Low-shot machine learning for medical image classification

AI and Deep Learning have seen many exciting real-world applications implemented today. The application focus for this project is on automatic medical image classification. Conventional deep learning requires huge amounts of data to be trained on to achieve high performance values. As such, this pro...

Full description

Saved in:
Bibliographic Details
Main Author: Yip, Chun Mun
Other Authors: Wang Lipo
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2020
Subjects:
Online Access:https://hdl.handle.net/10356/139377
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-139377
record_format dspace
spelling sg-ntu-dr.10356-1393772023-07-07T18:34:59Z Low-shot machine learning for medical image classification Yip, Chun Mun Wang Lipo School of Electrical and Electronic Engineering elpwang@ntu.edu.sg Engineering::Electrical and electronic engineering AI and Deep Learning have seen many exciting real-world applications implemented today. The application focus for this project is on automatic medical image classification. Conventional deep learning requires huge amounts of data to be trained on to achieve high performance values. As such, this project aims to develop strategies and models that can produce satisfactory accuracy in classifying medical images given a very small sample size. The performance will be evaluated on the task of classifying normal retinas against diabetic retinopathy retinas and normal lungs against pneumonia infected lungs. First the effects of low-shot training were studied in greater detail by iteratively training a basic convolution neural network model (CNN) with the sample size scaled down each time. The findings showed deteriorating accuracy performance for both tasks. However, for the same model, one task suggested underfitting and the other, overfitting. Three strategies were explored namely, Generative Adversarial Networks (GAN), Transfer Learning and Model Optimization. GAN has the ability to generate synthetic copies of real images. However, in this project our GAN was proven to be unfeasible. Transfer learning relies on previously trained models that are publicly available to use. Model optimization refers to the process of fine-tuning and adjusting the construct of the model to improve its performance. For this project we found that transfer learning and model optimization were successful in tackling our low-shot training problems. For future work and contribution, we hope the strategies developed in this project will be helpful in overcoming similar problem statements. These strategies have the potential to be viable and remarkable solutions to many deep learning challenges faced today. Bachelor of Engineering (Electrical and Electronic Engineering) 2020-05-19T05:48:20Z 2020-05-19T05:48:20Z 2020 Final Year Project (FYP) https://hdl.handle.net/10356/139377 en A3260-191 application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Electrical and electronic engineering
spellingShingle Engineering::Electrical and electronic engineering
Yip, Chun Mun
Low-shot machine learning for medical image classification
description AI and Deep Learning have seen many exciting real-world applications implemented today. The application focus for this project is on automatic medical image classification. Conventional deep learning requires huge amounts of data to be trained on to achieve high performance values. As such, this project aims to develop strategies and models that can produce satisfactory accuracy in classifying medical images given a very small sample size. The performance will be evaluated on the task of classifying normal retinas against diabetic retinopathy retinas and normal lungs against pneumonia infected lungs. First the effects of low-shot training were studied in greater detail by iteratively training a basic convolution neural network model (CNN) with the sample size scaled down each time. The findings showed deteriorating accuracy performance for both tasks. However, for the same model, one task suggested underfitting and the other, overfitting. Three strategies were explored namely, Generative Adversarial Networks (GAN), Transfer Learning and Model Optimization. GAN has the ability to generate synthetic copies of real images. However, in this project our GAN was proven to be unfeasible. Transfer learning relies on previously trained models that are publicly available to use. Model optimization refers to the process of fine-tuning and adjusting the construct of the model to improve its performance. For this project we found that transfer learning and model optimization were successful in tackling our low-shot training problems. For future work and contribution, we hope the strategies developed in this project will be helpful in overcoming similar problem statements. These strategies have the potential to be viable and remarkable solutions to many deep learning challenges faced today.
author2 Wang Lipo
author_facet Wang Lipo
Yip, Chun Mun
format Final Year Project
author Yip, Chun Mun
author_sort Yip, Chun Mun
title Low-shot machine learning for medical image classification
title_short Low-shot machine learning for medical image classification
title_full Low-shot machine learning for medical image classification
title_fullStr Low-shot machine learning for medical image classification
title_full_unstemmed Low-shot machine learning for medical image classification
title_sort low-shot machine learning for medical image classification
publisher Nanyang Technological University
publishDate 2020
url https://hdl.handle.net/10356/139377
_version_ 1772826646064136192