for the medical image corresponding to the segmentation image with the largest area, measuring according to a Hough transform straight line detection principle to obtain the intracranial maximum longitudinal diameter along the brain sickle in the medical image.
5. The ALVI automatic measurement method of claim 4, wherein the measurement of the intracranial maximum longitudinal diameter along the sickle of the brain in the medical image according to the Hough transform straight line detection principle comprises:
carrying out Canny edge detection processing on the medical image corresponding to the segmented image with the largest area to obtain a corresponding binary edge image;
carrying out Hough transformation on the binary edge image to map each pixel point to Hough space, and obtaining a maximum value in the Hough space;
and determining the intracranial maximum longitudinal diameter along the brain sickle in the medical image corresponding to the segmentation image with the maximum area according to the obtained maximum value.
6. The ALVI automated measurement method of claim 4, wherein determining the body maximum anteroposterior diameter of the left and right ventricles in the maximally-areal segmented image from the intracranial maximum longitudinal diameter comprises:
taking a straight line section where the intracranial maximum longitudinal diameter is located as a reference line, measuring the length of a left ventricle body part and the length of a right ventricle body part which are parallel to the reference line in a segmentation image with the largest area, and respectively obtaining a first body anteroposterior diameter and a second body anteroposterior diameter;
when the absolute difference value of the anteroposterior diameter of the first body and the anteroposterior diameter of the second body is smaller than a preset threshold value, taking the larger value of the two as the maximum anteroposterior diameter of the body of the left ventricle and the right ventricle;
and when the absolute difference value of the front and back diameters of the first body part and the second body part is equal to or larger than a preset threshold value, taking the average value of the front and back diameters of the first body part and the second body part as the maximum front and back diameter of the body part of the left ventricle and the right ventricle.
7. An apparatus for an aftertreatment workstation, comprising:
the communication unit is used for being connected with the medical imaging equipment and acquiring medical images of the cranium of the patient on a plurality of measuring layers from the medical imaging equipment;
a processor connected to the communication unit for obtaining ALVI parameters of the cranium of the patient according to the ALVI automated measurement method of any one of claims 1-6;
And the display is connected with the processor and is used for displaying the medical image corresponding to the segmented image with the largest area and displaying the ALVI parameter.
8. The apparatus according to claim 7, wherein the medical imaging device is a CT device or an MRI device, and the communication unit obtains medical images of the patient's cranium on a plurality of measurement levels by using communication during the scanning of the CT device or the MRI device from the basis of cranium to the top of cranium or from the top of cranium to the bottom of cranium of the patient, wherein the medical images comprise a sickle area and left and right ventricle areas.
9. The apparatus of claim 7, wherein the processor is further configured to preliminarily determine a size of a ventricle of a cranium of the patient based on the ALVI parameter, and to control the display to display the determination.
10. A computer-readable storage medium comprising a program executable by a processor to implement the ALVI automatic measurement method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010651711.0A CN111862014A (en) | 2020-07-08 | 2020-07-08 | ALVI automatic measurement method and device based on left and right ventricle segmentation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010651711.0A CN111862014A (en) | 2020-07-08 | 2020-07-08 | ALVI automatic measurement method and device based on left and right ventricle segmentation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111862014A true CN111862014A (en) | 2020-10-30 |
