Impact on Cost and Expert Time of Data-Efficient Deep Learning for Medical Image Segmentation

Jaiswal A, Rinneburger M, Meyer F, Arjune S, Oberlinkels L, Reimer PAWM, Lotter-Becker L, Akünal Ü, Bujotzek M, Denner S, Maier-Hein K, Stepansky L, May M, Habert M, Köhn A, Schöneck M, Reimer RPWM, Müller RU, Lennartz S, Hokamp NG, Bucher AM, Persigehl T, Caldeira LL (2026)

Publication Type: Journal article

Publication year: 2026

Journal

Radiology: Artificial Intelligence Radiological Society of North America

Book Volume: 8

Article Number: e250200

Journal Issue: 4

DOI: 10.1148/ryai.250200

Abstract

Purpose: To develop and systematically evaluate an iterative training approach, termed the expert-guided annotation loop, for efficient reference standard segmentation generation, including assessment of two sample selection strategies and real-world clinical implementation. Materials and Methods: This retrospective study included 10 datasets comprising 1941 CT and MRI scans from patients with autosomal dominant polycystic kidney disease, prostate cancer, uveal melanoma, thyroid eye disease, or non–small cell lung cancer. nnU-Net segmentation models were iteratively trained using an expert-guided annotation loop with random or active learning–based sample selection. In each iteration, additional samples were added to the training set, and model-generated presegmentations were corrected by expert radiologists to create reference standard annotations. Expert time required for manual segmentation versus presegmentations correction was measured. Model performance and efficiency were assessed using nonparametric tests, and cost savings were estimated for kidney and tumor segmentation using probabilistic sensitivity analysis. Feasibility of end-to-end no-code implementation was evaluated. Results: Fifty-seven segmentation models were trained and evaluated. Final model mean Dice scores ranged from 0.67 to 0.97 for organ segmentation and from 0.64 to 0.69 for lung tumor segmentation across internal and external test sets. Maximum expert time savings were 90.3% for kidney and 48.2% for tumor segmentation (P < .001 and P = .003, respectively), corresponding to estimated per-examination cost savings of $14.30 (95% CI: 5.94, 26.87) and $5.63 (95% CI: −7.26, 26.09), respectively. No-code execution of the expert-guided annotation loop was feasible. Conclusion: The expert-guided annotation loop reduced expert annotation time and enabled estimated cost savings while producing high-quality reference standard segmentations. The no-code workflow was implemented in a clinical environment.

Authors with CRIS profile

Leonard Stepansky Institute of Radiology Matthias May Institute of Radiology

Involved external institutions

Deutsches Krebsforschungszentrum (DKFZ) DE Germany (DE) ImFusion GmbH DE Germany (DE) Fraunhofer-Institut für Bildgestützte Medizin (MEVIS) DE Germany (DE) Universitätsklinikum Frankfurt am Main (KGU) DE Germany (DE)

How to cite

APA:

Jaiswal, A., Rinneburger, M., Meyer, F., Arjune, S., Oberlinkels, L., Reimer, P.A.W.M.,... Caldeira, L.L. (2026). Impact on Cost and Expert Time of Data-Efficient Deep Learning for Medical Image Segmentation. Radiology: Artificial Intelligence, 8(4). https://doi.org/10.1148/ryai.250200

MLA:

Jaiswal, Astha, et al. "Impact on Cost and Expert Time of Data-Efficient Deep Learning for Medical Image Segmentation." Radiology: Artificial Intelligence 8.4 (2026).

BibTeX: Download