WO2024113521A1 - 一种多模态数据融合的盆底功能整体评估方法及设备 - Google Patents

一种多模态数据融合的盆底功能整体评估方法及设备 Download PDF

Info

Publication number
WO2024113521A1
WO2024113521A1 PCT/CN2023/079183 CN2023079183W WO2024113521A1 WO 2024113521 A1 WO2024113521 A1 WO 2024113521A1 CN 2023079183 W CN2023079183 W CN 2023079183W WO 2024113521 A1 WO2024113521 A1 WO 2024113521A1
Authority
WO
WIPO (PCT)
Prior art keywords
module
pelvic floor
pelvic
fusion
vaginal
Prior art date
Application number
PCT/CN2023/079183
Other languages
English (en)
French (fr)
Inventor
牛晓宇
张�林
王涛
梅玲
魏冬梅
陈悦悦
崔陶
王倩
Original Assignee
四川大学华西第二医院
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 四川大学华西第二医院 filed Critical 四川大学华西第二医院
Publication of WO2024113521A1 publication Critical patent/WO2024113521A1/zh

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/12—Diagnosis using ultrasonic, sonic or infrasonic waves in body cavities or body tracts, e.g. by using catheters
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/5215—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data
    • A61B8/5238—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data for combining image data of patient, e.g. merging several images from different acquisition modes into one image
    • A61B8/5261—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data for combining image data of patient, e.g. merging several images from different acquisition modes into one image combining images from different diagnostic modalities, e.g. ultrasound and X-ray
    • 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
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • 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
    • G06T2207/10004—Still image; Photographic image
    • G06T2207/10012—Stereo images

