US20190290246A1 - Assisted detection model of breast tumor, assisted detection system thereof, and method for assisted detecting breast tumor - Google Patents
Assisted detection model of breast tumor, assisted detection system thereof, and method for assisted detecting breast tumor Download PDFInfo
- Publication number
- US20190290246A1 US20190290246A1 US16/217,844 US201816217844A US2019290246A1 US 20190290246 A1 US20190290246 A1 US 20190290246A1 US 201816217844 A US201816217844 A US 201816217844A US 2019290246 A1 US2019290246 A1 US 2019290246A1
- Authority
- US
- United States
- Prior art keywords
- breast
- tumor
- image
- breast tumor
- assisted detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 208000026310 Breast neoplasm Diseases 0.000 title claims abstract description 163
- 206010006187 Breast cancer Diseases 0.000 title claims abstract description 159
- 238000001514 detection method Methods 0.000 title claims abstract description 102
- 238000000034 method Methods 0.000 title description 6
- 206010028980 Neoplasm Diseases 0.000 claims abstract description 90
- 238000007781 pre-processing Methods 0.000 claims abstract description 28
- 238000012545 processing Methods 0.000 claims abstract description 5
- 238000002604 ultrasonography Methods 0.000 claims description 110
- 210000000481 breast Anatomy 0.000 claims description 108
- 239000011159 matrix material Substances 0.000 claims description 66
- 238000010606 normalization Methods 0.000 claims description 27
- 238000011176 pooling Methods 0.000 claims description 25
- 201000011510 cancer Diseases 0.000 claims description 16
- 238000013135 deep learning Methods 0.000 claims description 14
- 238000009966 trimming Methods 0.000 claims description 11
- 238000013527 convolutional neural network Methods 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 230000006870 function Effects 0.000 claims description 4
- 238000011282 treatment Methods 0.000 description 9
- 210000004027 cell Anatomy 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000009607 mammography Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 3
- 230000003211 malignant effect Effects 0.000 description 3
- 208000003200 Adenoma Diseases 0.000 description 2
- 238000001574 biopsy Methods 0.000 description 2
- 201000007295 breast benign neoplasm Diseases 0.000 description 2
- 208000031513 cyst Diseases 0.000 description 2
- 210000005075 mammary gland Anatomy 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 210000001519 tissue Anatomy 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000009534 blood test Methods 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 238000002512 chemotherapy Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 229940088597 hormone Drugs 0.000 description 1
- 239000005556 hormone Substances 0.000 description 1
- 210000001165 lymph node Anatomy 0.000 description 1
- 210000001365 lymphatic vessel Anatomy 0.000 description 1
- 230000009245 menopause Effects 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 230000035755 proliferation Effects 0.000 description 1
- 238000002601 radiography Methods 0.000 description 1
- 238000001959 radiotherapy Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 238000001356 surgical procedure Methods 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 230000001225 therapeutic effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- 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
- A61B8/0825—Detecting organic movements or changes, e.g. tumours, cysts, swellings for diagnosis of the breast, e.g. mammography
-
- 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
- A61B8/5207—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of raw data to produce diagnostic data, e.g. for generating an image
-
- 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
- A61B8/5246—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 the same or different imaging techniques, e.g. color Doppler and B-mode
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
-
- 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
- G06T7/0014—Biomedical image inspection using an image reference approach
-
- 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
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
-
- 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
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
- G16H40/63—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for local operation
-
- 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
-
- 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
- A61B8/0833—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures
- A61B8/085—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures for locating body or organic structures, e.g. tumours, calculi, blood vessels, nodules
-
- 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
- A61B8/5223—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data for extracting a diagnostic or physiological parameter from medical diagnostic data
-
- 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
- G06T2207/10132—Ultrasound image
-
- 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
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/30068—Mammography; Breast
-
- 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
- G06T2207/30096—Tumor; Lesion
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Theoretical Computer Science (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Heart & Thoracic Surgery (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Vascular Medicine (AREA)
- Physiology (AREA)
- General Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
Description
- This application claims priority to Taiwan Application Serial Number 107110127, filed Mar. 23, 2018, which is herein incorporated by reference.
- The present disclosure relates to a medical information analysis model, system and method thereof. More particularly, the present disclosure relates to an assisted detection model of breast tumor, an assisted detection system of breast tumor, and a method for assisted detecting breast tumor.
- Breast tumors are formed by abnormal division and proliferation of breast acinar cells or mammary gland cells. Most breast tumors are benign fibrous adenomas, fibrocysts or cysts, and only one tenths breast tumors may be malignant tumors. But if the benign tumor is too large or with clinical symptoms, it still need treatment. The properties of hard blocks appearing in different age groups are not the same. In general, breast tumors that occur before the age of 30 are mostly benign fibrous adenomas or cysts. Breast tumors that occur from 30 to 50 years old are usually fibrocystic or breast cancer. Breast tumors that occur after menopause are mostly breast cancer.
