What question did this study set out to answer?

The aim is to create an automated system for accurately segmenting the prostate and dominant intraprostatic lesions using automated techniques.

February 8, 2026Physics in Medicine and Biology0 citationsOpen Access

Reinforcement learning-guided segment anything model for MRI prostate and dominant intraprostatic lesions auto-segmentation

JCJingchu ChenJohns Hopkins University MHMingzhe HuEmory University MSMojtaba SafariEmory University

Puntos clave

The aim is to create an automated system for accurately segmenting the prostate and dominant intraprostatic lesions using automated techniques.
Developed a pipeline with two main stages: localization and segmentation.
Utilized a reinforcement learning-based localization network to predict bounding boxes.
Employed a fine-tuned Segment Anything Model (SAM) for actual image segmentation.
Evaluated performance using metrics such as Dice Similarity Coefficient and Intersection over Union.
Achieved a mean DSC of 0.896 for prostate segmentation and 0.592 for DIL segmentation.
For DILs, a detection rate of 89% was recorded.
Performance varied across lesion volume groups, with mean DSC and IoU indicating robust results.

Resumen

Abstract ObjectiveAccurate segmentation of the prostate and dominant intraprostatic lesions (DILs) on magnetic resonance imaging (MRI) is important for prostate cancer radiation therapy treatment planning and targeted dose escalation. However, DIL segmentation remains challenging due to small datasets, institutional bias, and variable imaging protocols. Although the Segment Anything Model (SAM) has shown promise in medical image segmentation, most prior work depends on manual prompts. This study developed a fully automated pipeline that combines localization with a fine-tuned SAM model to segment the prostate and DIL.ApproachTwo datasets were utilized: the PI-CAI dataset, comprising 1,476 patients, and the Cancer Imaging Archive dataset, comprising 803 patients. The pipeline consisted of two stages: (1) a reinforcement learning-based localization network predicted bounding boxes as segmentation inputs, and (2) a fine-tuned SAM model performed segmentation. Model performance was evaluated using the Dice Similarity Coefficient (DSC), Intersection over Union (IoU), and detection rates, with additional analysis based on lesion volumes.Main ResultsThe proposed method achieved a mean and median DSC of 0.896±0.070 and 0.915, and an IoU of 0.818±0.100 and 0.844 for prostate segmentation. For DIL segmentation, the mean and median DSC were 0.592±0.192 and 0.636, IoU of 0.446±0.190 and 0.466, with a detection rate of 89%. Four DIL groups were created based on lesion volume percentile. The mean/median DSC and IoU for each volume group is as follows: 0.5–1.0 cubic centimeters (cc): 0.555±0.201/0.562 1.0-1.8 cc: 0.603±0.185/0.660 1.8-4.0 cc: 0.588±0.183/0.627 >4.0 cc: 0.621±0.197/0.669 SignificanceThis study presented a fully automated prostate and DIL segmentation framework on MRI by integrating a localization network with fine-tuned SAM. The method achieved robust performance across large multi-institutional datasets and diverse lesion shapes. It shows strong potential for application to clinical workflows for prostate cancer radiation therapy planning and treatment.

Preguntar a la IA

Me gusta

Guardar

Ver artículo completo

Cite This Study

Chen et al. (Thu,) studied this question.

synapsesocial.com/papers/6988291e0fc35cd7a88492fb https://doi.org/https://doi.org/10.1088/1361-6560/ae4287

Preguntar a la IA

Me gusta

Guardar

Ver artículo completo