Definitions

  • the present invention relates to a method and device for overall evaluation of pelvic floor function based on multimodal data fusion, and specifically relates to a method for overall evaluation of human pelvic floor function based on acoustic, optical and electrical multimodality, and a computer device and a computer-readable storage medium for implementing the method, and belongs to the field of medical diagnosis and detection technology.
  • PFD pelvic floor dysfunction
  • POP pelvic organ prolapse
  • UI urinary incontinence
  • fecal incontinence sexual dysfunction and chronic pelvic pain. It has a long course and a high recurrence rate. It is a type of disease that seriously affects the quality of life and physical and mental health of patients, and imposes a serious burden on individuals, families and society.
  • PFD has been considered one of the five major chronic diseases that affect the quality of human life.
  • the pelvic floor structure is complex, involving multiple tissues such as muscles, fascia, ligaments, blood vessels and nerves, forming a whole body that is related to different organs such as urinary, reproductive, and anorectal.
  • the muscles and fascia of the pelvic floor form an overall supporting structure, among which the muscle state is particularly important for the realization of the pelvic floor support function.
  • the pelvic floor muscles are composed of three levels of muscles, including the pelvic diaphragm (levator ani muscles and upper and lower fascia), deep transverse muscles and fascia, and superficial perineal muscles and fascia. Only when the pelvic floor muscles contract normally and synergistically can the coordinated control function of the pelvic floor muscles be normally exerted.
  • pelvic floor muscles degenerate, relax, and lack muscle strength, it will cause anatomical displacement of pelvic organs, abnormal closing function, and other conditions, leading to functional disorders. Therefore, how to measure and evaluate the functional state of the pelvic floor is currently an important evaluation indicator in the diagnosis and treatment of pelvic floor diseases and one of the prerequisites for selecting treatment plans.
  • the vaginal surface electromyography probes on the market can roughly measure the average electromyography, but due to the influence of multiple factors such as the small number of detection pieces or rings on the probe rod (usually two), low measurement accuracy, and displacement in the vagina after activity, the measurement is not accurate and has poor repeatability, which may cause clinicians to make misjudgments and lead to deviations in clinical treatment plans.
  • pelvic floor muscle activity is mainly qualitative, focusing on the shape of the signal line, the number of signal amplitudes from rest to contraction, the duration and quality of surface electromyographic signals during endurance contraction, and comparing the changes in pelvic floor muscle function before and after treatment. Only a few studies have conducted quantitative studies on such changes. Keshwani et al. compared 16 commercial vaginal probes and found (Neurourol Urodyn. 2015 Feb;34(2):104-12) that all probes had design defects, including geometry, electrode size, number, position, and configuration.
  • the invention patent with publication number CN114343646A discloses a multi-source fusion probe used in a muscle strength assessment method.
  • the multi-source fusion probe includes an elastomer and a conductive electrode on the surface of the elastomer.
  • the elastomer is used to collect pressure values
  • the conductive electrode is used to collect electromyographic values, so that the electromyographic values can be matched with the pressure values, and the pelvic floor muscle status can be evaluated from multiple dimensions of pressure values and pelvic floor electromyographic signals, and the data is objective and quantifiable.
  • the invention patent with publication number CN110916656A discloses a multi-channel pelvic floor muscle strength assessment system.
  • the system utilizes multi-channel electrodes: a first acquisition electrode, a second acquisition electrode, and a third acquisition electrode to map the acquired pelvic floor muscle surface electromyographic signals into adjacent electromyographic signal matrix relationships, thereby realizing the precise positioning analysis of the pelvic floor muscles by multi-channel electrodes with maximum efficiency.
  • This can effectively solve the defect that the number of pelvic floor electromyographic signal detection channels is low under the existing technical conditions, and the defect that pathological muscle groups cannot be accurately positioned is solved.
  • the invention patent with publication number CN111012326A discloses a pelvic floor muscle calibration method, which obtains and analyzes the pelvic floor muscle ultrasonic image collected by an ultrasonic acquisition system to obtain a first pelvic floor muscle evaluation result value based on ultrasound; and obtains and analyzes the pelvic floor muscle bioelectric signal collected by a pelvic floor muscle biofeedback acquisition system to obtain a second pelvic floor evaluation result value based on the pelvic floor muscle bioelectric signal; then, based on the determined degree of match between the first pelvic floor muscle pelvic bone result value and the second pelvic floor evaluation result value, determines whether the degree of match falls within a preset range of match; if not, calibrates the current bioelectric stimulation output parameters of the pelvic floor muscle biofeedback acquisition system to improve the accuracy of the pelvic floor evaluation.
  • Pelvic floor muscle testing relies on the pressure signal and electrical signal provided by the vaginal probe as evaluation data. This data can only judge the strength of the pelvic floor muscles and cannot provide an overall evaluation value for the pelvic floor function.
  • the existing pelvic floor muscle assessment method can only obtain relevant numerical values collected by the vaginal probe, but cannot obtain visual and imaging test results, and still cannot understand the overall pelvic floor function status.
  • the purpose of the present invention is to provide a multimodal data fusion pelvic floor function overall assessment method, which can enhance the fine structure touch and shallow and deep image recognition capabilities under the pelvic diaphragm through the deep fusion of optical, ultrasonic and pressure detection platforms, and improve the accuracy of pelvic floor diagnosis and treatment.
  • the present invention also provides a computer device for implementing the pelvic floor function overall assessment method.
  • the optical imaging module is used to collect data including the condition of the vaginal and rectal mucosa, abnormal proliferation, mucosal bulging, loosening and stacking, and anatomical site displacement to obtain basic information about the pelvic floor.
  • the ultrasound module is used to collect data including the submucosal matrix structure of the pelvic cavity organs and the deep pelvic nerves, blood vessels, muscles, bones and their corresponding anatomical displacements, so as to obtain ultrasound-based evaluation information on the pelvic tissue and nerve, blood vessel, and musculoskeletal functions.
  • the muscle strength and electromyography detection module is used to collect the pelvic floor muscle strength and electromyography data around the vagina, rectum, and urethral levator ani hiatus to obtain pelvic floor muscle function assessment information based on muscle strength and electromyography data.
  • a multi-module fusion vaginal rectal probe is used to integrate the optical imaging module, ultrasound module, muscle strength and electromyography detection module into the multi-module fusion vaginal rectal probe.
  • the multi-module fusion vaginal rectal probe is used to send the obtained basic pelvic floor information, pelvic tissue and nerve, blood vessel, musculoskeletal function evaluation information, and pelvic floor muscle strength and electromyography data to the computer to achieve multi-module signal fusion and obtain the overall evaluation information of the pelvic floor function.
  • the multi-module fusion vaginal rectal probe is a detection finger sleeve
  • the optical imaging module is arranged at the fingertip of the detection finger sleeve
  • the ultrasound module is arranged at the finger web of the first knuckle of the detection finger sleeve
  • the muscle strength and electromyography detection module are arranged on the entire surface of the detection finger sleeve, and a transparent film inflation start working mode is adopted.
  • Data acquisition is achieved by controlling the tilt, push, pull and/or rotation of the multi-module fusion vaginal rectal probe.
  • the multi-module signal fusion described in the present invention is to extract the position information of the required area in the image, determine the basic positioning using the U-Net network multi-classification segmentation method, and then obtain a three-dimensional binary image for the overall functional evaluation of the pelvic floor through semi-automatic reconstruction and segmentation of the image. Finally, after manual error verification and image conversion, a pelvic floor function overall evaluation model can be constructed for the expression of pelvic floor function overall evaluation information.
  • a computer device comprises a memory, a processor and a computer program stored and executed on the processor, wherein the processor implements the above-mentioned method for overall assessment of pelvic floor function when executing the computer program.
  • a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the above-mentioned pelvic floor function overall assessment method.
  • a multi-modal data fusion pelvic floor function overall assessment system including a multi-module fusion vaginal rectal probe integrated with an optical imaging module, an ultrasound module, a muscle strength and electromyography detection module,
  • the optical imaging module is used to collect data including the condition of the vaginal and rectal mucosa, abnormal proliferation, mucosal swelling, loosening and stacking, and anatomical site displacement to obtain basic information of the pelvic floor;
  • the ultrasound module is used to collect data including the submucosal matrix structure of the pelvic cavity organs and the deep pelvic nerves, blood vessels, muscles, bones and their corresponding anatomical displacements, so as to obtain the pelvic tissue and nerve, blood vessel, muscle and bone function assessment information based on ultrasound images;
  • the muscle strength and electromyography detection module is used to collect pelvic floor muscle strength and electromyography data around the vagina, rectum, and urethral levator anal hiatus to obtain pelvic floor muscle function assessment information based on the muscle strength and electromyography data;
  • the multi-module fusion vaginal rectal probe is used to send the obtained basic pelvic floor information, pelvic tissue and nerve, blood vessel, musculoskeletal function assessment information, and pelvic floor muscle strength and electromyography data to a computer to achieve multi-module signal fusion and obtain overall pelvic floor function assessment information;
