Graph Representation Learning of Longitudinal Medical Imaging Trajectories for Treatment Response Prediction
2026-07-06 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial IntelligenceMachine Learning
AI summaryⓘ
The authors developed a new method that uses 3D medical images taken over time to predict how breast cancer patients will respond to chemotherapy before surgery. Their approach combines advanced graph neural networks with special learning techniques to better understand changes across multiple MRI scans. Tested on data from 585 patients, their method performed better than other image-based prediction tools. They also studied how the number and timing of scans affect prediction accuracy. The authors plan to share their tools publicly to help further research in this area.
pathological complete response (pCR)neoadjuvant chemotherapy (NACT)breast cancerDCE-MRIgraph neural networksself-supervised learninglongitudinal imagingtreatment response prediction
Authors
Johannes Kiechle, Richard Osuala, Daniel M. Lang, Stefan M. Fischer, Ivana Janíčková, Karim Lekadir, Julia A. Schnabel, Jan C. Peeken
Abstract
In patients with breast cancer, pathological complete response (pCR) has been established as a clinically meaningful surrogate marker for long-term outcomes. While commonly treated with neoadjuvant chemotherapy (NACT), effective treatment decision-making remains challenging, as therapeutic response can vary substantially across patients, calling for predictive models capable of accurately estimating individualized treatment response. To address this, we propose an imaging-based 3D spatio-temporal framework for treatment response prediction that integrates a state-of-the-art graph neural network with relational modeling of temporal interactions across timepoints alongside three novel complementary self-supervised treatment trajectory representation learning objectives. Experiments across a cohort of 585 patients from the public ISPY-2 dataset demonstrate that our method substantially outperforms both vision and self-supervised learning baselines across several classification metrics. Alongside establishing a breast cancer pCR prediction benchmark, we include a principled ablation of our method and further introduce and empirically assess the impact of the available number of DCE-MRI timepoints per patient trajectory and the inclusion of inter-scan time-differences. Overall, our study substantiates the utility of clinically meaningful longitudinal medical imagaging modeling for predicting NACT-induced pCR. We will publicly share our code repository and a user-friendly PyPI library for dataset curation upon publication, effectively promoting reproducible open-source research.