Vision transformers for small histological datasets learned through knowledge distillation
Kanwal, Neel; Eftestøl, Trygve Christian; Khoraminia, Farbod; Zuiverloon, Tahlita C M; Engan, Kjersti
Chapter
Accepted version
Permanent lenke
https://hdl.handle.net/11250/3118923Utgivelsesdato
2023Metadata
Vis full innførselSamlinger
Originalversjon
Kanwal, N., Eftestøl, T.C., Khoraminia, F., Zuiverloon, T.C.M, Engan, K. (2023). Vision transformers for small histological datasets learned through knowledge distillation. I: Advances in Knowledge Discovery and Data Mining 27th Pacific-Asia conference on knowledge discovery and data mining, PAKDD 2023, Osaka, Japan, May 25–28, 2023 : Proceedings, Part III. Springer Nature, s. 167-179. 10.1007/978-3-031-33380-4_13Sammendrag
Computational Pathology (CPATH) systems have the potential to automate diagnostic tasks. However, the artifacts on the digitized histological glass slides, known as Whole Slide Images (WSIs), may hamper the overall performance of CPATH systems. Deep Learning (DL) models such as Vision Transformers (ViTs) may detect and exclude artifacts before running the diagnostic algorithm. A simple way to develop robust and generalized ViTs is to train them on massive datasets. Unfortunately, acquiring large medical datasets is expensive and inconvenient, prompting the need for a generalized artifact detection method for WSIs. In this paper, we present a student-teacher recipe to improve the classification performance of ViT for the air bubbles detection task. ViT, trained under the student-teacher framework, boosts its performance by distilling existing knowledge from the high-capacity teacher model. Our best-performing ViT yields 0.961 and 0.911 F1-score and MCC, respectively, observing a 7% gain in MCC against stand-alone training. The proposed method presents a new perspective of leveraging knowledge distillation over transfer learning to encourage the use of customized transformers for efficient preprocessing pipelines in the CPATH systems.