  • the multi-module fusion vaginal rectal probe is a detection finger sleeve, wherein the optical imaging module (1) is arranged at the fingertip of the detection finger sleeve, the ultrasound module is arranged at the web of the first knuckle of the detection finger sleeve, the muscle strength and electromyography detection module is arranged on the entire surface of the detection finger sleeve, and a transparent film inflation start working mode is adopted.
  • the present invention has the following advantages and beneficial effects:
  • the present invention adopts for the first time a multimodal data fusion approach in the process of pelvic floor function assessment, namely the deep fusion of optical, ultrasonic and pressure detection, to achieve a comprehensive evaluation of the inner cavity of the pelvic organs and their surrounding nerves, blood vessels, and musculoskeletal systems from the mucosal surface to the internal tissue state of the interstitium, providing a more accurate and objective basis for pelvic floor diagnosis and treatment, and thus improving the accuracy of pelvic floor diagnosis and treatment.
  • the present invention uses a multi-module fusion vaginal-rectal probe to achieve comprehensive integrated detection of the vagina and rectum. It can simultaneously obtain basic pelvic floor information based on optical images, pelvic tissue and nerve, blood vessel, and musculoskeletal function assessment information based on ultrasound images, and pelvic floor muscle strength and electromyography data. After these data are deeply fused by a computer, a comprehensive and three-dimensional display of the overall status of the pelvic floor function can be obtained. This is a problem that has not been solved by all current instruments and equipment, and can provide a basis for scientific research on various pelvic floor functional diseases.
  • the present invention actually provides an integrated pelvic floor function detection and evaluation method that integrates optical, ultrasonic and pressure detection on a multi-module fusion vaginal rectal probe.
  • the method realizes the integration of optical, ultrasonic and pressure detection, and can improve the detection efficiency and detection accuracy (the integrated detection enables the body position to remain consistent during multiple detection processes); the method fuses the video image of optical detection, the two-dimensional image of ultrasonic detection and the curve graph of pressure detection to obtain a three-dimensional stereoscopic image, which can be used to realize the display of the all-round overall state of pelvic floor tissues, muscles, bones, nerves, etc., and provide an objective basis for pelvic floor diagnosis and treatment and scientific research.
  • FIG1 is a schematic diagram of the structure of the multi-module fusion vaginal rectal probe of the present invention.
  • FIG. 2 is a rectal endoscopy picture (I) of the test subject XX in Example 4 of the present invention.
  • FIG. 3 is a rectal endoscopy picture (II) of the test subject XX in Example 4 of the present invention.
  • FIG4 is a cross-sectional picture of the anal canal of the test subject XX in Example 4 of the present invention (A: internal anal sphincter; B: external anal sphincter).
  • FIG. 5 is a longitudinal section of the anal canal of the test subject XX in Example 4 of the present invention (C: sagittal section of the anal canal).
  • Example 6 is a longitudinal cross-sectional picture of the urethra and vagina of subject XX in Example 4 of the present invention (D: urethra; E: vagina).
  • FIG. 7 is a pelvic floor electromyography curve of subject XX in Example 4 of the present invention.
  • FIG8 is a three-dimensional schematic diagram of the pelvic floor of subject XX in Example 4 of the present invention (only the pelvic floor muscles are shown).
  • FIG. 9 is a three-dimensional schematic diagram (lateral view) of the pelvic floor tissue of the test subject XX in Example 4 of the present invention.
  • FIG. 10 is a three-dimensional schematic diagram (frontal view) of the pelvic floor blood vessels and muscles of the test subject XX in Example 4 of the present invention.
  • 1 is an optical imaging module
  • 2 is an ultrasonic module
  • 3 is a transparent film
  • 4 is a finger.
  • the fine structures under the pelvic diaphragm are: the anatomical structures of the pelvic floor muscles (puborectalis, iliocaudal, pubocaudal, pubovaginal, bulbospongiosus, superficial transverse perineal, ischiocavernosus, internal and external anal sphincters, internal and external urethral sphincters, etc.), fascia (pubourethral fascia, pubocervical fascia, rectovaginal fascia, superior fascia of the pelvic diaphragm, etc.), ligaments (levator ani tendon arch, sacrospinous ligament, anterior longitudinal ligament, pelvic fascia tendon arch, etc.), nerves (pudendal nerve, sacral plexus, hypogastric plexus, pelvic nerve, etc.), blood vessels (internal pudendal artery and vein, perineal artery and vein, inferior rectal artery and vein, etc.) and the organs that run through the pelvic diaphragm
  • Valsalva maneuver is to take a deep breath, close the glottis, and then exhale forcefully, fighting against the closed epiglottis when exhaling, increasing the intrathoracic and intra-abdominal pressures.
  • Kegal movement is one of the most commonly used ways to exercise the pelvic floor muscles. Kegel exercises are often used to reduce urinary incontinence and postpartum urinary incontinence in women.
  • the first step is to contract the pelvic floor muscles for 5 seconds. At the beginning, you may only be able to contract them for 2-3 seconds. This is normal. Just persist for a long time.
  • the second step is to relax the muscles for 10 seconds to give the pelvic floor muscles time to rest and avoid strain; the third step is to repeat the exercise 10 times, which is considered a set of Kegel training. You can rest after you finish. After training for a period of time, slowly increase the time of each contraction of the pelvic floor muscles to 10 seconds. Maintain the strength of 10 seconds of contraction and 10 seconds of relaxation to exercise, also 10 times a set, 3-4 sets a day.
  • point Aa is the distance between the midline of the anterior vaginal wall and the hymen edge in the Valsalva state, which is 3 cm away from the hymen edge in the relaxed state.
  • the value range is -3 to +3 cm.
  • the value of point Ba is the distance from the lowest point of the prolapsed part of the anterior vaginal wall to the edge of the hymen after point Aa in the Valsalva state.
  • the value range is -3 to the total length of the vagina.
  • the Ap point value is: the distance between the midline of the posterior vaginal wall and the hymen edge in the Valsalva state, 3 cm away from the hymen edge in the relaxed state.
  • the value range is -3 to +3 cm.
  • the Bp point value is the distance from the lowest point of the prolapsed part of the posterior vaginal wall to the edge of the hymen after the Ap point in the Vasalva state.
  • the value range is -3cm to the length of the vagina.
  • Point C value is: the distance from the external cervical opening (equivalent to the vaginal stump in patients with hysterectomy) to the edge of the hymen in the Vasalva state.
  • the value range is - vaginal length ⁇ vaginal length.
  • the value of point D is the distance between the posterior vaginal fornix and the edge of the hymen in the Valsalva state.
  • the value range is - vaginal length ⁇ vaginal length.
  • Example 1 Multi-module fusion vaginal rectal probe
  • the multi-module fusion vaginal rectal probe of the present embodiment is a detection finger sleeve
  • an optical imaging module 1 is integrated at the finger tip of the detection finger sleeve to collect data such as the condition of the mucosa in the vagina and rectal cavity, abnormal hyperplasia, mucosal swelling, loose stacking, and anatomical site displacement.
  • An ultrasound module 2 is integrated at the web of the first knuckle of the detection finger sleeve to collect data on the submucosal matrix structure of the pelvic cavity organs and the deep nerves, blood vessels, muscles, bones and their corresponding anatomical displacements in the pelvic cavity.
  • the entire outer surface of the detection finger sleeve is covered with a muscle strength and electromyography detection module, and the working mode is activated by inflating the transparent film 3 to collect pelvic floor muscle strength and electromyography data around the vagina, rectum, and urethral levator ani hiatus.
  • These data are collected in the storage of the multi-module fusion vaginal rectal probe and can be transmitted to a computer via Bluetooth or other wireless/wired transmission methods, and then a multi-modal data fusion pelvic floor function overall evaluation is performed.
  • the inflating and starting working mode through the transparent film 3 means that a layer of transparent film 3 is arranged on the outer surface of the detection finger sleeve.
  • the muscle force and electromyography detection module arranged on the outer surface of the detection finger sleeve is activated to collect the relevant signal generated by the pressure of the transparent film 3, which can be an electrical signal or other signals.
  • Figure 1 is only a simple example, and does not show the specific structure of the detection finger sleeve and the muscle force and electromyography detection module.
  • functional modules such as detection, transmission and storage are integrated into corresponding positions of the detection finger sleeve.
  • the detection finger sleeve is used to be mounted on the front knuckle of the finger 4.
  • the transparent film 3 outside the detection finger sleeve covers the entire finger 4. When not inflated, the transparent film 3 can be attached to the detection finger sleeve and the knuckle at the lower end of the detection finger sleeve. When inflated, it can bulge outward and contact the part to be tested.
  • Example 2 Overall assessment of pelvic floor function using multimodal data fusion
  • Step 1 After the multi-module fusion vaginal rectal probe is powered on and connected to a computer, the operator wears the multi-module fusion vaginal rectal probe and inserts it into the tester's vagina for testing.
  • Step 2 During the detection process, the multi-module fusion vaginal rectal probe is controlled to tilt, push, pull and/or rotate to realize data collection, and the basic information of the pelvic floor based on the optical image is obtained through the optical imaging module 1, and the pelvic tissue and nerve, blood vessel, and musculoskeletal function evaluation information based on the ultrasound image is obtained through the ultrasound module 2; the pelvic floor muscle function evaluation information based on muscle strength and electromyography data is obtained through the muscle strength and electromyography detection module.
  • the basic information of the pelvic floor includes but is not limited to the condition of the vaginal and rectal mucosa, abnormal proliferation, mucosal bulging, loosening and stacking, and anatomical site displacement.
