Cost-Sensitive Regularization for Diabetic Retinopathy Grading from Eye Fundus Images

Abstract

Assessing the degree of disease severity in biomedical images is a task similar to standard classification but constrained by an underlying structure in the label space. Such a structure reflects the monotonic relationship between different disease grades. In this paper, we propose a straightforward approach to enforce this constraint for the task of predicting Diabetic Retinopathy (DR) severity from eye fundus images based on the well-known notion of Cost-Sensitive classification. We expand standard classification losses with an extra term that acts as a regularizer, imposing greater penalties on predicted grades when they are farther away from the true grade associated to a particular image. Furthermore, we show how to adapt our method to the modelling of label noise in each of the sub-problems associated to DR grading, an approach we refer to as Atomic Sub-Task modeling. This yields models that can implicitly take into account the inherent noise present in DR grade annotations. Our experimental analysis on several public datasets reveals that, when a standard Convolutional Neural Network is trained using this simple strategy, improvements of 3-5% of quadratic-weighted kappa scores can be achieved at a negligible computational cost. Code to reproduce our results is released at github.com/agaldran/cost_sensitive_loss_classification.

GWAS of Retinal Vessel Tortuosity Identifies 173 Novel Loci Revealing Genes and Pathways Associated with Vascular Pathomechanics and Cardiometabolic Diseases

“…LWNET [2] converts raw RGB fundus images into a categorical image of three categories, which are defined by the following RGB values: 1) artery: red (255,0,0), 2) vein: blue (0,0,255) and 3) background: black (0,0,0) (Supplemental Figure 5a).…”

DeepDRiD: Diabetic Retinopathy—Grading and Image Quality Estimation Challenge

“…One team in sub-challenge 2 proposed and adopted cost-sensitive loss. 40 The teams that selected different training strategies to develop deep learning models are detailed in Table 7 .…”

“… 40 The teams that selected different training strategies to develop deep learning models are detailed in Table 7 .…”

“… SL1, smooth L1 loss; CE, cross-entropy loss; DV, dual view loss; PL, patient-level loss; CS, cost-sensitive loss; 40 L1, L1 loss; CE(5 class), mean loss of 5 class (one versus others); MMoE, multi-gate mixture of expert; 41 GMP, generalized mean pooling; 42 OHEM, online hard example mining; 43 , 44 CV, cross-validation; O, oversampling; ES, early stopping; TL, transfer learning; TTA, test time augmentation; 45 , 46 PLT, pseudo-labeled and labeled training. …”

Balanced-MixUp for Highly Imbalanced Medical Image Classification

“…Common solutions to address data imbalance involve data re-sampling to achieve a balanced class distribution [4], curriculum learning [13], adapted loss functions, e.g. cost-sensitive classification [30,9], or weighting the contribution of the different samples [7,19]. Another approach uses synthetic manipulation of data and/or labels to drive the learning process towards a more suitable solution, like label smoothing [8] or SMOTE [5].…”