Problem
Research questions and friction points this paper is trying to address.
Melanoma Diagnosis
Class Imbalance
Loss Function
Deep Neural Networks
Uncertainty Calibration
On Impact of Loss Function on the Performance of Neural Networks in Melanoma Diagnosis
🤖 AI Summary
This study addresses the degradation of deep learning diagnostic performance caused by class imbalance in melanoma datasets by systematically investigating the effects of various loss functions on neural network classification and uncertainty calibration. Specifically, deep networks are trained using Focal Loss, logit-adjusted softmax cross-entropy, and weighted softmax cross-entropy for comparative analysis. Experimental results demonstrate that Focal Loss significantly outperforms the alternative methods in improving the area under the curve (AUC) while reducing the expected calibration error (ECE). Consequently, it effectively achieves an optimal balance between diagnostic accuracy and uncertainty calibration. These findings provide a reliable optimization strategy for medical image diagnosis under severe class imbalance conditions.
📝 Abstract
Melanoma is the deadliest type of skin cancer, whose early diagnosis is crucial for patients'survival. Image classification using deep learning models has shown promising results for melanoma diagnosis. However, the performance of these models on the melanoma datasets such as SIIM-ISIC melanoma classification dataset is a challenge due to the class imbalance. One of the methods to deal with this challenge is using loss function modifications. In this work, we have investigated the effect of different loss functions on the performance of deep neural networks. We trained these networks using focal loss, logit-adjusted softmax cross-entropy (CE) loss, and weighted softmax CE loss, and we report different metrics for evaluating performance and uncertainty calibration. Our results suggest that focal loss delivers a good combination of performance in terms of AUC and uncertainty calibration in terms of expected calibration error (ECE) simultaneously.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Loss Function
Class Imbalance
Melanoma Diagnosis
Focal Loss
Uncertainty Calibration
🔎 Similar Papers
No similar papers found.