- Breasts are rich in blood vessels, lymphatic vessels, lymph nodes, so breast cancer cells easily spread to other organs. With the increase of the number of the breast cancer patients, the prognosis of breast cancer is related to the three factors of “type of cancer cells”, “time of discovery” and “method of treatment”. But as long as the early detection and early treatment can be as early as possible, the therapeutic effect will be better. Clinical statistics show that the 5-year survival rate of early breast cancer can reach more than 80%, and the cure rate in the first phase is more than 97%. The American medical community promotes breast self-examination and mammography, so that the discovery rate of early breast cancer can be as high as 60%. However, the discovery rate of early breast cancer is only 15-20% in the Chinese people due to the conservative nature of the Chinese people.
- Breast tumor examination and clinical identification of benign and malignant breast tumor include special radiography (mammography), breast ultrasound, blood test and biopsy sampling. Because the mammary glands of Asian women are usually dense, it is necessary to squeeze the breast force strongly in mammography, which is likely to cause discomfort to the patient and the possibility of cancer cells spreading due to squeezing. If the specialist is unable to confirm the tumor status by the result of the mammography or breast ultrasound examination, the living tissue will be sampled in an invasive manner for diagnosis.
- It can be seen that the conventional technology lacks tools with high index, better sensitivity and can be used as clinically assisted tool for grouping breast tumor types. Therefore, it is necessary to improve the conventional techniques to improve the accuracy of the diagnosis of breast tumor types by using breast ultrasound image, reduce the discomfort caused by other invasive examinations, and reduce the spread of cancer cells that may be caused by the examination.
- According to one aspect of the present disclosure, an assisted detection model of breast tumor includes following establishing steps. A reference database is obtained, wherein the reference database includes a plurality of reference breast ultrasound images. An image preprocessing step is performed, wherein the image preprocessing step is for dividing an image matrix value of each of the reference breast ultrasound images by a first normalization factor to obtain a reference value interval, and the reference value interval is between 0 and 1. A feature selecting step is performed, wherein the feature selecting step is for selecting a feature matrix according to the reference database by using an autoencoder module, and the autoencoder module includes an encoder and a decoder. The encoder is for compressing the reference value interval to obtain the feature matrix, wherein the encoder includes a plurality of convolution layers and a plurality of pooling layers. The decoder is for reducing the feature matrix and comparing the reduced feature matrix with the reference breast ultrasound images to confirm that the feature matrix includes key information in each of the reference breast ultrasound images, wherein the decoder includes a plurality of convolution layers and a plurality of upsampling layers. A classifying step is performed, wherein the classifying step is for achieving a convergence of the feature matrix by using a deep learning classifier to obtain the assisted detection model of breast tumor. The assisted detection model of breast tumor is used to determine a breast tumor type of a subject and predict a probability of a tumor location of the subject.
- According to another aspect of the present disclosure, an assisted detection method of breast tumor includes following steps. The assisted detection model of breast tumor of the aforementioned aspect is provided. A target breast ultrasound image of a subject is provided. An image matrix value of the target breast ultrasound image is divided by a second normalization factor to obtain a target value interval. The assisted detection model of breast tumor is used to analyze the target value interval to determine a breast tumor type of the subject and predict a probability of a tumor location of the subject.
- According to still another aspect of the present disclosure, an assisted detection system of breast tumor includes an image capturing unit and a non-transitory machine readable medium. The image capturing unit is for obtaining a target breast ultrasound image of a subject. The non-transitory machine readable medium storing a program which, when executed by at least one processing unit, determines a breast tumor type of the subject and predicts a probability of a tumor location of the subject. The program includes a reference database obtaining module, a first image preprocessing module, an autoencoder module, a classifying module, a second image preprocessing module and a comparing module. The reference database obtaining module is for obtaining a reference database, wherein the reference database includes a plurality of reference breast ultrasound images. The first image preprocessing module is for normalizing an image matrix value of each of the reference breast ultrasound images to obtain a reference value interval, wherein the reference value interval is between 0 and 1. The autoencoder module is for selecting a feature matrix according to the reference database, and the autoencoder module includes an encoder and a decoder. The encoder is for compressing the reference value interval to obtain the feature matrix, wherein the encoder includes a plurality of convolution layers and a plurality of pooling layers. The decoder is for reducing the feature matrix and comparing the reduced feature matrix with the reference breast ultrasound images to confirm that the feature matrix includes key information in each of the reference breast ultrasound images, wherein the decoder includes a plurality of convolution layers and a plurality of upsampling layers. The classifying module is for achieving a convergence of the feature matrix by using a deep learning classifier to obtain an assisted detection model of breast tumor. The second image preprocessing module is for normalizing an image matrix value of each of the target breast ultrasound image to obtain a target value interval, wherein the target value interval is between 0 and 1. The comparing module is for analyzing the target value interval by the assisted detection model of breast tumor to determine the breast tumor type of the subject and predict the probability of the tumor location of the subject.