Family
ID=73153142
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010651711.0A Pending CN111862014A (en) | 2020-07-08 | 2020-07-08 | ALVI automatic measurement method and device based on left and right ventricle segmentation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111862014A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469941A (en) * | 2021-05-27 | 2021-10-01 | 武汉楚精灵医疗科技有限公司 | Method for measuring width of bile-pancreatic duct in ultrasonic bile-pancreatic duct examination |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010117573A2 (en) * | 2009-04-07 | 2010-10-14 | Virginia Commonwealth University | Automated measurement of brain injury indices using brain ct images, injury data, and machine learning |
| CN102016911A (en) * | 2008-03-03 | 2011-04-13 | 新加坡科技研究局 | A method and system of segmenting CT scan data |
| CN104414680A (en) * | 2013-08-21 | 2015-03-18 | 深圳迈瑞生物医疗电子股份有限公司 | Three-dimensional ultrasonic imaging method and system |
| CN105825508A (en) * | 2016-03-17 | 2016-08-03 | 电子科技大学 | Hydrocephalus auxiliary diagnosis method based on brain medical image segmentation |
| CN107203997A (en) * | 2016-03-16 | 2017-09-26 | 上海联影医疗科技有限公司 | A kind of dividing method of half brain of left and right |
| US20180144467A1 (en) * | 2016-11-22 | 2018-05-24 | Hyperfine Research, Inc. | Systems and methods for automated detection in magnetic resonance images |
| CN108369642A (en) * | 2015-12-18 | 2018-08-03 | 加利福尼亚大学董事会 | Acute disease feature is explained and quantified according to head computer tomography |
| CN108765447A (en) * | 2018-04-26 | 2018-11-06 | 深圳博脑医疗科技有限公司 | A kind of image partition method, image segmentation device and electronic equipment |
| CN109686426A (en) * | 2018-12-29 | 2019-04-26 | 上海商汤智能科技有限公司 | Medical imaging processing method and processing device, electronic equipment and storage medium |
| CN110279417A (en) * | 2019-06-25 | 2019-09-27 | 沈阳东软智能医疗科技研究院有限公司 | Identify the method, device and equipment of aorta vessel |
| CN110338844A (en) * | 2015-02-16 | 2019-10-18 | 深圳迈瑞生物医疗电子股份有限公司 | The display processing method and 3-D supersonic imaging method and system of three-dimensional imaging data |
| CN110517766A (en) * | 2019-08-09 | 2019-11-29 | 上海依智医疗技术有限公司 | Identify the method and device of encephalatrophy |
| CN110956636A (en) * | 2019-11-28 | 2020-04-03 | 北京推想科技有限公司 | Image processing method and device |
| CN111161275A (en) * | 2018-11-08 | 2020-05-15 | 腾讯科技（深圳）有限公司 | Method and device for segmenting target object in medical image and electronic equipment |
-
2020
- 2020-07-08 CN CN202010651711.0A patent/CN111862014A/en active Pending
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102016911A (en) * | 2008-03-03 | 2011-04-13 | 新加坡科技研究局 | A method and system of segmenting CT scan data |
| WO2010117573A2 (en) * | 2009-04-07 | 2010-10-14 | Virginia Commonwealth University | Automated measurement of brain injury indices using brain ct images, injury data, and machine learning |
| CN104414680A (en) * | 2013-08-21 | 2015-03-18 | 深圳迈瑞生物医疗电子股份有限公司 | Three-dimensional ultrasonic imaging method and system |
| CN110338844A (en) * | 2015-02-16 | 2019-10-18 | 深圳迈瑞生物医疗电子股份有限公司 | The display processing method and 3-D supersonic imaging method and system of three-dimensional imaging data |
| CN108369642A (en) * | 2015-12-18 | 2018-08-03 | 加利福尼亚大学董事会 | Acute disease feature is explained and quantified according to head computer tomography |
| CN107203997A (en) * | 2016-03-16 | 2017-09-26 | 上海联影医疗科技有限公司 | A kind of dividing method of half brain of left and right |
| CN105825508A (en) * | 2016-03-17 | 2016-08-03 | 电子科技大学 | Hydrocephalus auxiliary diagnosis method based on brain medical image segmentation |
| CN109983474A (en) * | 2016-11-22 | 2019-07-05 | 海珀菲纳研究股份有限公司 | For the system and method detected automatically in magnetic resonance image |
| US20180144467A1 (en) * | 2016-11-22 | 2018-05-24 | Hyperfine Research, Inc. | Systems and methods for automated detection in magnetic resonance images |
| CN108765447A (en) * | 2018-04-26 | 2018-11-06 | 深圳博脑医疗科技有限公司 | A kind of image partition method, image segmentation device and electronic equipment |
| CN111161275A (en) * | 2018-11-08 | 2020-05-15 | 腾讯科技（深圳）有限公司 | Method and device for segmenting target object in medical image and electronic equipment |