  • Examples include: mucosal color, congestion, bleeding, erosion, ulcer, epithelial thickness, pigmentation, abnormal vascular proliferation, abnormal tissue proliferation, sinus tract, fistula, degree of mucosal prolapse, static and Valsalva POPQ prolapse grading, etc.
  • the information on the evaluation of pelvic tissues and nerve, blood vessel, and musculoskeletal functions includes, but is not limited to, data on the submucosal matrix structure of pelvic cavity organs and deep pelvic nerves, blood vessels, muscles, bones, and their corresponding anatomical displacement.
  • the pelvic floor hiatus and organ displacement landmarks that can be listed include: measuring the vertical distance between the lowest point of the bladder, cervix, and rectal ampulla and the posterior lower edge of the pubic symphysis to the horizontal line of the sacrococcygeal joint in resting and Valsalva state, bladder neck mobility, urethral internal orifice funnel formation, bladder posterior angle/urethra rotation angle, prostate volume; as well as pelvic floor muscle volume, levator ani hiatus area, urethra and anal sphincter integrity, bladder residual urine; as well as the tortuosity of the internal and external iliac blood vessels in the pelvis and the internal iliac vein and its branches in the pelvic floor, internal pudendal blood vessels and nerves, sciatic nerve course and abnormality, pubic symphysis separation, and sacrococcygeal joint abnormality, etc.
  • the pelvic floor muscle function assessment information includes but is not limited to the pelvic floor muscle strength and electromyographic data around the vagina, rectum, and urethral levator ani hiatus. Examples include: static and Valsalva pelvic floor levator ani muscle-based first-class muscle and sphincter-based second-class muscle strength, electromyographic signals, and electromyographic variability.
  • the patient was kept static, and the color, congestion, bleeding, erosion, ulcer, epithelial thickness, pigmentation, abnormal vascular proliferation, abnormal tissue proliferation, sinus, fistula of the cavity mucosa; submucosal matrix structure and deep pelvic nerves, blood vessels, muscles, bones and other related data were collected.
  • POPQ grading score including the acquisition of Aa point value, Ba point value, Ap point value, Bp point value, C point value and D point value
  • the vertical distance from the lowest point of the bladder, cervix and rectal ampulla to the posterior lower edge of the pubic symphysis to the horizontal line of the sacrococcygeal joint bladder neck activity, urethral internal orifice funnel formation, bladder posterior angle/urethral rotation angle, pelvic floor muscle volume, anal levator hiatus area, urethra and anal sphincter integrity and other related data were collected.
  • relevant data on the muscle strength, electromyographic signals and electromyographic variability of the first-class muscles and the second-class sphincters are collected.
  • the optical imaging module 1, the ultrasonic module 2, the muscle strength and electromyography detection module used in this step can all be implemented by conventional functional modules.
  • Step three the collected data is transmitted to the computer via the Bluetooth module of the multi-module fusion vaginal rectal probe.
  • the computer processor analyzes and processes the obtained overall assessment information of the pelvic floor function to obtain a three-dimensional image that fully displays the pelvic floor function, which serves as the basis for the overall assessment of the pelvic floor function.
  • Example 3 Constructing a comprehensive assessment model for pelvic floor function
  • Step 1 Extract DICOM images (i.e., the medical images and related information obtained in the detection process in step 2 of embodiment 2), use window-leveling algorithm or other feasible methods to analyze the images, and extract the image and position features of the levator ani muscles, uterus and vagina, rectum and anal canal, bladder and urethra in the image through analysis.
  • DICOM images i.e., the medical images and related information obtained in the detection process in step 2 of embodiment 2
  • window-leveling algorithm or other feasible methods to analyze the images, and extract the image and position features of the levator ani muscles, uterus and vagina, rectum and anal canal, bladder and urethra in the image through analysis.
  • Step 2 Semi-automatic reconstruction and segmentation
  • the U-Net network multi-classification segmentation method is used to segment the left coccygeus muscle, right coccygeus muscle, left iliococcygeus muscle, right iliococcygeus muscle, left levator anus muscle, right levator anus muscle, intestine, uterus and bladder into 9 regions on the extracted image (T2) sequence.
  • the T2 sequence is used as the model input, and the input dimension is increased from three dimensions to four dimensions, that is, the original image dimension is improve to .
  • the area in the image that is equal to 0 is cut off, and only the non-zero area is retained.
  • normalization processing is performed, and then the image is standardized to establish an image segmentation algorithm, as shown in formula (1):
  • HU value of each point in the image is the HU value of each point in the image
  • the mean HU value of all points in the image is: is the HU standard deviation of all points in the image, is the result after standardization transformation.
  • Dice loss and cross entropy (CE) loss are used as cost functions to train the network
  • Gaussian weight code splicing Patch and morphological operations are used to post-process the mask obtained by the pelvic floor muscle tissue segmentation module to achieve semi-automatic reconstruction and segmentation of the image.
  • the post-processing is divided into three steps: Gaussian weight code splicing Patch, connected domain analysis, and image scaling. Finally, a three-dimensional binary image for the overall functional evaluation of the pelvic floor can be obtained.
  • the manually verified three-dimensional binary image is converted into a Mesh surface model, and part of the model surface is smoothed using algorithms such as Gaussian smoothing to improve the observation experience.
  • a holistic evaluation model of pelvic floor function is constructed to achieve the fusion of multimodal data and ultimately obtain an overall objective image that is beneficial as a basis for pelvic floor diagnosis and treatment.
  • the multi-module fusion vaginal rectal probe shown in Figure 1 was used to detect the pelvic floor function of the test subject XX and perform an overall assessment.
  • the test subject is required to empty the rectum, and the multi-module fusion vaginal rectal probe is used for examination by vaginal ultrasound or rectal digital examination.
  • the basic pelvic floor information and ultrasound image-based assessment information are obtained and displayed in real time on the host, as shown in Figures 2 and 3, which are rectoscope images; Figures 4 and 5, which are cross-sectional and longitudinal images of the anal canal; and Figure 6, which is a longitudinal section of the urethra and vagina.
  • the multi-module fusion vaginal rectal probe of this embodiment can be used to detect the pressure values of multiple sites, by setting a pressure measuring point at the top of the multi-module fusion vaginal rectal probe, and then setting three groups in sequence around it, each group evenly distributed with 4 pressure measuring points, a total of 13 pressure measuring points. As shown in Figure 7, the pressure curve corresponding to each pressure measuring point within the sampling time of 10 seconds can be displayed separately, where the numbers 0 to 12 represent different pressure measuring points, and the pressure value (kPa) represents the average pressure corresponding to different pressure measuring points.
  • Figure 8 is a three-dimensional schematic diagram showing the pelvic floor muscles.
  • Figure 9 is a three-dimensional schematic diagram showing the pelvic floor tissues.
  • Figure 10 is a three-dimensional schematic diagram showing the pelvic floor blood vessels and muscles.
  • the corresponding images or pictures can be fused into a display interface for display through a computer, for example: the image used to reflect the patient's colonoscope is displayed in a display interface, that is, the pictures shown in Figures 2 and 3 appear in the display interface of a computer at the same time, and the pictures shown in Figures 4, 5, 6 and 7 appear in the display interface of a computer at the same time.
  • the present invention has the following advantages:
  • Multimodal fusion data is proposed for the first time (such as the three-dimensional schematic diagrams of Figures 8, 9, and 10 shown in Example 4), which can reflect the relevant functions of the pelvic floor structure as a whole.
  • the existing traditional pelvic floor function assessment method is still that the pelvic floor muscle assessment uses a vaginal probe to collect pressure signals and electrical signals as evaluation values, and the pelvic tissue assessment uses ultrasonic detection to obtain measurement data of relevant tissues.
  • doctors still need to analyze and judge based on the evaluation values or measurement data obtained by their respective tests in order to obtain overall assessment data.
  • the experience of clinicians is the key to judging the accuracy of the overall pelvic floor assessment.
  • Multimodal fusion data can be obtained by using the finger cuff detection method.
  • the data acquisition method is simpler than the existing traditional pelvic floor function assessment method. It omits the separate detection methods such as vaginal digital examination, pelvic floor ultrasound, and muscle strength and electromyography detection, which greatly saves clinical detection time and contributes to the improvement of the overall efficiency of pelvic floor function assessment.
  • Multimodal fusion data can also solve the shortcomings of current pelvic floor dysfunction diseases or key big data. The reason is that: ultrasound examination is completed at the same time as digital examination, and multimodal data is effectively integrated, which can timely and accurately judge the overall functional status of the pelvic floor.
  • the fusion referred to in the present invention includes two types. One is to fuse multimodal data through the constructed pelvic floor function overall evaluation model to obtain a three-dimensional image that can reflect the overall function of the pelvic floor, see Figures 8, 9 and 10; the other is to use a multi-module fusion probe to simultaneously perform optical, ultrasonic, electromyographic and other tests, and fuse the images or pictures obtained from the tests on the same computer display interface to avoid the tediousness of separate tests and the inconvenience of data retrieval.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Radiology & Medical Imaging (AREA)
  • Physics & Mathematics (AREA)
  • Biomedical Technology (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Pathology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Biophysics (AREA)
  • Veterinary Medicine (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Quality & Reliability (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)