- The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by Office upon request and payment of the necessary fee. The present disclosure can be more fully understood by reading the following detailed description of the embodiment, with reference made to the accompanying drawings as follows:
-
FIG. 1 is a flowchart of establishing steps of an assisted detection model of breast tumor according to one embodiment of the present disclosure. -
FIG. 2 is a flowchart of an assisted detection method of breast tumor according to another embodiment of the present disclosure. -
FIG. 3 is a block diagram of an assisted detection system of breast tumor according to still another embodiment of the present disclosure. -
FIG. 4 is a structural diagram of an assisted detection model of breast tumor according to yet another embodiment of the present disclosure. -
FIG. 5 shows a receiver operating characteristic curve (ROC) diagram of the assisted detection model of breast tumor used to determine a breast tumor type of a subject. -
FIGS. 6A, 6B and 6C show analysis result charts of the assisted detection model of breast tumor used to predict a probability of a tumor location of a subject. - Reference will now be made in detail to the present embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings.
- Please refer to
FIG. 1 , which is a flowchart of establishing steps of an assisted detection model ofbreast tumor 100 according to one embodiment of the present disclosure. The establishing steps of the assisted detection model ofbreast tumor 100 includesStep 110,Step 120,Step 130 andStep 140. The established assisted detection model of breast tumor can be used to determine a breast tumor type of a subject and predict a probability of a tumor location of the subject. The breast tumor type can be no tumor, benign tumor and malignant tumor. - In
Step 110, a reference database is obtained, wherein the reference database includes a plurality of reference breast ultrasound images. - In
Step 120, an image preprocessing step is performed, wherein the image preprocessing step is for dividing an image matrix value of each of the reference breast ultrasound images by a first normalization factor to obtain a reference value interval, and the reference value interval is between 0 and 1. The first normalization factor can be 255. The image preprocessing step can further include trimming the reference breast ultrasound images and resetting the image size of the trimmed reference breast ultrasound images. In detail, in image preprocessing step, each of the reference breast ultrasound images is trimmed to remove the text mark on the periphery of the reference breast ultrasound images first. The image matrix value of each of the reference breast ultrasound images is then divided by the first normalization factor of 255 to obtain the reference value interval between 0 and 1. Finally, the image size of each of the reference ultrasound images after trimming is reset to 128 pixels×128 pixels. - In
Step 130, a feature selecting step is performed, wherein the feature selecting step is for selecting a feature matrix according to the reference database by using an autoencoder module, and the autoencoder module includes an encoder and a decoder. The encoder is for compressing the reference value interval to obtain the feature matrix, wherein the encoder includes a plurality of convolution layers and a plurality of pooling layers. The decoder is for reducing the feature matrix and comparing the reduced feature matrix with the reference breast ultrasound images to confirm that the feature matrix includes key information in each of the reference breast ultrasound images, wherein the decoder includes a plurality of convolution layers and a plurality of upsampling layers. A pooling function of the pooling layers can be a max pooling. - In
Step 140, a classifying step is performed, wherein the classifying step is for achieving a convergence of the feature matrix by using a deep learning classifier to obtain the assisted detection model of breast tumor. The assisted detection model of breast tumor is used to determine the breast tumor type of the subject and predict the probability of the tumor location of the subject through breast ultrasound image. The breast tumor type can be no tumor, benign tumor and malignant tumor. The deep learning classifier can be a deep neural network (DNN), a convolutional neural network (CNN) or a deep belief network (DBN). Preferably, the deep learning classifier can be the convolutional neural network. - Please refer to
FIG. 2 , which is a flowchart of an assisted detection method ofbreast tumor 200 according to another embodiment of the present disclosure. The assisted detection method ofbreast tumor 200 includesStep 210,Step 220,Step 230 andStep 240. - In
Step 210, the assisted detection model of breast tumor is provided, wherein the assisted detection model of breast tumor is established by theaforementioned Steps 110 to 140. - In
Step 220, a target breast ultrasound image of a subject is provided. - In
Step 230, an image matrix value of the target breast ultrasound image is divided by a second normalization factor to obtain a target value interval. The second normalization factor can be 255. In detail, the target breast ultrasound image is trimmed to remove the text mark on the periphery of the target breast ultrasound image first. The image matrix value of the target breast ultrasound image is then divided by the second normalization factor of 255 to obtain the target value interval between 0 and 1. Finally, the image size of the target ultrasound image after trimming is reset to 128 pixels×128 pixels. - In
Step 240, the assisted detection model of breast tumor is used to analyze the target value interval to determine the breast tumor type of the subject and predict the probability of the tumor location of the subject. The breast tumor type can be no tumor, benign tumor and malignant tumor. - Please refer to
