Newest papers

Evaluation of Uncertainty-Aware Multi-Software Ensembles for Hippocampal Segmentation

Gabriel Oliveira-Stahl, Anna Schroder, James C. Moggridge, Hamza A. Salhab, Caroline Micallef, Josephine Barnes, M. Jorge Cardoso, Carole H. Sudre, Matthew Grech‐Sollars · 2026 · The Journal of Machine Learning for Biomedical Imaging

DOI: 10.59275/j.melba.2026-e77b · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Segmen hippocampal yang akurat dapat menjadi alat yang berguna untuk mendiagnosis dan memantau kondisi neurologis seperti penyakit Alzheimer dan epilepsi. Sementara banyak metode segmentasi otomatis ada, adopsi klinis mereka tetap terbatas. Penilaian ketidakpastian yang dapat diandalkan dapat meningkatkan kepercayaan dan memfasilitasi terjemahan klinis. Studi ini mengevaluasi seperangkat lima metode segmen hippocampal yang heterogen — InnerEye, ASHS, FastSurfer, HippoSeg, dan FreeSurfer — di dua dataset demensia dan satu dataset epilepsi. Sub-set yang berisi InnerEye, FastSurfer, dan HippoSeg studi muncul sebagai akurat dan efisien, menekankan kelayakan untuk menyeimbangkan biaya komputasi dan kinerja. Selain itu, kuantifikasi

Machine translation (Meta M2M100); the English original follows.

Abstract

Accurate hippocampal segmentation can be a useful tool for diagnosing and monitoring neurological conditions such as Alzheimer’s disease and epilepsy. While numerous automated segmentation methods exist, their clinical adoption remains limited. Reliable uncertainty assessment can enhance trust and facilitate clinical translation. This study evaluates ensembles of five heterogeneous hippocampal segmentation methods — InnerEye, ASHS, FastSurfer, HippoSeg, and FreeSurfer — across two dementia datasets and one epilepsy dataset. The sub-ensemble containing InnerEye, FastSurfer, and HippoSeg emerged as both accurate and efficient, highlighting the feasibility of balancing computational cost and performance. Additionally, ensemble-derived uncertainty quantification with sample variance, mutual information, and predictive entropy is shown to reduce inaccurate segmentations by flagging low-confidence cases, potentially providing a mechanism for automatically escalating ambiguous cases for expert assessment. Lastly, this study explores the downstream clinical utility of ensemble-derived uncertainty, finding a significant correlation with motion artefact severity and a modest contribution to Alzheimer’s disease classification.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Segmentation, Flagging, Computer science, Artificial intelligence, Entropy (arrow of time), Machine learning

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange