# Inflation of test accuracy due to data leakage in deep learning-based classification of OCT images

arXiv: https://arxiv.org/abs/2202.12267 · HTML: https://mlsanity.com/paper/2202.12267/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 46 references that are in the map.
- First posted on arXiv 21 Feb 2022.

## Builds on (most cited first, 4 of 4)

- [A Survey on Deep Learning in Medical Image Analysis](https://mlsanity.com/paper/1702.05747/) (2017)
- [Medical Image Analysis using Convolutional Neural Networks: A Review](https://mlsanity.com/paper/1709.02250/) (2017)
- [Why rankings of biomedical image analysis competitions should be interpreted with care](https://mlsanity.com/paper/1806.02051/) (2018)
- [Optic-Net: A Novel Convolutional Neural Network for Diagnosis of Retinal Diseases from Optical Tomography Images](https://mlsanity.com/paper/1910.05672/) (2019)

## Built on by (most cited first, 17 of 17)

- [Localization Is All You Evaluate: Data Leakage in Online Mapping Datasets and How to Fix It](https://mlsanity.com/paper/2312.06420/) (2023)
- [Coarse race data conceals disparities in clinical risk score performance](https://mlsanity.com/paper/2304.09270/) (2023)
- [AutoPrognosis 2.0: Democratizing Diagnostic and Prognostic Modeling in Healthcare with Automated Machine Learning](https://mlsanity.com/paper/2210.12090/) (2022)
- [Investigating the Quality of DermaMNIST and Fitzpatrick17k Dermatological Image Datasets](https://mlsanity.com/paper/2401.14497/) (2024)
- [Efficacy of MRI data harmonization in the age of machine learning. A multicenter study across 36 datasets](https://mlsanity.com/paper/2211.04125/) (2022)
- [SpanSeq: Similarity-based sequence data splitting method for improved development and assessment of deep learning projects](https://mlsanity.com/paper/2402.14482/) (2024)
- [Integration of nested cross-validation, automated hyperparameter optimization, high-performance computing to reduce and quantify the variance of test performance estimation of deep learning models](https://mlsanity.com/paper/2503.08589/) (2025)
- [Advanced Multi-Architecture Deep Learning Framework for BIRADS-Based Mammographic Image Retrieval: Comprehensive Performance Analysis with Super-Ensemble Optimization](https://mlsanity.com/paper/2508.04790/) (2025)
- [A geometric and deep learning reproducible pipeline for monitoring floating anthropogenic debris in urban rivers using in situ cameras](https://mlsanity.com/paper/2510.23798/) (2025)
- [Beyond Perfect Scores: Proof-by-Contradiction for Trustworthy Machine Learning](https://mlsanity.com/paper/2601.06704/) (2026)
- [Early Linguistic Pattern of Anxiety from Social Media Using Interpretable Linguistic Features: A Multi-Faceted Validation Study with Author-Disjoint Evaluation](https://mlsanity.com/paper/2601.11758/) (2026)
- [On Procrustes Contamination in Machine Learning Applications of Geometric Morphometrics](https://mlsanity.com/paper/2601.18448/) (2026)
- [fastml: Guarded Resampling Workflows for Safer Automated Machine Learning in R](https://mlsanity.com/paper/2604.05225/) (2026)
- [Certification of Machine Learning Models via Directional Sharpness](https://mlsanity.com/paper/2606.25004/) (2026)
- [Do ECG Foundation Models Transfer to Rare Cardiac Diseases? Evidence from Brugada Syndrome Detection](https://mlsanity.com/paper/2607.03009/) (2026)
- [Monte Carlo Dropout Uncertainty and Entropy-Thresholded Selective Prediction for Architecture-Agnostic Brain Tumor MRI Triage](https://mlsanity.com/paper/2607.16317/) (2026)
- [Patient-Level, Leakage-Aware Deep Learning for Cross-Center Periapical Radiograph Classification](https://mlsanity.com/paper/2609.14703/) (2026)

## Abstract (arXiv)

In the application of deep learning on optical coherence tomography (OCT) data, it is common to train classification networks using 2D images originating from volumetric data. Given the micrometer resolution of OCT systems, consecutive images are often very similar in both visible structures and noise. Thus, an inappropriate data split can result in overlap between the training and testing sets, with a large portion of the literature overlooking this aspect. In this study, the effect of improper dataset splitting on model evaluation is demonstrated for three classification tasks using three OCT open-access datasets extensively used, Kermany's and Srinivasan's ophthalmology datasets, and AIIMS breast tissue dataset. Results show that the classification performance is inflated by 0.07 up to 0.43 in terms of Matthews Correlation Coefficient (accuracy: 5% to 30%) for models tested on datasets with improper splitting, highlighting the considerable effect of dataset handling on model evaluation. This study intends to raise awareness on the importance of dataset splitting given the increased research interest in implementing deep learning on OCT data.