Class-Balanced Affinity Loss for Highly Imbalanced Tissue Classification in Computational Pathology

Mahbub, Taslim; Obeid, Ahmad; Javed, Sajid; Dias, Jorge; Werghi, Naoufel

doi:10.1007/978-3-031-37660-3_35

Cited by 1 publication

(1 citation statement)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As seen in Table XI and Table XII, our proposed loss (L cf al ) outperforms the vanilla implementation as well as the class-balanced version of the loss across both the datasets. Preliminary results of the class-balanced variant of the loss function were introduced in our ICPR workshop paper [36]. The best-determined hyperparameters (Table IX) were reutilized in this investigation.…”

Section: Ablation Study 1) Loss Functionmentioning

confidence: 99%

Center-Focused Affinity Loss for Class Imbalance Histology Image Classification

Mahbub,

Obeid,

Javed

et al. 2024

IEEE J. Biomed. Health Inform.

Self Cite

View full text Add to dashboard Cite

Early-stage cancer diagnosis potentially improves the chances of survival for many cancer patients worldwide. Manual examination of Whole Slide Images (WSIs) is a timeconsuming task for analyzing tumor-microenvironment. To overcome this limitation, the conjunction of deep learning with computational pathology has been proposed to assist pathologists in efficiently prognosing the cancerous spread. Nevertheless, the existing deep learning methods are ill-equipped to handle finegrained histopathology datasets. This is because these models are constrained via conventional softmax loss function, which cannot expose them to learn distinct representational embeddings of the similarly textured WSIs containing an imbalanced data distribution. To address this problem, we propose a novel center-focused affinity loss (CFAL) function that exhibits 1) constructing uniformly distributed class prototypes in the feature space, 2) penalizing difficult samples, 3) minimizing intra-class variations, and 4) placing greater emphasis on learning minority class features. We evaluated the performance of the proposed CFAL loss function on two publicly available breast and colon cancer datasets having varying levels of imbalanced classes. The proposed CFAL function shows better discrimination abilities as compared to the popular loss functions such as ArcFace, CosFace, and Focal loss. Moreover, it outperforms several SOTA methods for histology image classification across both datasets. The paper code is publicly available at https://github.com/Taslim-M/CFALmax-margin-keras.

show abstract

Section: Ablation Study 1) Loss Functionmentioning

confidence: 99%