Abstract
dc:description.abstract<p>This robustness of histopathology classification models under adversarial and real-world perturbations resembling clinical artifacts is being investigated.</p><p>Using whole-slide images from the CAMELYON17 cohort, four representative architectures—ResNet-18, ResNet-50, HIPT-2MLP, and ViT-B/16 —are benchmarked across controlled pixel-level distortions and artifact-like transformations. Adversarial methods include iterative Fast Gradient Sign, Projected Gradient Descent, Salt-and-Pepper noise, and the Adversarial Watermark—Stain Shift (AWSS). Three defense strategies—Randomized Smoothing, Adversarial Training, and an Artifact Detector—are evaluated for their ability to preserve diagnostic accuracy and model reliability. Structured perturbations consistently degrade performance, with transformer-based models showing the greatest sensitivity. The benchmark developed here offers a reproducible framework for assessing robustness at both tile and slide levels, supporting safer deployment of histopathology AI in clinical workflows.</p>
Degree
thesis:*- Name thesis:degree_name
- Master of Science in Computer Science - (M.S.)
- Discipline thesis:degree_discipline
- Computer Science
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Yajnik, Ruchik N.
- Contributors dc:contributor
-
- Cong Shi
- Lijing Wang
- Xuan Liu
Subjects
dc:subject × 7Identifiers
dc:identifier.*- Repository record dc:identifier
- https://digitalcommons.njit.edu/theses/3555