Back to Search
Start Over
Gastric precancerous diseases classification using CNN with a concise model
- Source :
- PLoS ONE, PLoS ONE, Vol 12, Iss 9, p e0185508 (2017)
- Publication Year :
- 2017
- Publisher :
- Public Library of Science (PLoS), 2017.
-
Abstract
- Gastric precancerous diseases (GPD) may deteriorate into early gastric cancer if misdiagnosed, so it is important to help doctors recognize GPD accurately and quickly. In this paper, we realize the classification of 3-class GPD, namely, polyp, erosion, and ulcer using convolutional neural networks (CNN) with a concise model called the Gastric Precancerous Disease Network (GPDNet). GPDNet introduces fire modules from SqueezeNet to reduce the model size and parameters about 10 times while improving speed for quick classification. To maintain classification accuracy with fewer parameters, we propose an innovative method called iterative reinforced learning (IRL). After training GPDNet from scratch, we apply IRL to fine-tune the parameters whose values are close to 0, and then we take the modified model as a pretrained model for the next training. The result shows that IRL can improve the accuracy about 9% after 6 iterations. The final classification accuracy of our GPDNet was 88.90%, which is promising for clinical GPD recognition.
- Subjects :
- Image Processing
lcsh:Medicine
02 engineering and technology
Pathology and Laboratory Medicine
computer.software_genre
Convolutional neural network
030218 nuclear medicine & medical imaging
Machine Learning
Mathematical and Statistical Techniques
0302 clinical medicine
Medicine and Health Sciences
0202 electrical engineering, electronic engineering, information engineering
Reinforcement learning
Medicine
lcsh:Science
Ulcers
Multidisciplinary
Artificial neural network
Oncology
Engineering and Technology
020201 artificial intelligence & image processing
Algorithms
Research Article
Computer and Information Sciences
Neural Networks
Surgical and Invasive Medical Procedures
Image processing
Research and Analysis Methods
Machine learning
03 medical and health sciences
Signs and Symptoms
Diagnostic Medicine
Artificial Intelligence
Stomach Neoplasms
Support Vector Machines
Gastrointestinal Tumors
Humans
business.industry
Deep learning
lcsh:R
Biology and Life Sciences
Cancers and Neoplasms
Endoscopy
Models, Theoretical
Convolution
Support vector machine
Gastric Cancer
Signal Processing
Lesions
lcsh:Q
Artificial intelligence
business
Mathematical Functions
Precancerous Conditions
computer
Neuroscience
Subjects
Details
- ISSN :
- 19326203
- Volume :
- 12
- Database :
- OpenAIRE
- Journal :
- PLOS ONE
- Accession number :
- edsair.doi.dedup.....97e04f23184365f81c7921086ebbd534