# Does deep learning model calibration improve performance in class-imbalanced medical image classification?

arXiv: https://arxiv.org/abs/2110.00918 · HTML: https://mlsanity.com/paper/2110.00918/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 34 references that are in the map.
- First posted on arXiv 29 Sep 2021.

## Builds on (most cited first, 10 of 10)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size](https://mlsanity.com/paper/1602.07360/) (2016)
- [Measuring Calibration in Deep Learning](https://mlsanity.com/paper/1904.01685/) (2019)
- [Improved Trainable Calibration Method for Neural Networks on Medical Imaging Classification](https://mlsanity.com/paper/2009.04057/) (2020)
- [Spline-Based Probability Calibration](https://mlsanity.com/paper/1809.07751/) (2018)

## Built on by (most cited first, 9 of 9)

- [On the Alignment of Group Fairness with Attribute Privacy](https://mlsanity.com/paper/2211.10209/) (2022)
- [Enhancing Learning with Label Differential Privacy by Vector Approximation](https://mlsanity.com/paper/2405.15150/) (2024)
- [Mpox-AISM: AI-Mediated Super Monitoring for Mpox and Like-Mpox](https://mlsanity.com/paper/2303.09780/) (2023)
- [Towards Clinical Prediction with Transparency: An Explainable AI Approach to Survival Modelling in Residential Aged Care](https://mlsanity.com/paper/2312.00271/) (2023)
- [Data Alchemy: Mitigating Cross-Site Model Variability Through Test Time Data Calibration](https://mlsanity.com/paper/2407.13632/) (2024)
- [CTDGSI: A comprehensive exploitation of instance selection methods for automatic text classification. VII Concurso de Teses, Dissertacc~oes e Trabalhos de Graduacc~ao em SI -- XXI Simp'osio Brasileiro de Sistemas de Informacc~ao](https://mlsanity.com/paper/2506.07169/) (2025)
- [Predicting Time Pressure of Powered Two-Wheeler Riders for Proactive Safety Interventions](https://mlsanity.com/paper/2601.03173/) (2026)
- [Multiclass Calibration Assessment and Recalibration of Probability Predictions via the Linear Log Odds Calibration Function](https://mlsanity.com/paper/2602.18573/) (2026)
- [C2LUE: Clinically Aligned Patient-Level Ultrasound Diagnosis via Evidence-Aware Multi-Instance Learning](https://mlsanity.com/paper/2610.14143/) (2026)

## Abstract (arXiv)

In medical image classification tasks, it is common to find that the number of normal samples far exceeds the number of abnormal samples. In such class-imbalanced situations, reliable training of deep neural networks continues to be a major challenge. Under these circumstances, the predicted class probabilities may be biased toward the majority class. Calibration has been suggested to alleviate some of these effects. However, there is insufficient analysis explaining when and whether calibrating a model would be beneficial in improving performance. In this study, we perform a systematic analysis of the effect of model calibration on its performance on two medical image modalities, namely, chest X-rays and fundus images, using various deep learning classifier backbones. For this, we study the following variations: (i) the degree of imbalances in the dataset used for training; (ii) calibration methods; and (iii) two classification thresholds, namely, default decision threshold of 0.5, and optimal threshold from precision-recall curves. Our results indicate that at the default operating threshold of 0.5, the performance achieved through calibration is significantly superior (p < 0.05) to using uncalibrated probabilities. However, at the PR-guided threshold, these gains are not significantly different (p > 0.05). This finding holds for both image modalities and at varying degrees of imbalance.