FIG. 3 , which is a block diagram of an assisted detection system ofbreast tumor 300 according to still another embodiment of the present disclosure. The assisted detection system ofbreast tumor 300 includes animage capturing unit 400 and a non-transitory machinereadable medium 500. The assisted detection system ofbreast tumor 300 can be used to determine the breast tumor type of the subject and predict the probability of the tumor location of the subject, wherein the breast tumor type can be no tumor, benign tumor and malignant tumor. - The
image capturing unit 400 is for obtaining the target breast ultrasound image of the subject and the reference breast ultrasound images. Theimage capturing unit 400 can be a breast ultrasound image capturing device, which can detect high-density breasts according to the physical characteristics of the tissue by using the returned sound waves. Preferably, theimage capturing unit 400 can be a handheld ultrasonic scanner or an automated breast ultrasound system (ABUS). - The non-transitory machine
readable medium 500 storing a program which, when executed by at least one processing unit, determines the breast tumor type of the subject and predicts the probability of the tumor location of the subject. The program includes a referencedatabase obtaining module 510, a firstimage preprocessing module 520, anautoencoder module 530, a classifyingmodule 540, a secondimage preprocessing module 550 and a comparingmodule 560. - The reference
database obtaining module 510 is for obtaining a reference database, wherein the reference database includes a plurality of reference breast ultrasound images. - The first
image preprocessing module 520 is for normalizing an image matrix value of each of the reference breast ultrasound images to obtain a reference value interval, wherein the reference value interval is between 0 and 1. The firstimage preprocessing module 520 can includes sets of instructions for trimming the reference breast ultrasound images, dividing the image matrix value of each of the reference breast ultrasound images by the first normalization factor to obtain the reference value interval, and resetting the image size of the trimmed reference breast ultrasound images. The first normalization factor can be 255. In detail, each of the reference breast ultrasound images is trimmed by the firstimage preprocessing module 520 to remove the text mark on the periphery of the reference breast ultrasound images first. The image matrix value of each of the reference breast ultrasound images is then divided by the first normalization factor of 255 to obtain the reference value interval between 0 and 1. Finally, the image size of each of the reference ultrasound images after trimming is reset to 128 pixels×128 pixels. - The
autoencoder module 530 is for selecting a feature matrix according to the reference database, and theautoencoder module 530 includes anencoder 532 and adecoder 534. Theencoder 532 is for compressing the reference value interval to obtain the feature matrix, wherein theencoder 532 includes a plurality of convolution layers and a plurality of pooling layers. Thedecoder 534 is for reducing the feature matrix and comparing the reduced feature matrix with the reference breast ultrasound images to confirm that the feature matrix includes key information in each of the reference breast ultrasound images, wherein thedecoder 534 includes a plurality of convolution layers and a plurality of upsampling layers. - The classifying
module 540 is for achieving a convergence of the feature matrix by using a deep learning classifier to obtain an assisted detection model of breast tumor. The deep learning classifier can be the deep neural network, the convolutional neural network or the deep belief network. Preferably, the deep learning classifier can be the convolutional neural network. - The second
image preprocessing module 550 is for normalizing an image matrix value of each of the target breast ultrasound image to obtain the target value interval, wherein the target value interval is between 0 and 1. The secondimage preprocessing module 550 can includes sets of instructions for trimming the target breast ultrasound image, dividing the image matrix value of the target breast ultrasound image by the second normalization factor to obtain the target value interval, and resetting the image size of the trimmed target breast ultrasound image. The second normalization factor can be 255. In detail, the target breast ultrasound image is trimmed by the secondimage preprocessing module 550 to remove the text mark on the periphery of the target breast ultrasound image first. The image matrix value of the target breast ultrasound image is then divided by the second normalization factor of 255 to obtain the target value interval between 0 and 1. Finally, the image size of the target ultrasound image after trimming is reset to 128 pixels×128 pixels. - The comparing
module 560 is for analyzing the target value interval by the assisted detection model of breast tumor to determine the breast tumor type of the subject and predict the probability of the tumor location of the subject. - The reference database used in the present disclosure is the retrospective breast ultrasound image data collected by the China Medical University Hospital. This clinical trial program is approved by China Medical University & Hospital Research Ethics Committee, which is numbered as CMUH106-REC1-087. A total of 330 reference subjects include 88 reference subjects with non-tumor, 148 reference subjects with benign breast tumors, and 94 reference subjects with malignant breast tumors. Biopsy results of aforementioned reference subjects are the further reference to determine whether the tumor is benign or malignant.