| CN109686426A (en) * | 2018-12-29 | 2019-04-26 | 上海商汤智能科技有限公司 | Medical imaging processing method and processing device, electronic equipment and storage medium |
| CN110279417A (en) * | 2019-06-25 | 2019-09-27 | 沈阳东软智能医疗科技研究院有限公司 | Identify the method, device and equipment of aorta vessel |
| CN110517766A (en) * | 2019-08-09 | 2019-11-29 | 上海依智医疗技术有限公司 | Identify the method and device of encephalatrophy |
| CN110956636A (en) * | 2019-11-28 | 2020-04-03 | 北京推想科技有限公司 | Image processing method and device |
Non-Patent Citations (6)
| Title |
|---|
| MOHSEN GHAFOORIAN 等: "Student Beats the Teacher: Deep Neural Networks for Lateral Ventricles Segmentation in Brain MR", 《ARXIV》, pages 1 - 6 * |
| MUHAN SHAO 等: "Brain ventricle parcellation using a deep neural network: Application to patients with ventriculomegaly", 《NEUROIMAGE: CLINICAL》, pages 1 - 10 * |
| QINGNENG LI 等: "Glioma Segmentation With a Unified Algorithm in Multimodal MRI Images", 《IEEE ACCESS》, vol. 6, pages 9843 - 9553 * |
| WENJIE HE 等: "A new index for assessing cerebral ventricular volume in idiopathic normal-pressure hydrocephalus: a comparison with Evans’ index", 《NEURORADIOLOGY》, vol. 62, no. 6, pages 661 - 667, XP037145371, DOI: 10.1007/s00234-020-02361-8 * |
| 刘亚龙 等: "基于精细化残差U-Net的新生儿局灶性脑白质损伤分割模型", 《计算机应用》, vol. 39, no. 12, pages 3456 - 3461 * |
| 张静 等: "脑部MRI 图像自动分割算法的研究", 《信息技术》, no. 6, pages 76 - 80 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469941A (en) * | 2021-05-27 | 2021-10-01 | 武汉楚精灵医疗科技有限公司 | Method for measuring width of bile-pancreatic duct in ultrasonic bile-pancreatic duct examination |
| CN113469941B (en) * | 2021-05-27 | 2022-11-08 | 武汉楚精灵医疗科技有限公司 | Method for measuring width of bile-pancreatic duct in ultrasonic bile-pancreatic duct examination |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11967072B2 (en) | Three-dimensional object segmentation of medical images localized with object detection | |
| US10453191B2 (en) | Automated intravascular plaque classification | |
| CN110338844B (en) | Three-dimensional imaging data display processing method and three-dimensional ultrasonic imaging method and system | |
| Namburete et al. | Learning-based prediction of gestational age from ultrasound images of the fetal brain | |
| JP4469594B2 (en) | System for measuring disease-related tissue changes | |
| JP4310099B2 (en) | Method and system for lung disease detection | |
| JP5081390B2 (en) | Method and system for monitoring tumor burden | |
| Koh et al. | An automatic segmentation method of the spinal canal from clinical MR images based on an attention model and an active contour model | |
| DE102012108121A1 (en) | Method and system for ultrasound-assisted automatic detection, quantification and tracking of pathologies | |
| JP2016531709A (en) | Image analysis technology for diagnosing disease | |
| US20080304616A1 (en) | Segmenting colon wall via level set techniques | |
| Balasooriya et al. | Intelligent brain hemorrhage diagnosis using artificial neural networks | |
| CN112884759A (en) | Method and related device for detecting metastasis state of axillary lymph nodes of breast cancer | |
| Zeng et al. | A 2.5 D deep learning-based method for drowning diagnosis using post-mortem computed tomography | |
| CN110738633A (en) | organism tissue three-dimensional image processing method and related equipment | |
| US7961923B2 (en) | Method for detection and visional enhancement of blood vessels and pulmonary emboli | |
| CN111862014A (en) | ALVI automatic measurement method and device based on left and right ventricle segmentation | |
| EP3624058A1 (en) | Method and system of analyzing symmetry from image data | |
| US20220249060A1 (en) | Method for processing 3d image data and 3d ultrasonic imaging method and system | |
| Rahmany et al. | A priori knowledge integration for the detection of cerebral aneurysm | |
| CN117036302B (en) | Method and system for determining calcification degree of aortic valve | |
| TWI548401B (en) | Method for reconstruction of blood vessels 3d structure | |
| CN115147378B (en) | CT image analysis and extraction method | |
| KR102647251B1 (en) | Method for evaluating low limb alignment and device for evaluating low limb alignment using the same | |
| Fabijańska et al. | IMAGE PROCESSING AND ANALYSIS ALGORITHMS FOR ASSESSMENT AND DIAGNOSIS OF BRAIN DISEASES |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20201030 |