Abstract

一种多模态数据融合的盆底功能整体评估方法及设备，利用光学成像模块（1），获得盆底基本信息；利用超声模块（2），获得基于超声影像的盆腔组织和神经、血管、肌骨功能评估信息；利用肌肉肌力和肌电检测模块，获得基于肌力肌电数据的盆底肌功能评估信息，然后再根据光学成像模块（1）、超声模块（2）、肌肉肌力和肌电检测模块的多模块信号融合，获得盆底功能整体评估信息。通过光学、超声和压力检测平台的深度融合，能够提升盆膈下精细结构感触和浅深层影像辨识能力，实现盆底诊疗精准度的提升。

Description

一种多模态数据融合的盆底功能整体评估方法及设备 技术领域
本发明是一种多模态数据融合的盆底功能整体评估方法及设备，具体涉及一种基于声光电多模态实现人体盆底功能整体评估的方法及实现该方法的计算机设备和计算机可读存储介质，属于医疗诊断检测技术领域。
 背景技术
女性盆底功能障碍性疾病（pelvic floor disfunction, PFD)是由多因素导致盆底支持结构薄弱，造成盆腔脏器位置和功能异常的一组疾病，包括：盆腔器官脱垂（pelvic organ prolapse, POP)、尿失禁（urinary incontinence, UI）、粪失禁、性功能障碍和慢性盆腔疼痛等，病程长而且复发率高，是一类严重影响患者生活质量和身心健康的疾病，对个人、家庭和社会造成严重负担，目前PFD已被认为是影响人类生活质量的五大慢性病之一。
盆底结构复杂，涉及肌肉、筋膜、韧带、血管和神经等多重组织，构成泌尿、生殖、肛肠等不同器官关联的整体。盆底的肌肉和筋膜形成了整体支撑结构，其中肌肉状态对于盆底支撑功能的实现尤为重要。盆底肌肉由3个层面的肌肉组成，包括盆膈（肛提肌及上下筋膜）、深横肌及筋膜、会阴浅层肌群及筋膜，只有盆底肌肉正常协同收缩，才能正常发挥盆底肌肉的协调控制作用。当盆底肌肉退化、松弛、肌力不足时，会引起盆腔脏器解剖移位、关闭功能异常等情况而导致功能障碍，所以如何衡量评价盆底功能状态是目前盆底疾病诊疗中的重要评价指标和选择治疗方案的前提内容之一。
国际指南建议对女性PFD患者首先进行盆底功能整体评估再制订适宜的诊疗方案。临床上最常用的肌力评估法方法是手指触诊和简易气囊阴道测压法。盆底手测肌力很大程度上依靠临床医师的主观经验进行判断，准确度和重复度不高；阴道测压装置简易粗糙，仅能大致反应阴道收缩后气囊内的整体压力变化，并不能真实反映盆底深层和浅层肌肉的收缩力和维持时间。目前临床上常用的间接衡量盆底肌肉收缩能力的表面肌电检测，来大致评估盆底肌功能。市面上的阴道内表面肌电探头可以粗略测量平均肌电，但由于探测棒上的探测片或环位置少（一般两个）、测量精度低、活动后阴道内位移等多个因素的影响，测量并不准确，可重复性差，可能使临床医生判断失误，导致临床治疗方案产生偏差。
目前盆底肌活性的评估主要是定性的，关注于信号线的形状，从休息到收缩阶段的信号振幅的数量，耐力收缩期间表面肌电信号的持续时间和质量，比较治疗前后盆底肌功能的变化，仅有少数研究对此类变化进行定量研究。Keshwani等对比16款商业化阴道探头发现（Neurourol Urodyn. 2015 Feb;34(2):104-12），所有探头在设计上均有缺陷，包括几何形状、电极大小、数量、位置和配置。
现有技术中，针对阴道探头的设计缺陷，公开号为CN114343646A的发明专利公开了一种肌力评定方法使用的多源融合探头，该多源融合探头包括弹性体和弹性体表面的导电电极，利用弹性体采集压力值，导电电极采集肌电值，从而能够将肌电值对应压力值进行数据匹配，从压力值及盆底肌电信号多维度评价盆底肌状态且数据客观、可量化。
公开号为CN110916656A的发明专利公开了一种多通道盆底肌肌力评估系统，该系统利用多通道电极：第一采集电极、第二采集电极和第三采集电极，将采集到的盆底肌表面肌电信号进行相邻肌电信号矩阵关系映射，从而最大效率的实现了多通道电极对盆底肌的精准定位分析，可有效解决现有技术条件下盆底肌电信号检测通道数低，无法实现精准定位病态肌群的缺陷。
公开号为CN111012326A的发明专利公开了一种盆底肌校准方法，该方法通过获取超声采集系统采集到的盆底肌超声图像并进行分析，得到基于超声的第一盆底肌评估结果值；以及获取盆底肌生物反馈采集系统采集到的盆底肌生物电信号并进行分析，得到基于盆底肌生物电信号的第二盆底评估结果值；然后根据确定的第一盆底肌盆骨结果值和第二盆底评估结果值的匹配程度，判断该匹配程度是否落在预设的匹配程度范围内，若否，则校准所述盆底肌生物反馈采集系统当前的生物电刺激输出参数，以提高盆底评估的准确性。
由此可知，现有阴道探头虽然从信息采集方式、信息采集数量以及信息准确程度等方面有所提升，但仍然存在以下问题：
（1）盆底肌检测依靠阴道探头提供的压力信号及电信号作为评估数据，该数据仅能对盆底肌的肌力情况进行判断，无法从整体上对盆底功能提供评估价值。
（2）现有盆底肌评估方法仅能获得阴道探头所采集而得到的相关数值，无法获得可视化及影像化的检测结果，依然无法了解整体盆底功能状态。
（3）现有技术虽然提出了基于超声影像对盆底肌评估结果进行修正，但需要通过超声采集系统单独对患者采集盆底肌超声图像，增加了检测流程，并无法直接和盆底肌肉参数指标作对映或者整体评价。
（4）现有的单独阴道镜或直肠镜可以对黏膜表面做评价，但又是单独的两项检测，因不同检查者和检测时空的变化，导致疾病整体对应性很差。
综上所述，目前仍缺乏一体化多模态全方位盆底功能整体评估方法和系统。
 发明内容
本发明的目的是提供一种多模态数据融合的盆底功能整体评估方法，该评估方法通过光学、超声和压力检测平台的深度融合，能够提升盆膈下精细结构感触和浅深层影像辨识能力，实现盆底诊疗精准度的提升。为此，本发明还提供了实现该盆底功能整体评估方法的计算机设备。
本发明通过下述技术方案实现：
利用光学成像模块，采集包括阴道内及直肠腔内黏膜状况、异常增生、黏膜膨隆松弛堆叠及解剖位点移位的数据，获得盆底基本信息，
利用超声模块，采集包括盆腔空腔脏器黏膜下基质结构及盆腔深部神经、血管、肌肉、骨骼及其相应解剖位移的数据，以获得基于超声影像的盆腔组织和神经、血管、肌骨功能评估信息，
利用肌肉肌力和肌电检测模块，采集阴道、直肠、尿道肛提肌裂孔周围的盆底肌肉肌力肌电数据，以获得基于肌力肌电数据的盆底肌功能评估信息，
采用多模块融合阴道直肠探头，将光学成像模块、超声模块、肌肉肌力和肌电检测模块集成于多模块融合阴道直肠探头上，多模块融合阴道直肠探头用于将获得的盆底基本信息，盆腔组织和神经、血管、肌骨功能评估信息，盆底肌肉肌力肌电数据发送至计算机，实现多模块信号融合, 获得盆底功能整体评估信息，
所述多模块融合阴道直肠探头为探测指套，将光学成像模块设于探测指套的指尖部位，将超声模块设于探测指套第一指节的指腹部位，将肌肉肌力和肌电检测模块设于整个探测指套表面，并采用透明膜充气启动工作模式。
通过控制多模块融合阴道直肠探头倾斜、推拉和/或旋转来实现数据采集。
在采用多模块融合阴道直肠探头进行数据采集时，患者行静态、Valsalva动态和/或kegal动态。
本发明所述多模块信号融合是通过提取影像中所需区域的位置信息，利用U-Net网络多分类分割方法确定基本定位，然后再通过图像的半自动重建分割，获得盆底整体功能评估的三维二值图像，最后经人工核验误差和图像转换后，即可构建得到盆底功能整体评估模型，用于盆底功能整体评估信息的表达。
一种计算机设备，包括储存器、处理器以及存储在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如上述所述的盆底功能整体评估方法。
一种计算机可读存储介质，所述计算机可读存储介质储存有计算机程序，所述计算机程序被处理器执行时实现如上述的盆底功能整体评估方法。
一种多模态数据融合的盆底功能整体评估系统，包括集成有光学成像模块、超声模块、肌肉肌力和肌电检测模块的多模块融合阴道直肠探头，
所述光学成像模块用于采集包括阴道内及直肠腔内黏膜状况、异常增生、黏膜膨隆松弛堆叠及解剖位点移位的数据，获得盆底基本信息；
所述超声模块用于采集包括盆腔空腔脏器黏膜下基质结构及盆腔深部神经、血管、肌肉、骨骼及其相应解剖位移的数据，以获得基于超声影像的盆腔组织和神经、血管、肌骨功能评估信息；
所述肌肉肌力和肌电检测模块用于采集阴道、直肠、尿道肛提肌裂孔周围的盆底肌肉肌力肌电数据，以获得基于肌力肌电数据的盆底肌功能评估信息；
所述多模块融合阴道直肠探头用于将获得的盆底基本信息，盆腔组织和神经、血管、肌骨功能评估信息，盆底肌肉肌力肌电数据发送至计算机，实现多模块信号融合, 获得盆底功能整体评估信息；
所述多模块融合阴道直肠探头为探测指套，将光学成像模块（1）设于探测指套的指尖部位，将超声模块设于探测指套第一指节的指腹部位，将肌肉肌力和肌电检测模块设于整个探测指套表面，并采用透明膜充气启动工作模式。
本发明与现有技术相比，具有以下优点及有益效果：
（1）本发明首次在盆底功能评估过程中采用多模态数据融合的方式，即光学、超声和压力检测的深度融合，实现了盆腔脏器的内腔及其周围神经、血管、肌骨系统从黏膜表面到间质内部组织状态的全面评价，为盆底诊疗提供更加精确的客观依据，可提升盆底诊疗的精准度。
（2）本发明采用多模块融合阴道直肠探头可实现阴道及直肠内全方位的一体化检测，可以同时获取基于光学影像的盆底基本信息、基于超声影像的盆腔组织和神经、血管、肌骨功能评估信息、盆底肌肉肌力肌电数据，这些数据经计算机深度融合后，即可获得全方位立体展示盆底功能的整体状态，是目前所有器械装备还未能解决的问题，可为各种盆底功能疾病的科学研究提供依据。
综上所述，本发明实际提供了一种将光学、超声和压力检测集成在多模块融合阴道直肠探头上而进行的一体化盆底功能检测评估方法，该方法实现了光学、超声和压力检测的一体化，可提高检测效率及检测精度（一体化检测使得多项检测过程中体位均能保持一致）；该方法将光学检测的视频影像、超声检测的二维图像和压力检测的曲线图融合得到三维立体图像，可用于实现盆底组织、肌骨、神经等全方位整体状态的展示，为盆底诊疗和科学研究提供客观依据。
 附图说明
图1为本发明所述多模块融合阴道直肠探头的结构示意图。
图2为本发明实施例4中测试者XX的直肠镜图片（一）。
图3为本发明实施例4中测试者XX的直肠镜图片（二）。
图4为本发明实施例4中测试者XX的肛管横断面图片（A：肛门内括约肌；B肛门外括约肌）。
图5为本发明实施例4中测试者XX的肛管纵断面图片（C：肛管矢状切面）。
图6为本发明实施例4中测试者XX的尿道阴道纵断面图片（D：尿道；E：阴道）。
图7为本发明实施例4中测试者XX的盆底肌电曲线。
图8为本发明实施例4中测试者XX的盆底三维示意图（仅显示盆底肌肉）。
图9为本发明实施例4中测试者XX的盆底组织的三维示意图（侧位）。
图10为本发明实施例4中测试者XX的盆底血管和肌肉的三维示意图（正位）。
上述图1中，1—光学成像模块，2—超声模块，3—透明膜，4—手指。
上述图2、3、8、9、10中实际获得的图像均为彩图，更便于作为组织形态评价的依据。
 实施方式
下面将本发明的发明目的、技术方案和有益效果作进一步详细的说明。
应该指出，以下详细说明都是示例性的，旨在对所要求的本发明提供进一步的说明，除非另有说明，本文使用的所有技术和科学术语具有与本发明所属技术领域的普通技术人员通常理解的相同含义。