- In this example, the assisted detection model of breast tumor needs to be established first. Please refer to
FIG. 4 , which is a structural diagram of an assisted detection model of breast tumor according to yet another embodiment of the present disclosure. - First, the reference database is obtained, and the reference database includes a plurality of reference
breast ultrasound images 610. The image processing is performed by using the first image preprocessing module (its reference numeral is omitted) with the referencebreast ultrasound images 610, which normalizes different types of original reference breast ultrasound images. In the normalization process, each of the reference breast ultrasound images is trimmed to remove the text mark on the periphery of the reference breast ultrasound images first. The image matrix value of each of the reference breast ultrasound images is then divided by the first normalization factor of 255 to obtain the reference value interval between 0 and 1. Finally, the image size of each of the reference ultrasound images after trimming is reset to 128 pixels×128 pixels to complete the normalization process. - The obtained reference value interval is input into an autoencoder module (its reference numeral is omitted), wherein the autoencoder module includes an
encoder 620 and adecoder 630. Theencoder 620 is for compressing the reference value interval to obtain the feature matrix, wherein theencoder 620 includes a plurality of convolution layers and a plurality of pooling layers. In detail, the different features of the input are extracted by two convolution layers convolution operation (Conv2D) first. The input reference ultrasonic image is divided into several rectangular regions by one pooling layer with a pooling function of max pooling (MaxPooling2D), and the maximum value of each sub-region is outputted. The two convolution layers convolution operation and the maximum value outputted by one pooling layer are repeated twice to obtain the feature matrix. - The obtained feature matrix is reduced by the
decoder 630 and then compared with the reference breast ultrasound images to confirm that the feature matrix contains key information in each of the reference breast ultrasound images. In addition, the obtained feature matrix is trained to achieve convergence using adeep learning classifier 650 to obtain the assisted detection model of breast tumor. - The
decoder 630 includes a plurality of convolution layers and a plurality of upsampling layers. In detail, after extracting different features of the input by the two convolution layers convolution operation (Conv2D), the sampling frequency is increased by one upsampling layer (UpSampling2D). After repeating the two convolution layers convolution operation and one up-sampling layer for increasing the sampling frequency twice, the reference breast ultrasound reducedimage 640 is obtained by one convolution layer convolution operation. The reference breast ultrasound reducedimage 640 is the image obtained by the referencebreast ultrasound image 610 processed by the autoencoder module. Comparing the referencebreast ultrasound image 610 and the reference breast ultrasound reducedimage 640, the reference breast ultrasound reducedimage 640 includes all key information of the referencebreast ultrasound images 610. - The
deep learning classifier 650 is trained by using the convolutional neural network. In detail, after inputting the selected feature matrix, the training flow for convolution layer of the convolution operation and the maximum output by the pooling layer in the convolutional neural network are as follows. The feature matrix extracted by theencoder 620 is respectively subjected to the operations including 1×1 convolution layer (Cony), 1×1 and 3×3 convolution layers (Cony), 1×1 and 5×5 convolution layers (Cony), and 3×3 pooling layer (MaxPooling) and 1×1 convolution layer (Cony), and the values of the calculated feature matrices in each group are deeply stacked (DepthConcat). Then, the operations including 1×1 convolution layer, 1×1 and 3×3 convolution layers, 1×1 and 5×5 convolution layers, 3×3 pooling layer and 1×1 convolution layer operation are repeated, and the values of the calculated feature matrices in each group are deeply stacked. Then the global average pooling (Global Average Pooling 2D) is performed to obtain the trained assisted detection model of breast tumor. The obtained assisted detection model of breast tumor can be used to determine the breast tumor type of the subject and to predict the probability of the tumor location of the subject. - In this example, the established assisted detection model of breast tumor is used to determine the breast tumor type of the subject including steps as follows. The established assisted detection model of breast tumor is provided. The target breast ultrasound image of the subject is provided. The image matrix value of the target breast ultrasound image is divided by the second normalization factor of 255 to obtain the target value interval. The obtained target value interval is between 0 and 1. The assisted detection model of breast tumor is used to analyze the target value interval to determine whether the breast tumor type of the subject is no tumor, benign tumor or malignant tumor. The criterion for the determination is to compare the similarity between the data of the assisted detection model of breast tumor with tumor type classification data of the reference database, and a probability value for each target breast ultrasound image is given to determine whether the breast tumor type of the subject is no tumor, benign tumor or malignant tumor.
