English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep learning models have revolutionized the field of medical image analysis,offering significant promise for improved diagnostics and patient care.However, their performance can be misleadingly optimistic due to a hiddenpitfall called 'data leakage'. In this study, we investigate data leakage in 3Dmedical imaging, specifically using 3D Convolutional Neural Networks (CNNs) forbrain MRI analysis. While 3D CNNs appear less prone to leakage than 2Dcounterparts, improper data splitting during cross-validation (CV) can stillpose issues, especially with longitudinal imaging data containing repeatedscans from the same subject. We explore the impact of different data splittingstrategies on model performance for longitudinal brain MRI analysis andidentify potential data leakage concerns. GradCAM visualization helps revealshortcuts in CNN models caused by identity confounding, where the model learnsto identify subjects along with diagnostic features. Our findings, consistentwith prior research, underscore the importance of subject-wise splitting andevaluating our model further on hold-out data from different subjects to ensurethe integrity and reliability of deep learning models in medical imageanalysis.

How You Split Matters: Data Leakage and Subject Characteristics Studies in Longitudinal Brain MRI Analysis