An active-learning based training-schedule for biomedical image segmentation on deep neural networks
This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2021.
| Autores principales: | , , |
|---|---|
| Otros Autores: | |
| Formato: | Tesis |
| Lenguaje: | en_US |
| Publicado: |
Brac University
2021
|
| Materias: | |
| Acceso en línea: | http://hdl.handle.net/10361/14809 |
| id |
10361-14809 |
|---|---|
| record_format |
dspace |
| spelling |
10361-148092023-01-24T09:03:42Z An active-learning based training-schedule for biomedical image segmentation on deep neural networks Hassan, Mehadi Das, Shemonto Dipu, Shoaib Ahmed Majumdar, Mahbubul Alam Das, Sowmitra Ahmed, Shahnewaz Department of Computer Science and Engineering, Brac University Active learning Deep learning Uncertainty metric Biomedical image segmentation This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2021. Cataloged from PDF version of thesis. Includes bibliographical references (pages 40-43). Biomedical image classification and segmentation are quite important tasks for medical diagnosis. Many Deep Neural Networks (U-Net, V-Net, etc) have been used in recent years to segment biomedical images. For classification of biomedical images or 3D data (X-Ray, CT scan, MRI), ResNet, DenseNet, Xception, Inception, etc. have been in use for automatic disease diagnosis. But all of these networks are trained end-to-end and they do not accumulate anatomical information that is required to interpret similar data in the same way Radiologists do. A new research direction would be to make the network aware of key anatomical locations and their relative positions while generating predictions. We investigated the roles that Active Learning can play in the development and deployment of Deep Learning enabled diagnostic applications and focus on techniques that will retain significant input from a human end-user. In order to practically understand the drawbacks of existing approaches using different networks, we benchmarked the MICCAI BraTS 2019 dataset on different Neural Networks. To overcome the drawbacks of existing approaches of different networks we have incorporated an uncertainty-based Active Learning Training Schedule to segment biomedical images. Through this approach, we have achieved a much better performance than the traditional end-to-end approaches on Deep Neural Networks for biomedical image segmentation. Mehadi Hassan Shemonto Das Shoaib Ahmed Dipu B. Computer Science 2021-07-15T06:01:13Z 2021-07-15T06:01:13Z 2021 2021-01 Thesis ID: 17101177 ID: 17101447 ID: 17101482 http://hdl.handle.net/10361/14809 en_US Brac University theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. 43 Pages application/pdf Brac University |
| institution |
Brac University |
| collection |
Institutional Repository |
| language |
en_US |
| topic |
Active learning Deep learning Uncertainty metric Biomedical image segmentation |
| spellingShingle |
Active learning Deep learning Uncertainty metric Biomedical image segmentation Hassan, Mehadi Das, Shemonto Dipu, Shoaib Ahmed An active-learning based training-schedule for biomedical image segmentation on deep neural networks |
| description |
This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2021. |
| author2 |
Majumdar, Mahbubul Alam |
| author_facet |
Majumdar, Mahbubul Alam Hassan, Mehadi Das, Shemonto Dipu, Shoaib Ahmed |
| format |
Thesis |
| author |
Hassan, Mehadi Das, Shemonto Dipu, Shoaib Ahmed |
| author_sort |
Hassan, Mehadi |
| title |
An active-learning based training-schedule for biomedical image segmentation on deep neural networks |
| title_short |
An active-learning based training-schedule for biomedical image segmentation on deep neural networks |
| title_full |
An active-learning based training-schedule for biomedical image segmentation on deep neural networks |
| title_fullStr |
An active-learning based training-schedule for biomedical image segmentation on deep neural networks |
| title_full_unstemmed |
An active-learning based training-schedule for biomedical image segmentation on deep neural networks |
| title_sort |
active-learning based training-schedule for biomedical image segmentation on deep neural networks |
| publisher |
Brac University |
| publishDate |
2021 |
| url |
http://hdl.handle.net/10361/14809 |
| work_keys_str_mv |
AT hassanmehadi anactivelearningbasedtrainingscheduleforbiomedicalimagesegmentationondeepneuralnetworks AT dasshemonto anactivelearningbasedtrainingscheduleforbiomedicalimagesegmentationondeepneuralnetworks AT dipushoaibahmed anactivelearningbasedtrainingscheduleforbiomedicalimagesegmentationondeepneuralnetworks AT hassanmehadi activelearningbasedtrainingscheduleforbiomedicalimagesegmentationondeepneuralnetworks AT dasshemonto activelearningbasedtrainingscheduleforbiomedicalimagesegmentationondeepneuralnetworks AT dipushoaibahmed activelearningbasedtrainingscheduleforbiomedicalimagesegmentationondeepneuralnetworks |
| _version_ |
1814307419271987200 |