arXiv:2510.26703v1 Announce Type: cross 
Abstract: Purpose: Medical foundation models (FMs) offer a path to build high-performance diagnostic systems. However, their application to prostate cancer (PCa) detection from micro-ultrasound ({\mu}US) remains untested in clinical settings. We present ProstNFound+, an adaptation of FMs for PCa detection from {\mu}US, along with its first prospective validation. Methods: ProstNFound+ incorporates a medical FM, adapter tuning, and a custom prompt encoder that embeds PCa-specific clinical biomarkers. The model generates a cancer heatmap and a risk score for clinically significant PCa. Following training on multi-center retrospective data, the model is prospectively evaluated on data acquired five years later from a new clinical site. Model predictions are benchmarked against standard clinical scoring protocols (PRI-MUS and PI-RADS). Results: ProstNFound+ shows strong generalization to the prospective data, with no performance degradation compared to retrospective evaluation. It aligns closely with clinical scores and produces interpretable heatmaps consistent with biopsy-confirmed lesions. Conclusion: The results highlight its potential for clinical deployment, offering a scalable and interpretable alternative to expert-driven protocols.

ProstNFound+ هي دراسة رائدة تستكشف استخدام نماذج الأساس الطبية للكشف عن سرطان البروستاتا من خلال الميكرو-ألتراساوند. لم يتم اختبار هذا النهج المبتكر في البيئات السريرية من قبل، مما يجعله خطوة مهمة إلى الأمام في تكنولوجيا التشخيص. تسلط المصادقة الاستباقية للدراسة الضوء على إمكانياتها في تعزيز الكشف المبكر وتحسين نتائج المرضى، وهو أمر بالغ الأهمية نظرًا لارتفاع معدل الإصابة بسرطان البروستاتا. من خلال دمج تقنيات الذكاء الاصطناعي المتقدمة، يمكن أن تحدث ProstNFound+ ثورة في كيفية تعاملنا مع تشخيصات السرطان.

ProstNFound+ es un estudio innovador que explora el uso de modelos de fundación médica para la detección del cáncer de próstata a través de micro-ultrasonido. Este enfoque no se ha probado en entornos clínicos antes, lo que lo convierte en un avance significativo en la tecnología de diagnóstico. La validación prospectiva del estudio destaca su potencial para mejorar la detección temprana y los resultados de los pacientes, lo cual es crucial dada la creciente incidencia del cáncer de próstata. Al integrar técnicas avanzadas de IA, ProstNFound+ podría revolucionar nuestra forma de abordar los diagnósticos del cáncer.

ProstNFound+ est une étude révolutionnaire qui explore l'utilisation de modèles de fondation médicale pour détecter le cancer de la prostate par micro-ultrason. Cette approche innovante n'a jamais été testée dans des contextes cliniques auparavant, ce qui en fait une avancée significative dans la technologie de diagnostic. La validation prospective de l'étude souligne son potentiel à améliorer la détection précoce et les résultats pour les patients, ce qui est crucial compte tenu de l'incidence croissante du cancer de la prostate. En intégrant des techniques avancées d'IA, ProstNFound+ pourrait révolutionner notre approche des diagnostics du cancer.

ProstNFound+ is a groundbreaking study that explores the use of medical foundation models for detecting prostate cancer through micro-ultrasound. This innovative approach has not been tested in clinical settings before, making it a significant step forward in diagnostic technology. The study's prospective validation highlights its potential to enhance early detection and improve patient outcomes, which is crucial given the rising incidence of prostate cancer. By integrating advanced AI techniques, ProstNFound+ could revolutionize how we approach cancer diagnostics.

ProstNFound+: A Prospective Study using Medical Foundation Models for Prostate Cancer Detection

arXiv:2512.03196v1 Announce Type: cross 
Abstract: Diffusion MRI (dMRI) enables non-invasive assessment of prostate microstructure but conventional metrics such as the Apparent Diffusion Coefficient in multiparametric MRI lack specificity to underlying histology. Integrating dMRI with the compartment-based biophysical VERDICT (Vascular, Extracellular, and Restricted Diffusion for Cytometry in Tumours) framework offers richer microstructural insights, though clinical gradient systems (40-80 mT/m) suffer from poor signal-to-noise ratio (SNR) at stronger diffusion weightings due to prolonged echo times. Ultra-strong gradients (up to 300 mT/m) can mitigate these limitations by improving SNR and contrast-to-noise ratios (CNR) but their adoption has until recently been limited to research environments due to challenges with peripheral nerve stimulation thresholds and gradient non-uniformity. This study investigates whether physics-informed self-supervised VERDICT (ssVERDICT) fitting applied to ultra-strong gradients enhances prostate cancer characterization relative to current clinical acquisitions. We developed enhanced ssVERDICT fitting approaches using dense multilayer perceptron (Dense MLP) and convolutional U-Net architectures, benchmarking them against non-linear least-squares (NLLS) fitting and Diffusion Kurtosis Imaging across clinical- to ultra-strong gradient systems. Dense ssVERDICT at ultra-strong gradient notably outperformed NLLS VERDICT, boosting median CNR by 47%, cutting inter-patient Coefficient of Variation by 52%, and reducing pooled f_ic variation by 50%. Overall, it delivered the highest CNR, the most stable parameter estimates, and the clearest tumour-normal contrast compared with conventional methods and clinical gradient systems. These findings highlight the potential of advanced gradient systems and deep learning-based modelling to improve non-invasive prostate cancer characterization and reduce unnecessary biopsies.

قدمت دراسة حديثة تقنية تصوير بالرنين المغناطيسي باستخدام تشتت قوي للغاية مع التعلم الذاتي لتحسين تصنيف سرطان البروستاتا. تعتمد هذه الطريقة على إطار VERDICT لتوفير رؤى دقيقة حول البنية الدقيقة، متجاوزةً قيود طرق التصوير التقليدية التي غالبًا ما تفتقر إلى التحديد بالنسبة للخصائص النسيجية.

Un estudio reciente ha presentado una técnica de IRM de difusión con gradientes ultra fuertes combinada con aprendizaje auto-supervisado para mejorar la caracterización del cáncer de próstata. Este enfoque utiliza el marco VERDICT para proporcionar información microestructural detallada, superando las limitaciones de los métodos de imagen tradicionales que a menudo carecen de especificidad hacia las características histológicas.

Une étude récente a introduit une technique d'IRM de diffusion à gradient ultra-fort combinée à un apprentissage auto-supervisé pour améliorer la caractérisation du cancer de la prostate. Cette approche s'appuie sur le cadre VERDICT pour fournir des informations microstructurales détaillées, surmontant les limitations des méthodes d'imagerie traditionnelles qui manquent souvent de spécificité par rapport aux caractéristiques histologiques.

A recent study has introduced an ultra-strong gradient diffusion MRI technique combined with self-supervised learning to enhance the characterization of prostate cancer. This approach leverages the VERDICT framework to provide detailed microstructural insights, overcoming limitations of traditional imaging methods that often lack specificity to histological features.

ProstNFound+: A Prospective Study using Medical Foundation Models for Prostate Cancer Detection

Was this article worth reading? Share it

PlantFCE Model Builder

Pageradar

Hypertune