ProstAtlasDiff: Prostate cancer detection on MRI using Diffusion Probabilistic Models guided by population spatial cancer atlases

• Published in: Medical image analysis Vol. 101; p. 103486
• Main Authors: Li, Cynthia Xinran, Bhattacharya, Indrani, Vesal, Sulaiman, Ghanouni, Pejman, Jahanandish, Hassan, Fan, Richard E., Sonn, Geoffrey A., Rusu, Mirabela
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.04.2025
• Subjects: Algorithms; Denoising Diffusion Probabilistic Models; Humans; Image Interpretation, Computer-Assisted - methods; Magnetic Resonance Imaging; Magnetic Resonance Imaging - methods; Male; Models, Statistical; Population atlases; Prostate cancer detection; Prostatic Neoplasms - diagnostic imaging; Prostatic Neoplasms - pathology; Sensitivity and Specificity; Denoising Diffusion Probabilistic Models; Magnetic Resonance Imaging; Population atlases; Prostate cancer detection
• ISSN: 1361-8415; 1361-8423; 1361-8423
• DOI: 10.1016/j.media.2025.103486

Magnetic Resonance Imaging (MRI) is increasingly being used to detect prostate cancer, yet its interpretation can be challenging due to subtle differences between benign and cancerous tissue. Recently, Denoising Diffusion Probabilistic Models (DDPMs) have shown great utility for medical image segmentation, modeling the process as noise removal in standard Gaussian distributions. In this study, we further enhance DDPMs by introducing the knowledge that the occurrence of cancer varies across the prostate (e.g., ∼70% of prostate cancers occur in the peripheral zone). We quantify such heterogeneity with a registration pipeline to calculate voxel-level cancer distribution mean and variances. Our proposed approach, ProstAtlasDiff, relies on DDPMs that use the cancer atlas to model noise removal and segment cancer on MRI. We trained and evaluated the performance of ProstAtlasDiff in detecting clinically significant cancer in a multi-institution multi-scanner dataset, and compared it with alternative models. In a lesion-level evaluation, ProstAtlasDiff achieved statistically significantly higher accuracy (0.91 vs. 0.85, p<0.001), specificity (0.91 vs. 0.84, p<0.001), positive predictive value (PPV, 0.50 vs. 0.35, p<0.001), compared to alternative models. ProstAtlasDiff also offers more accurate cancer outlines, achieving a higher Dice Coefficient (0.33 vs. 0.31, p<0.01). Furthermore, we evaluated ProstAtlasDiff in an independent cohort of 91 patients who underwent radical prostatectomy to compare its performance to that of radiologists, relative to whole-mount histopathology ground truth. ProstAtlasDiff detected 16% (15 lesions out of 93) more clinically significant cancers compared to radiologists (sensitivity: 0.90 vs. 0.75, p<0.01), and was comparable in terms of ROC-AUC, PR-AUC, PPV, accuracy, and Dice coefficient (p≥0.05). Furthermore, we evaluated ProstAtlasDiff in a second independent cohort of 537 subjects and observed that ProsAtlasDiff outperformed alternative approaches. These results suggest that ProstAltasDiff has the potential to assist in localizing cancer for biopsy guidance and treatment planning. [Display omitted] •We developed Diffusion Models to find clinically significant prostate cancer on MRI•We assessed the spatial statistical distribution of cancer and used it as a prior•We tested our proposed approach in multi-institutional, multi-vendor datasets•The model can assist clinicians with cancer localization and treatment planning