盆膈下精细结构是：盆底各组肌肉（耻骨直肠肌、髂尾肌、耻尾肌、耻骨阴道肌、球海绵体肌、会阴浅横肌、坐骨海绵体肌、肛门内外括约肌、尿道内外口括约肌等）、筋膜（耻骨尿道筋膜、耻骨宫颈筋膜、直肠阴道筋膜、盆膈上筋膜等）、韧带（肛提肌腱弓、骶棘韧带、前纵韧带、盆筋膜腱弓等）、神经（阴部神经、骶神经丛、腹下神经丛、盆神经等）、血管（阴部内动静脉、会阴动静脉、直肠下动静脉等）及贯穿盆膈的各器官（膀胱尿道、子宫阴道、肛管直肠）的解剖结构和在完成控尿、控便和生殖等生理功能的各种病理生理状态。
Valsalva动作是：深吸气后紧闭声门，再用力做呼气动作，呼气时对抗紧闭的会厌，增加胸内压和腹腔内压力的状态。
kegal动作是：凯格尔运动又称盆骨运动，是盆底肌锻炼的一种最常用的方式。凯格尔运动常被用来降低尿失禁、妇女的产后尿失禁问题。第一步，收缩盆底肌肉5秒钟，一开始做可能只能收缩2-3秒，这是正常的，长期坚持就好。第二步，放松肌肉10秒钟，给盆底肌肉休息的时间、避免拉伤；第三步，重复练习10次，算是一组凯格尔训练，做完就可以休息了。训练一段时间后，慢慢把每次收缩盆底肌的时间增加到10秒。保持10秒收缩，10秒放松的力度来锻炼，同样是10次一组，每天3-4组。 
Aa点数值是：放松状态下阴道前壁中线距处女膜缘3cm处在Valsalva状态下相对于处女膜缘的距离。数值范围为-3～+3cm。
Ba点数值是：Valsalva状态下，Aa点后阴道前壁脱出部最低点距处女膜缘的距离。数值范围-3～阴道总长度。
Ap点数值是：放松状态下阴道后壁中线距处女膜缘3cm处在Valsalva状态下相对于处女膜缘的距离。数值范围为-3～+3cm。
Bp点数值是：Vasalva状态下，Ap点后阴道后壁脱出部最低点距处女膜缘的距离。数值范围为-3cm～阴道长度。
C点数值是：Vasalva状态下，子宫颈外口（子宫切除者则相当于阴道残端）据处女膜缘的距离。数值范围为-阴道长度～阴道长度。
D点数值是：Valsalva状态下，阴道后穹窿据处女膜缘的距离。数值范围为-阴道长度～阴道长度。
实施例1：多模块融合阴道直肠探头
如图1所示，本实施例的多模块融合阴道直肠探头为探测指套，在探测指套的指尖部位集成有光学成像模块1，以用于采集阴道内及直肠腔内黏膜状况、异常增生、黏膜膨隆松弛堆叠及解剖位点移位等数据。在探测指套的第一指节的指腹部位集成有超声模块2，以用于采集盆腔空腔脏器黏膜下基质结构及盆腔深部神经、血管、肌肉、骨骼及其相应解剖位移的数据。在探测指套的整个外表面覆盖有肌肉肌力和肌电检测模块，通过透明膜3充气启动工作模式，以用于采集阴道、直肠、尿道肛提肌裂孔周围的盆底肌肉肌力肌电数据。这些数据被收集在多模块融合阴道直肠探头的储存器中，可通过蓝牙或其他无线/有线输送方式传送至计算机，再进行多模态数据融合的盆底功能整体评估。
通过透明膜3充气启动工作模式是指探测指套外套设一层透明膜3，当透明膜3充气鼓起时，启动设置于探测指套外表面的肌肉肌力和肌电检测模块工作，以用于收集透明膜3因压力而产生的相关信号，该信号可以是电信号或其他信号等。（图1仅为简单示例，并未示出探测指套以及肌肉肌力和肌电检测模块的具体结构。）
在一个具体的实施例中，将检测、传输和储存等功能模块集成于探测指套的相应位置上，探测指套用于套设在手指4前端指节上，探测指套外的透明膜3覆盖于整个手指4上，未充气时，透明膜3可贴合于探测指套及探测指套下端的指节上，充气时可向外鼓起并接触待测部位。
实施例2：多模态数据融合的盆底功能整体评估
采用图1所示多模块融合阴道直肠探头进行以下操作：
步骤一，将多模块融合阴道直肠探头开启电源后与计算机连通后，操作者佩戴多模块融合阴道直肠探头后插入测试者的阴道内进行检测。
步骤二，检测过程中，控制多模块融合阴道直肠探头倾斜、推拉和/或旋转来实现数据采集，并通过光学成像模块1获得基于光学影像的盆底基本信息，通过超声模块2获得基于超声影像的盆腔组织和神经、血管、肌骨功能评估信息；通过肌肉肌力和肌电检测模块获得基于肌力肌电数据的盆底肌功能评估信息，
其中，盆底基本信息包括但不限于阴道内及直肠腔内黏膜状况、异常增生、黏膜膨隆松弛堆叠及解剖位点移位等数据。可列举的如：黏膜色泽、充血、出血、糜烂、溃疡、上皮厚度、色素沉着、异常血管增生、异常组织增生、窦道、瘘孔、黏膜脱垂程度、静态与Valsalva下POPQ脱垂分度情况，等等。
盆腔组织和神经、血管、肌骨功能评估信息包括但不限于盆腔空腔脏器黏膜下基质结构及盆腔深部神经、血管、肌肉、骨骼及其相应解剖位移等数据。可列举的盆底裂孔及脏器移位标志性指标如：测量安静和Valsalva状态时膀胱、宫颈和直肠壶腹部最低点与耻骨联合后下缘到骶尾关节水平线的垂直距离，膀胱颈活动度，尿道内口漏斗形成，膀胱后角/尿道旋转角，前列腺体积；以及盆底肌肉体积，肛提肌裂孔面积，尿道及肛门括约肌完整度，膀胱残余尿；以及盆腔髂内外血管及盆底髂内静脉及其分支迂曲情况，阴部内血管及神经，坐骨神经走形及异常，耻骨联合分离及骶尾关节异常，等等。
盆底肌功能评估信息包括但不限于阴道、直肠、尿道肛提肌裂孔周围的盆底肌肉肌力肌电数据。可列举的如：静态和Valsalva下盆底肛提肌为主的一类肌和括约肌为主的二类肌肌力、肌电信号和肌电变异度情况。
进一步的，在采集上述数据时，使患者保持静态时，进行腔道黏膜色泽、充血、出血、糜烂、溃疡、上皮厚度、色素沉着、异常血管增生、异常组织增生、窦道、瘘孔；黏膜下基质结构及盆腔深部神经、血管、肌肉、骨骼等相关数据采集。使患者行Valsalva动态时，进行黏膜及组织脱垂情况，POPQ分度评分情况（包括Aa点数值、Ba点数值、Ap点数值、Bp点数值、C点数值及D点数值的获取），膀胱、宫颈和直肠壶腹部最低点与耻骨联合后下缘到骶尾关节水平线的垂直距离，膀胱颈活动度，尿道内口漏斗形成，膀胱后角/尿道旋转角，盆底肌肉体积，肛提肌裂孔面积，尿道及肛门括约肌完整度等相关数据采集。使患者行kegal动态时，进行一类肌和括约肌为主的二类肌肌力、肌电信号和肌电变异度情况的相关数据采集。
在一个具体的实施例中，该步骤中采用的光学成像模块1、超声模块2、肌肉肌力和肌电检测模块均可采用常规功能模块即可实现。
步骤三，采集到的数据经多模块融合阴道直肠探头的蓝牙模块传送至计算计中，计算机的处理器将获得的盆底功能整体评估信息进行分析和处理后，即可获得全方位展示盆底功能的立体影像，以作为盆底功能整体评估的依据。
实施例3：构建盆底功能整体评估模型
步骤一：提取DICOM图像（即实施例2步骤二中检测过程所获取的医学图像和相关信息） ，采用Window-leveling 算法或其他可行方法进行图像分析，通过分析来提取影像中肛提肌、子宫及阴道、直肠及肛管、膀胱及尿道等影像及位置特征。
步骤二：半自动重建分割
基于U-Net网络多分类分割方法，通过计算机在低分辨率上进行粗分割，确定上述器官和组织的基本定位；然后在原像图上精分割，对每一帧图像进行自标注，通过人工处理和重度结合，完成图像的半自动重建分割，以获得盆底整体功能评估的三维二值图像，准确度约95%以上。
进一步的，利用U-Net网络多分类分割方法是在提取的影像图像（T2）序列上分割出左尾骨肌、右尾骨肌、左髂尾肌、右髂尾肌、左肛提肌、右肛提肌、肠道、子宫和膀胱9个区域。将T2序列作为模型输入，并将输入维度从三维提高到四维，即：图像原始维度为 提高到 。将影像图像中等于0的区域裁剪掉，只保留非0的区域。为保证分割的准确性，进行归一化处理，然后对图像进行标准化处理，建立图像分割算法，如公式（1）所示：
          （1）
其中， 为影像图像中的每一个点的HU值， 该影像图像中所有点的HU均值， 为该影像图像中所有点的HU标准差， 为进过标准化变换后的结果。
然后采用Dice损失和交叉熵（CE）损失作为代价函数进行网络的训练，再使用高斯权重码拼接Patch及形态学操作对盆底肌组织分割模块得到的mask进行后处理，实现图像的半自动重建分割，其中，后处理又分为高斯权重码拼接Patch、连通域分析、图像缩放三个步骤，最终即可得到盆底整体功能评估的三维二值图像。
步骤三：人工核验 
临床医生和超声医生、盆底核磁共振医生进行人工核验，确定图像识别与定位是否正确，使组织器官定位的误差≤1.0mm。
步骤四：图像转化 
将人工核验后的三维二值图像转化为Mesh表面模型，使用Gaussian smoothing等算法平滑处理部分模型表面，以提升观察体验，即构建得到盆底功能整体评估模型，从而实现多模态数据的融合，并最终得到有利于作为盆底诊疗依据的整体客观图像。
实施例4：针对测试者XX进行的盆底功能整体评估
采用图1所示多模块融合阴道直肠探头对测试者XX的盆底功能进行检测同时进行整体评估。
首选，需要测试者XX排空直肠，多模块融合阴道直肠探头按阴道超声或直肠指诊方式进行检查，同时获得的盆底基本信息和基于超声影像的评估信息，并实时显示在主机上，如图2、图3所示，为直肠镜图片；如图4、图5所示，为肛管横断面和纵断面图片；如图6所示，为尿道阴道纵断面图片。
在获得上述检测数据后，停止多模块融合阴道直肠探头的动作，使测试者XX行平卧位，启动鼓气模式，气囊充气，同时展开测压点，即可获得盆底肌功能评估信息，并实时显示在主机上，参见图7。本实施例的多模块融合阴道直肠探头可用于检测多个位点的压强值，通过在多模块融合阴道直肠探头顶端设置一个测压点，周边再依次设置三组，每组均布4个测压点，共计13个测压点。如图7所示，可以分别显示每个测压点在采样时间10秒内对应的压强曲线，其中，数字0至12代表不同的测压点，压强值（kPa）代表不同测压点对应的平均压强。
将上述数据信息输入计算机中，利用实施例3的盆底功能整体评估模型进行分析和处理，即可获得全方位展示盆底功能的立体影像，如图8、图9和图10所示。其中，图8为显示盆底肌肉的三维示意图。图9为显示盆底组织的三维示意图。图10为显示盆底血管和肌肉的三维示意图。
在本实施例的应用过程中，在通过多模块融合阴道直肠探头获得所需的信息数据后，可通过计算机将对应的影像或图像融合在一个显示界面中进行展示，例如：用于反映患者直肠镜的图像在一个显示界面中展示，即：图2和图3所示图片同时出现在一个计算机的显示界面中，图4、图5、图6和图7所示图片同时出现在一个计算机的显示界面中。
基于以上实施例，本发明具有以下优点：
（1）首次提出多模态融合数据（如实施例4所示图8、图9、图10的三维示意图），能够从整体上反映盆底结构的相关功能，而现有传统盆底功能评估方法仍然是盆底肌评估采用阴道探头采集压力信号及电信号作为评估值，盆腔组织评估采用超声检测用于获取相关组织的测量数据，在临床工作中，仍然需要医生根据各自检测获得的评估值或测量数据进行分析和判断，才能获得整体评估数据，临床医生的经验值是判断盆底整体评估准确性的关键。
（2）采用指套探测方式即可获得多模态融合数据，数据的获取方式较现有传统盆底功能评估的检测方式更加简单，省略了阴道指诊、盆底超声、肌力肌电检测分开进行等检测方式，极大程度的节省了临床检测时间，为盆底功能整体评估效率的提升做出贡献。
（3）多模态融合数据还能解决目前盆底功能障碍性疾病或者关键大数据的短板，原因是： 指检的同时完成超声检查，将多模态数据进行了有效融合，能够及时准确判断盆底整体功能状态。
需要说明的是，本发明所指融合包括两种，一种是通过构建的盆底功能整体评估模型，将多模态数据融合而得到可反映盆底整体功能的立体影像，参见图8、图9和图10；一种是通过多模块融合探头同时进行光学、超声、肌电等检测，将检测所得影像或图像融合在同一计算机的显示界面中展示，避免单独检测的繁琐和数据查阅的不便。
以上所述，仅是本发明的较佳实施例，并非对本发明做任何形式上的限制，凡是依据本发明的技术实质对以上实施例所作的任何简单修改、等同变化，均落入本发明的保护范围之内。

