Nourhan Bayasi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Nourhan Bayasi has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Ghassan Hamarneh
Simon Fraser University
8 shared publications - Rafeef Garbi
University of British Columbia
7 shared publications - Siyi Du
Imperial College London
4 shared publications Jamil Fayyad
American University of Sharjah
3 shared publications- Alceu Bissoto
University of Bern
2 shared publications Rafeef Garbi
University of British Columbia, Vancouver, BC, Canada
2 shared publications
Latest Publications
- Open Access
- 18 August 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 106, 103764
BiasPruner: Mitigating bias transfer in continual learning for fair medical image analysis
- 18 August 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 106, 103764
Open AccessBiasPruner: Mitigating bias transfer in continual learning for fair medical image analysis
- 3 October 2024
- Published by Springer Nature
- p. 90-101
BiasPruner: Debiased Continual Learning for Medical Image Classification
- 3 October 2024
- Published by Springer Nature
- p. 90-101
BiasPruner: Debiased Continual Learning for Medical Image Classification
- Open Access
BiasPruner: Debiased Continual Learning for Medical Image Classification
Continual Learning (CL) is crucial for enabling networks to dynamically adapt as they learn new tasks sequentially, accommodating new data and classes without catastrophic forgetting. Diverging from conventional perspectives on CL, our paper introduces a new perspective wherein forgetting could actually benefit the sequential learning paradigm. Specifically, we present BiasPruner, a CL framework that intentionally forgets spurious correlations in the training data that could lead to shortcut learning. Utilizing a new bias score that measures the contribution of each unit in the network to learning spurious features, BiasPruner prunes those units with the highest bias scores to form a debiased subnetwork preserved for a given task. As BiasPruner learns a new task, it constructs a new debiased subnetwork, potentially incorporating units from previous subnetworks, which improves adaptation and performance on the new task. During inference, BiasPruner employs a simple task-agnostic approach to select the best debiased subnetwork for predictions. We conduct experiments on three medical datasets for skin lesion classification and chest X-Ray classification and demonstrate that BiasPruner consistently outperforms SOTA CL methods in terms of classification performance and fairness. Our code is available here.
Open AccessBiasPruner: Debiased Continual Learning for Medical Image Classification
Continual Learning (CL) is crucial for enabling networks to dynamically adapt as they learn new tasks sequentially, accommodating new data and classes without catastrophic forgetting. Diverging from conventional perspectives on CL, our paper introduces a new perspective wherein forgetting could actually benefit the sequential learning paradigm. Specifically, we present BiasPruner, a CL framework that intentionally forgets spurious correlations in the training data that could lead to shortcut learning. Utilizing a new bias score that measures the contribution of each unit in the network to learning spurious features, BiasPruner prunes those units with the highest bias scores to form a debiased subnetwork preserved for a given task. As BiasPruner learns a new task, it constructs a new debiased subnetwork, potentially incorporating units from previous subnetworks, which improves adaptation and performance on the new task. During inference, BiasPruner employs a simple task-agnostic approach to select the best debiased subnetwork for predictions. We conduct experiments on three medical datasets for skin lesion classification and chest X-Ray classification and demonstrate that BiasPruner consistently outperforms SOTA CL methods in terms of classification performance and fairness. Our code is available here.
- 8 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 43 (11) , 3767-3779
GC2: Generalizable Continual Classification of Medical Images
Deep learning models have achieved remarkable success in medical image classification. These models are typically trained once on the available annotated images and thus lack the ability of continually learning new tasks (i.e., new classes or data distributions) due to the problem of catastrophic forgetting. Recently, there has been more interest in designing continual learning methods to learn different tasks presented sequentially over time while preserving previously acquired knowledge. However, these methods focus mainly on preventing catastrophic forgetting and are tested under a closed-world assumption; i.e., assuming the test data is drawn from the same distribution as the training data. In this work, we advance the state-of-the-art in continual learning by proposing GC 2 for medical image classification, which learns a sequence of tasks while simultaneously enhancing its out-of-distribution robustness. To alleviate forgetting, GC 2 employs a gradual culpability-based network pruning to identify an optimal subnetwork for each task. To improve generalization, GC 2 incorporates adversarial image augmentation and knowledge distillation approaches for learning generalized and robust representations for each subnetwork. Our extensive experiments on multiple benchmarks in a task-agnostic inference demonstrate that GC 2 significantly outperforms baselines and other continual learning methods in reducing forgetting and enhancing generalization. Our code is publicly available at the following link: https://github.com/ nourhanb/TMI2024-GC2.
- 8 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 43 (11) , 3767-3779
GC2: Generalizable Continual Classification of Medical Images
Deep learning models have achieved remarkable success in medical image classification. These models are typically trained once on the available annotated images and thus lack the ability of continually learning new tasks (i.e., new classes or data distributions) due to the problem of catastrophic forgetting. Recently, there has been more interest in designing continual learning methods to learn different tasks presented sequentially over time while preserving previously acquired knowledge. However, these methods focus mainly on preventing catastrophic forgetting and are tested under a closed-world assumption; i.e., assuming the test data is drawn from the same distribution as the training data. In this work, we advance the state-of-the-art in continual learning by proposing GC 2 for medical image classification, which learns a sequence of tasks while simultaneously enhancing its out-of-distribution robustness. To alleviate forgetting, GC 2 employs a gradual culpability-based network pruning to identify an optimal subnetwork for each task. To improve generalization, GC 2 incorporates adversarial image augmentation and knowledge distillation approaches for learning generalized and robust representations for each subnetwork. Our extensive experiments on multiple benchmarks in a task-agnostic inference demonstrate that GC 2 significantly outperforms baselines and other continual learning methods in reducing forgetting and enhancing generalization. Our code is publicly available at the following link: https://github.com/ nourhanb/TMI2024-GC2.
- 1 December 2023
- Published by Springer Nature
- p. 3-13
Continual-GEN: Continual Group Ensembling for Domain-agnostic Skin Lesion Classification
- 1 December 2023
- Published by Springer Nature
- p. 3-13
Continual-GEN: Continual Group Ensembling for Domain-agnostic Skin Lesion Classification
- 1 December 2023
- Published by Springer Nature
- p. 25-36
AViT: Adapting Vision Transformers for Small Skin Lesion Segmentation Datasets
- 1 December 2023
- Published by Springer Nature
- p. 25-36
AViT: Adapting Vision Transformers for Small Skin Lesion Segmentation Datasets
- 1 October 2023
- Published by Springer Nature
- p. 448-458
MDViT: Multi-domain Vision Transformer for Small Medical Image Segmentation Datasets
- 1 October 2023
- Published by Springer Nature
- p. 448-458
MDViT: Multi-domain Vision Transformer for Small Medical Image Segmentation Datasets
- 14 February 2023
- Published by Springer Nature
- p. 185-202
FairDisCo: Fairer AI in Dermatology via Disentanglement Contrastive Learning
- 14 February 2023
- Published by Springer Nature
- p. 185-202
FairDisCo: Fairer AI in Dermatology via Disentanglement Contrastive Learning
- 21 September 2021
- Published by Springer Nature
- p. 165-175
Culprit-Prune-Net: Efficient Continual Sequential Multi-domain Learning with Application to Skin Lesion Classification
- 21 September 2021
- Published by Springer Nature
- p. 165-175
Culprit-Prune-Net: Efficient Continual Sequential Multi-domain Learning with Application to Skin Lesion Classification