Semi-Supervised Learning for limited medical data using Generative Adversarial Network and Transfer Learning

Deep Learning is progressively becoming popular for computer based automated diagnosis of diseases. Deep Learning algorithms necessitate a large amount of data for training which is hard to acquire for medical problems. Similarly, annotation of medical images can be done with the help of specialized doctors only. This paper presents a semi-supervised learning based model that combines the capabilities of generative adversarial network (GAN) and transfer learning. The proposed model does not demand a large amount of data and can be trained using a small number of images. To evaluate the performance of the model, it is trained and tested on publicly available chest Xray dataset. Better classification accuracy of 94.73% is achieved for normal X-ray images and the ones with pneumonia.

[1]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[2]  Adel Khelifi,et al.  A Comparison of Transfer Learning Performance Versus Health Experts in Disease Diagnosis From Medical Imaging , 2020, IEEE Access.

[3]  Daniel S. Kermany,et al.  Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning , 2018, Cell.

[4]  Yaël Frégier,et al.  Geometric Science of Information: 5th International Conference, GSI 2021, Paris, France, July 21–23, 2021, Proceedings , 2019, GSI.

[5]  Ting Liu,et al.  Recent advances in convolutional neural networks , 2015, Pattern Recognit..

[6]  Natan T. Shaked,et al.  TOP-GAN: Label-Free Cancer Cell Classification Using Deep Learning with a Small Training Set , 2018, ArXiv.

[7]  U. Rajendra Acharya,et al.  Automated detection of diabetic subject using pre-trained 2D-CNN models with frequency spectrum images extracted from heart rate signals , 2019, Comput. Biol. Medicine.

[8]  Joel J. P. C. Rodrigues,et al.  A novel deep learning based framework for the detection and classification of breast cancer using transfer learning , 2019, Pattern Recognit. Lett..

[9]  Bogdan Raducanu,et al.  Transferring GANs: generating images from limited data , 2018, ECCV.

[10]  Lixin Zheng,et al.  A transfer learning method with deep residual network for pediatric pneumonia diagnosis , 2020, Comput. Methods Programs Biomed..

[11]  Muhammad Imran Razzak,et al.  Deep Learning for Medical Image Processing: Overview, Challenges and Future , 2017, ArXiv.

[12]  Burhan Ergen,et al.  A Deep Feature Learning Model for Pneumonia Detection Applying a Combination of mRMR Feature Selection and Machine Learning Models , 2020, IRBM.

[13]  U. Rajendra Acharya,et al.  Application of deep transfer learning for automated brain abnormality classification using MR images , 2019, Cognitive Systems Research.

[14]  Wojciech Zaremba,et al.  Improved Techniques for Training GANs , 2016, NIPS.

[15]  Ioannis D. Apostolopoulos,et al.  Covid-19: automatic detection from X-ray images utilizing transfer learning with convolutional neural networks , 2020, Physical and Engineering Sciences in Medicine.

[16]  Rachna Jain,et al.  Convolutional neural network based Alzheimer’s disease classification from magnetic resonance brain images , 2019, Cognitive Systems Research.

[17]  Arun Ross,et al.  On automated source selection for transfer learning in convolutional neural networks , 2018, Pattern Recognit..

[18]  Kai Zhang,et al.  Deep learning for image-based cancer detection and diagnosis - A survey , 2018, Pattern Recognit..

[19]  Christoph Meinel,et al.  Deep Learning for Medical Image Analysis , 2018, Journal of Pathology Informatics.

[20]  Paul Babyn,et al.  Generative Adversarial Network in Medical Imaging: A Review , 2018, Medical Image Anal..