Claims (5)

  1. 一种多模态数据融合的盆底功能整体评估方法，其特征在于： 
    利用光学成像模块（1），采集包括阴道内及直肠腔内黏膜状况、异常增生、黏膜膨隆松弛堆叠及解剖位点移位的数据，获得盆底基本信息，
    利用超声模块（2），采集包括盆腔空腔脏器黏膜下基质结构及盆腔深部神经、血管、肌肉、骨骼及其相应解剖位移的数据，以获得基于超声影像的盆腔组织和神经、血管、肌骨功能评估信息，
    利用肌肉肌力和肌电检测模块，采集阴道、直肠、尿道肛提肌裂孔周围的盆底肌肉肌力肌电数据，以获得基于肌力肌电数据的盆底肌功能评估信息，
    采用多模块融合阴道直肠探头，将光学成像模块（1）、超声模块（2）、肌肉肌力和肌电检测模块集成于多模块融合阴道直肠探头上，多模块融合阴道直肠探头用于将获得的盆底基本信息，盆腔组织和神经、血管、肌骨功能评估信息，盆底肌肉肌力肌电数据发送至计算机，实现多模块信号融合, 获得盆底功能整体评估信息，
    所述多模块融合阴道直肠探头为探测指套，将光学成像模块（1）设于探测指套的指尖部位，将超声模块（2）设于探测指套第一指节的指腹部位，将肌肉肌力和肌电检测模块设于整个探测指套表面，并采用透明膜（3）充气启动工作模式。
  2. 根据权利要求1所述的一种多模态数据融合的盆底功能整体评估方法，其特征在于：通过控制多模块融合阴道直肠探头倾斜、推拉和/或旋转来实现数据采集。
  3. 根据权利要求1所述的一种多模态数据融合的盆底功能整体评估方法，其特征在于：在采用多模块融合阴道直肠探头进行数据采集时，患者行静态、Valsalva动态和/或kegal动态。
  4. 一种计算机设备，其特征在于：包括储存器、处理器以及存储在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如权利要求1～3任一项所述的盆底功能整体评估方法。
  5. 一种多模态数据融合的盆底功能整体评估系统，其特征在于：包括集成有光学成像模块（1）、超声模块（2）、肌肉肌力和肌电检测模块的多模块融合阴道直肠探头，
    所述光学成像模块（1）用于采集包括阴道内及直肠腔内黏膜状况、异常增生、黏膜膨隆松弛堆叠及解剖位点移位的数据，获得盆底基本信息；
    所述超声模块（2）用于采集包括盆腔空腔脏器黏膜下基质结构及盆腔深部神经、血管、肌肉、骨骼及其相应解剖位移的数据，以获得基于超声影像的盆腔组织和神经、血管、肌骨功能评估信息；
    所述肌肉肌力和肌电检测模块用于采集阴道、直肠、尿道肛提肌裂孔周围的盆底肌肉肌力肌电数据，以获得基于肌力肌电数据的盆底肌功能评估信息；
    所述多模块融合阴道直肠探头用于将获得的盆底基本信息，盆腔组织和神经、血管、肌骨功能评估信息，盆底肌肉肌力肌电数据发送至计算机，实现多模块信号融合, 获得盆底功能整体评估信息；
    所述多模块融合阴道直肠探头为探测指套，将光学成像模块（1）设于探测指套的指尖部位，将超声模块（2）设于探测指套第一指节的指腹部位，将肌肉肌力和肌电检测模块设于整个探测指套表面，并采用透明膜（3）充气启动工作模式。
PCT/CN2023/079183 2022-11-29 2023-03-02 一种多模态数据融合的盆底功能整体评估方法及设备 WO2024113521A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211503654.7 2022-11-29
CN202211503654.7A CN115530881B (zh) 2022-11-29 2022-11-29 一种多模态数据融合的盆底功能整体评估方法及设备