- Please refer to
FIG. 5 , which shows a receiver operating characteristic curve (ROC) diagram of the assisted detection model of breast tumor used to determine the breast tumor type of the subject. InFIG. 5 , when the assisted detection model of breast tumor is used to determine the breast tumor type of the subject, the area under the macro-average curve and the area under the micro-average curve are both 0.98. When the breast tumor type is no tumor, the AUC (area under the curve) is 1. When the breast tumor type is benign tumor, the AUC is 0.96. When the breast tumor type is malignant tumor, the AUC is 0.97. The results indicate that the assisted detection model of breast tumor, the assisted detection method of breast tumor and the assisted detection system of breast tumor of the present disclosure can be used to accurately determine the breast tumor type of the subject by the breast ultrasound image. - In this example, the established assisted detection model of breast tumor is used to predict the probability of the tumor location of the subject including steps as follows. The established assisted detection model of breast tumor is provided. The target breast ultrasound image of the subject is provided. The image matrix value of the target breast ultrasound image is divided by the second normalization factor of 255 to obtain the target value interval. The obtained target value interval is between 0 and 1. The assisted detection model of breast tumor is used to analyze the target value interval to predict the probability of the tumor location of the subject. The criterion for the determination is to compare the similarity between the data of the assisted detection model of breast tumor with the tumor location information of the reference database, and a probability distribution heat map for each target breast ultrasound image is given to predict the probability of the tumor location of the subject.
- Please refer to
FIGS. 6A, 6B, and 6C , which show analysis result charts of the assisted detection model of breast tumor used to predict the probability of the tumor location of the subject. A targetbreast ultrasound image 710, a tumor position circledimage 720, a tumorposition prediction image 730, a merged image of the tumor position circled image and the tumorposition prediction image 740, and a merged image of the target breast ultrasound image, the tumor position circled image and the tumorposition prediction image 750 are included inFIGS. 6A to 6C . The targetbreast ultrasound image 710 is the original target ultrasound image of the subject. The tumor location circledimage 720 is obtained by the physician circling the tumor position appearing in the targetbreast ultrasound image 710. The tumorposition prediction image 730 is a probability distribution heat map of the tumor position generated after analyzing the targetbreast ultrasound image 710 via the assisted detection model of breast tumor of the present disclosure, wherein the different colors indicate the different probabilities of the tumor location, please refer to the color correspondence table next to the tumorposition prediction image 730. - In
FIG. 6A , there is no tumor in the targetbreast ultrasound image 710, so there is no circled image block in the tumor position circledimage 720. In the tumorposition prediction image 730 generated after analysis by the assisted detection model of breast tumor of the present disclosure, the probability of tumor position is also not shown. InFIG. 6B , there is a tumor in the targetbreast ultrasound image 710 at the upper left position, so there is a breasttumor image block 701 in the tumor position circledimage 720. The tumorposition prediction image 730 generated after analysis by the assisted detection model of breast tumor of the present disclosure shows different probabilities of the tumor position. In the merged image of the tumor position circled image and the tumorposition prediction image 740, the area with the highest incidence of the tumor position in the tumorposition prediction image 730 has high degree of overlap with the breasttumor image block 701 in the tumor position circledimage 720. InFIG. 6C , there is a tumor in the upper middle position of the targetbreast ultrasound image 710, so there is a breasttumor image block 701 in the tumor position circledimage 720. The tumorposition prediction image 730 generated after analysis by the assisted detection model of breast tumor of the present disclosure shows different probabilities of the tumor position. In the merged image of the tumor position circled image and the tumorposition prediction image 740, the area with the highest incidence of the tumor position in the tumorposition prediction image 730 has high degree of overlap with the breasttumor image block 701 in the tumor position circledimage 720. The results indicate that the assisted detection model of breast tumor, the assisted detection method of breast tumor and the assisted detection system of breast tumor of the present disclosure can be used to predict the probability of the tumor location of the subject as an assisted tool for breast tumor detection. - To sum up, the assisted detection model of breast tumor, the assisted detection method of breast tumor and the assisted detection system of breast tumor are provided in the present disclosure. The assisted detection model of breast tumor is based on autoencoder deep neural network, which can effectively improve the sensitivity and the specificity in the identification of the benign tumor and the malignant tumor by the ultrasound image. The assisted detection model of breast tumor of the present disclosure can be used as the assisted tool for improving the accuracy of the diagnosis of breast tumor types and providing a second opinion to the specialist to reduce the patient's discomfort caused by examinations, and reduce the spread of cancer cells that may be caused by the examination. Therefore, the assisted detection model of breast tumor, the assisted detection method of breast tumor and the assisted detection system of breast tumor of the present disclosure can perform automatic and rapid data analysis by using an individual breast ultrasound image in a non-invasive detection manner, and can assist medical personnel to diagnose early for improving the discovery rate of early breast cancer.
