Method and system for medical image interpretation
AssigneeAcer Incorporated
InventorsMeng-Che Cheng, Ming-Tzuo Yin
Life of this patent
Estimating the term…
Abstract
A method and a system for medical image interpretation are provided. A medical image is provided to a convolutional neural network model. The convolutional neural network model includes a feature extraction part, a first classifier, and N second classifiers. N feature maps are generated by using the last layer of the feature extraction part of the convolutional neural network model. N symptom interpretation results of N symptoms of a disease are obtained based on the N feature maps through the N second classifiers. A disease interpretation result of the disease is obtained based on the N feature maps through the first classifier.
Claims20
A method for medical image interpretation, adapted for a computer device, the method comprising:
- providing a medical image to a convolutional neural network model, wherein the convolutional neural network model comprises a feature extraction part, a first classifier, and N second classifiers, wherein N is a positive integer;
- generating N feature maps by using the feature extraction part of the convolutional neural network model;
- obtaining N symptom interpretation results of N symptoms of a disease based on the N feature maps through the N second classifiers;
- and
- obtaining a disease interpretation result of the disease based on the N feature maps through the first classifier,
- wherein one of the N second classifiers outputs one of the N symptom interpretation results only according to part of the N feature maps, and another one of the N second classifiers outputs another of the N symptom interpretation results only according to another part of the N feature maps.
A system for medical image interpretation, comprising:
- a storage device;
- and
- a processor, coupled to the storage device and configured to:
- provide a medical image to a convolutional neural network model, wherein the convolutional neural network model comprises a feature extraction part, a first classifier, and N second classifiers, wherein N is a positive integer;
- generate N feature maps by using the feature extraction part of the convolutional neural network model;
- obtain N symptom interpretation results of N symptoms of a disease based on the N feature maps through the N second classifiers;
- and
- obtain a disease interpretation result of the disease based on the N feature maps through the first classifier,
- wherein one of the N second classifiers outputs one of the N symptom interpretation results only according to part of the N feature maps, and another one of the N second classifiers outputs another of the N symptom interpretation results only according to another part of the N feature maps.
Dependencies are inferred from the claim text.