Publications (1)

Publication Number Publication Date
WO2024113521A1 true WO2024113521A1 (zh) 2024-06-06

Family

ID=84721744

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/079183 WO2024113521A1 (zh) 2022-11-29 2023-03-02 一种多模态数据融合的盆底功能整体评估方法及设备

Country Status (2)

Country Link
CN (1) CN115530881B (zh)
WO (1) WO2024113521A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115530881B (zh) * 2022-11-29 2023-03-07 四川大学华西第二医院 一种多模态数据融合的盆底功能整体评估方法及设备

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040092821A1 (en) * 2000-11-24 2004-05-13 Steffen Hering Ultrasonic probe with positioning device for examination devices and operation devices
US20120108918A1 (en) * 2008-09-19 2012-05-03 Physiosonics, Inc. Acoustic Palpation Using Non-Invasive Ultrasound Techniques for Identification of Target Sites and Assessment of Chronic Pain Disorders
US20130158365A1 (en) * 2011-12-16 2013-06-20 The Regents Of The University Of Michigan Digital manometry finger-mountable sensor device
CN106683159A (zh) * 2016-12-22 2017-05-17 深圳开立生物医疗科技股份有限公司 三维盆底超声图像处理方法及系统
CN109512444A (zh) * 2018-12-28 2019-03-26 四川大学华西医院 一种具有盆底肌肉收缩监测及康复训练功能的探头
CN109893146A (zh) * 2019-03-07 2019-06-18 深圳大学 一种女性盆底功能障碍评估方法及其系统
US20190200851A1 (en) * 2017-12-29 2019-07-04 The Regents Of The University Of California Optical biopsy applicators for treatment planning, monitoring, and image-guided therapy
CN111383212A (zh) * 2020-03-06 2020-07-07 深圳度影医疗科技有限公司 一种盆底超声视频图像的分析方法
CN115530881A (zh) * 2022-11-29 2022-12-30 四川大学华西第二医院 一种多模态数据融合的盆底功能整体评估方法及设备

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6862480B2 (en) * 2001-11-29 2005-03-01 Biocontrol Medical Ltd. Pelvic disorder treatment device
US20060100529A1 (en) * 2004-02-02 2006-05-11 Siemens Corporate Research Inc. Combined intra-rectal optical-MR and intra-rectal optical-US device for prostate-, cevix-, rectum imaging diagnostics
WO2014082040A1 (en) * 2012-11-26 2014-05-30 Mayo Foundation For Medical Education And Research Finger cot camera system
CN104027893B (zh) * 2013-03-08 2021-08-31 奥赛拉公司 用于多焦点超声治疗的装置和方法
GB2523991B (en) * 2014-02-19 2018-07-18 Papirov Eduard Device, system and method for ambulatory urodynamic analysis
US11491342B2 (en) * 2015-07-01 2022-11-08 Btl Medical Solutions A.S. Magnetic stimulation methods and devices for therapeutic treatments
US20210015451A1 (en) * 2015-09-08 2021-01-21 Advanced Tactile Imaging, Inc. Methods for vaginal tactile and ultrasound image fusion
US20170065249A1 (en) * 2015-09-08 2017-03-09 Advanced Tactile Imaging Inc. Methods and probes for vaginal tactile and ultrasound imaging
CN105999637B (zh) * 2016-07-05 2018-01-09 山东省千佛山医院 一种形状自适应盆底肌群训练装置
JP2018143403A (ja) * 2017-03-03 2018-09-20 セイコーエプソン株式会社 超音波探触子及び超音波装置
CA3075432C (en) * 2017-09-18 2023-05-23 Derek SHAM Methods and systems for vaginal therapeutic device fitting
CN111012326B (zh) * 2018-10-09 2022-07-05 深圳市理邦精密仪器股份有限公司 盆底校准方法、设备及计算机可读存储介质
US11813446B2 (en) * 2019-04-05 2023-11-14 University Of Louisville Research Foundation, Inc. Methods of for improvement of lower urinary tract function
CN110338790B (zh) * 2019-07-03 2020-10-30 浙江大学 一种用于采集表面肌电和多种生理参数的柔性指套
CN211511798U (zh) * 2019-11-26 2020-09-18 深圳市理邦精密仪器股份有限公司 阴道电极及盆底肌检测仪器
CN215874651U (zh) * 2021-05-07 2022-02-22 武汉维珍医疗有限公司 一种阴道截面测量仪
CN113729740B (zh) * 2021-08-03 2022-10-14 浙江大学 一种基于高密度气囊电极的盆底肌群功能精准诊断分析方法

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040092821A1 (en) * 2000-11-24 2004-05-13 Steffen Hering Ultrasonic probe with positioning device for examination devices and operation devices
US20120108918A1 (en) * 2008-09-19 2012-05-03 Physiosonics, Inc. Acoustic Palpation Using Non-Invasive Ultrasound Techniques for Identification of Target Sites and Assessment of Chronic Pain Disorders
US20130158365A1 (en) * 2011-12-16 2013-06-20 The Regents Of The University Of Michigan Digital manometry finger-mountable sensor device
CN106683159A (zh) * 2016-12-22 2017-05-17 深圳开立生物医疗科技股份有限公司 三维盆底超声图像处理方法及系统
US20190200851A1 (en) * 2017-12-29 2019-07-04 The Regents Of The University Of California Optical biopsy applicators for treatment planning, monitoring, and image-guided therapy
CN109512444A (zh) * 2018-12-28 2019-03-26 四川大学华西医院 一种具有盆底肌肉收缩监测及康复训练功能的探头
CN109893146A (zh) * 2019-03-07 2019-06-18 深圳大学 一种女性盆底功能障碍评估方法及其系统
CN111383212A (zh) * 2020-03-06 2020-07-07 深圳度影医疗科技有限公司 一种盆底超声视频图像的分析方法
CN115530881A (zh) * 2022-11-29 2022-12-30 四川大学华西第二医院 一种多模态数据融合的盆底功能整体评估方法及设备

Also Published As

Publication number Publication date
CN115530881A (zh) 2022-12-30
CN115530881B (zh) 2023-03-07

Similar Documents

Publication Publication Date Title
Frawley et al. An International Continence Society (ICS) report on the terminology for pelvic floor muscle assessment
Carrington et al. Advances in the evaluation of anorectal function
Sultan et al. An International Urogynecological Association (IUGA)/International Continence Society (ICS) joint report on the terminology for female anorectal dysfunction
US8052622B2 (en) Methods for characterizing vaginal tissue elasticity
US8419659B2 (en) Methods for assessment of improvements in pelvic organ conditions after an interventional procedure
Bump et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction
US9861316B2 (en) Methods and probes for vaginal tactile and electromyographic imaging and location-guided female pelvic floor therapy
Griffiths et al. Dynamic testing
US20170065249A1 (en) Methods and probes for vaginal tactile and ultrasound imaging
US20210353195A1 (en) Devices, systems, and methods for monitoring bladder function
Wagenlehner et al. Surgical reconstruction of pelvic floor descent: anatomic and functional aspects
WO2024113521A1 (zh) 一种多模态数据融合的盆底功能整体评估方法及设备
Doumouchtsis et al. An International Continence Society (ICS)/International Urogynecological Association (IUGA) joint report on the terminology for the assessment and management of obstetric pelvic floor disorders
US20200037950A1 (en) Methods for biomechanical mapping of the female pelvic floor
CN117481670A (zh) 一种基于盆底肌电的评估及训练康复系统
US20210015451A1 (en) Methods for vaginal tactile and ultrasound image fusion
CN107896482B (zh) 用于测量内腔变形的诊断探头
CN211985439U (zh) 一种尿道膀胱后角测量仪
Vignoli Urodynamics for Urogynecologists
Martínez Bustelo Effect of perineal and abdominal exercises in the vesical displacement in postpartum women
Constantinou Dynamics of the Female Pelvic Floor
Tabbakha et al. A Customized System to Test Pelvic Floor Muscles Contraction: A Proof of Concept
Bø Visual observation
Thorsen et al. Common Tests for the Pelvic Floor
Nijland Functional ultrasound assessment of external anal sphincter in healthy women and patients with previous obstetric anal sphincter injury