- In addition, the assisted detection model of breast tumor of the present disclosure can automatically mark the location of the breast tumor, provide the tumor location information to doctor, and assist the physician in formulating the treatments of tumor. Therefore, the physician can formulate the treatments of the patient according to the benign and malignant of the breast tumor and location of the breast tumor by using the assisted detection method of breast tumor and the assisted detection system of breast tumor of the present disclosure. Further treatments may include surgical treatment, radiation therapy, hormone control treatment, chemotherapy, and target treatment. Therefore, the physician can formulate a subsequent treatment of the patient according to the benign and malignant of the breast tumor and location of the breast tumor by using the assisted detection model of breast tumor, the assisted detection method of breast tumor and the assisted detection system of breast tumor of the present disclosure.
- Although the present disclosure has been described in considerable detail with reference to certain embodiments thereof, other embodiments are possible. Therefore, the spirit and scope of the appended claims should not be limited to the description of the embodiments contained herein.
- It will be apparent to those skilled in the art that various modifications and variations can be made to the structure of the present disclosure without departing from the scope or spirit of the disclosure. In view of the foregoing, it is intended that the present disclosure cover modifications and variations of this disclosure provided they fall within the scope of the following claims.
Claims (17)
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW107110127 | 2018-03-23 | ||
TW107110127A TWI667996B (en) | 2018-03-23 | 2018-03-23 | Assisted detection model of breast tumor, and assisted detection system of breast tumor |
TW107110127A | 2018-03-23 |
Publications (2)
Publication Number | Publication Date |
---|---|
US10420535B1 US10420535B1 (en) | 2019-09-24 |
US20190290246A1 true US20190290246A1 (en) | 2019-09-26 |
Family
ID=67983080
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US16/217,844 Expired - Fee Related US10420535B1 (en) | 2018-03-23 | 2018-12-12 | Assisted detection model of breast tumor, assisted detection system thereof, and method for assisted detecting breast tumor |
Country Status (2)
Country | Link |
---|---|
US (1) | US10420535B1 (en) |
TW (1) | TWI667996B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021250986A1 (en) * | 2020-06-10 | 2021-12-16 | 日立造船株式会社 | Inspection device, inspection method, and inspection program |
WO2022042201A1 (en) * | 2020-08-31 | 2022-03-03 | Ping An Technology (Shenzhen) Co., Ltd. | Method and apparatus for mammographic multi-view mass identification |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110728674B (en) * | 2019-10-21 | 2022-04-05 | 清华大学 | Image processing method and device, electronic equipment and computer readable storage medium |
CN110827345B (en) * | 2019-10-31 | 2020-12-11 | 推想医疗科技股份有限公司 | Cardiothoracic ratio determining method, cardiothoracic ratio determining device, cardiothoracic ratio determining equipment, storage medium and computer equipment |
CN110827296B (en) * | 2019-11-01 | 2023-06-23 | 南京信息工程大学 | Mammary X-ray image analysis method of multi-target integrated deep neural network |
TWI743693B (en) * | 2020-02-27 | 2021-10-21 | 國立陽明交通大學 | Benign tumor development trend assessment system, server computing device thereof and computer readable storage medium |
CN111556017B (en) * | 2020-03-25 | 2021-07-27 | 中国科学院信息工程研究所 | Network intrusion detection method based on self-coding machine and electronic device |
CN113034436B (en) * | 2021-02-26 | 2023-07-14 | 中山大学孙逸仙纪念医院 | Breast cancer molecular transformation prediction device based on breast MR image histology |
CN113053523A (en) * | 2021-04-23 | 2021-06-29 | 广州易睿智影科技有限公司 | Continuous self-learning multi-model fusion ultrasonic breast tumor precise identification system |
CN114098819A (en) * | 2021-12-07 | 2022-03-01 | 四川大学华西医院 | Multifunctional detection and massage nursing system and nursing method for breasts |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW319763B (en) * | 1995-02-01 | 1997-11-11 | Epix Medical Inc | |
JPH11502357A (en) * | 1995-02-10 | 1999-02-23 | カーディアク・マリナーズ・インコーポレイテッド | Scanning beam type X-ray imaging system |
JP6115643B2 (en) * | 2013-08-21 | 2017-04-19 | 株式会社島津製作所 | Image processing method |
JP6291813B2 (en) * | 2013-11-29 | 2018-03-14 | コニカミノルタ株式会社 | Medical image system and program |
US9538974B2 (en) * | 2014-11-20 | 2017-01-10 | General Electric Company | Methods and systems for correcting table deflection |
US11832969B2 (en) * | 2016-12-22 | 2023-12-05 | The Johns Hopkins University | Machine learning approach to beamforming |
US11302003B2 (en) * | 2017-10-26 | 2022-04-12 | Wisconsin Alumni Research Foundation | Deep learning based data-driven approach for attenuation correction of pet data |
-
2018
- 2018-03-23 TW TW107110127A patent/TWI667996B/en active
- 2018-12-12 US US16/217,844 patent/US10420535B1/en not_active Expired - Fee Related
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021250986A1 (en) * | 2020-06-10 | 2021-12-16 | 日立造船株式会社 | Inspection device, inspection method, and inspection program |
JP2021196205A (en) * | 2020-06-10 | 2021-12-27 | 日立造船株式会社 | Inspection device, inspection method, and inspection program |
JP7385529B2 (en) | 2020-06-10 | 2023-11-22 | 日立造船株式会社 | Inspection equipment, inspection methods, and inspection programs |
WO2022042201A1 (en) * | 2020-08-31 | 2022-03-03 | Ping An Technology (Shenzhen) Co., Ltd. | Method and apparatus for mammographic multi-view mass identification |
Also Published As
Publication number | Publication date |
---|---|
TWI667996B (en) | 2019-08-11 |
TW201940124A (en) | 2019-10-16 |
US10420535B1 (en) | 2019-09-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10420535B1 (en) | Assisted detection model of breast tumor, assisted detection system thereof, and method for assisted detecting breast tumor | |
KR101857624B1 (en) | Medical diagnosis method applied clinical information and apparatus using the same | |
Hakim et al. | Thermal imaging-an emerging modality for breast cancer detection: a comprehensive review | |
US11937973B2 (en) | Systems and media for automatically diagnosing thyroid nodules | |
WO2019102829A1 (en) | Image analysis method, image analysis device, image analysis system, image analysis program, and storage medium | |
KR102321487B1 (en) | artificial intelligence-based diagnostic device for shapes and characteristics of detected tumors | |
CN112071418B (en) | Gastric cancer peritoneal metastasis prediction system and method based on enhanced CT image histology | |
Nautiyal et al. | Different approaches of ann for detection of cancer | |
CN110276740A (en) | Mammary tumor assists detection model, its system and mammary tumor aided detection method | |
WO2023198166A1 (en) | Image detection method, system and device, and storage medium | |
Vocaturo et al. | Artificial intelligence approaches on ultrasound for breast cancer diagnosis | |
CN112002407A (en) | Breast cancer diagnosis device and method based on ultrasonic video | |
CN103815926B (en) | Breast cancer detection method and apparatus | |
Kahaki et al. | Weakly supervised deep learning for predicting the response to hormonal treatment of women with atypical endometrial hyperplasia: a feasibility study | |
CN113393938B (en) | Breast cancer risk prediction system fusing image and clinical characteristic information | |
Yongping et al. | Evaluation of the quadri-planes method in computer-aided diagnosis of breast lesions by ultrasonography: prospective single-center study | |
Ahmed et al. | Detection of uterine fibroids in medical images using deep neural networks | |
Zhang et al. | Analysis of the application value of ultrasound imaging diagnosis in the clinical staging of thyroid cancer | |
Aravinda et al. | A demystifying convolutional neural networks using Grad-CAM for prediction of coronavirus disease (COVID-19) on X-ray images | |
Zhang et al. | A Weakly Supervised-Guided Soft Attention Network for Classification of Intracranial Hemorrhage | |
Santhosh et al. | Deep Learning Techniques for Brain Tumor Diagnosis: A Review | |
Xiong et al. | Deep Ensemble Learning Network for Kidney Lesion Detection | |
Gowri et al. | An improved classification of MR images for cervical cancer using convolutional neural networks | |
KR102620046B1 (en) | Method and system for breast ultrasonic image diagnosis using weakly-supervised deep learning artificial intelligence | |
Cañada et al. | Assistance to lung cancer detection on histological images using Convolutional Neural Networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
FEPP | Fee payment procedure |
Free format text: ENTITY STATUS SET TO UNDISCOUNTED (ORIGINAL EVENT CODE: BIG.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
FEPP | Fee payment procedure |
Free format text: ENTITY STATUS SET TO SMALL (ORIGINAL EVENT CODE: SMAL); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
AS | Assignment |
Owner name: CHINA MEDICAL UNIVERSITY HOSPITAL, TAIWAN Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:HUANG, TZUNG-CHI;LIAO, KEN YING-KAI;YU, JIAXIN;AND OTHERS;REEL/FRAME:048319/0168 Effective date: 20181203 |
|
FEPP | Fee payment procedure |
Free format text: ENTITY STATUS SET TO UNDISCOUNTED (ORIGINAL EVENT CODE: BIG.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
FEPP | Fee payment procedure |
Free format text: MAINTENANCE FEE REMINDER MAILED (ORIGINAL EVENT CODE: REM.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
LAPS | Lapse for failure to pay maintenance fees |
Free format text: PATENT EXPIRED FOR FAILURE TO PAY MAINTENANCE FEES (ORIGINAL EVENT CODE: EXP.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
STCH | Information on status: patent discontinuation |
Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362 |
|
FP | Lapsed due to failure to pay maintenance fee |
Effective date: 20230924 |