jump to content jump to footer
Radiologist analyzing whole-body MRI scans of prostate cancer bone metastases using AI-assisted tumor load quantification in mint Lesion

AI-Supported Tumor Load Quantification for Bone Metastasis in Prostate Cancer

To assess treatment response in patients with advanced prostate cancer, radiologists rely on advanced medical imaging. Conventional modalities, such as bone scans and CT, present limitations in detecting early disease progression and lack standardized criteria for evaluating therapy response [1]. By utilizing Whole-Body MRI, radiologists can detect and monitor metastatic spread patterns, evaluating tumor load non-invasively, radiation free, and without the use of contrast agents or tracers.

However, translating this imaging data into routine clinical insights presents significant workflow hurdles:

  • Manual delineation of bone metastasis is a time-consuming process that often takes an hour or more per scan [2].
  • Even semi-automated tools can require between 12 and 23 minutes to finalize results, making routine use difficult [2].
  • Without structured data, assessing therapy effectiveness across multiple time points can be inconsistent.

 

AI Capabilities Integrated into mint Lesion

To streamline this process, mint Lesion integrates with a specialized AI algorithm*, developed in collaboration with The Royal Marsden NHS Foundation Trust and The Institute of Cancer Research, London. The algorithm performs the automated image analysis; mint Lesion delivers the quantitative readout, the structured report and the longitudinal comparison.

  • Targeted Data Extraction: mint Lesion presents total diffusion volume (TDV) and global ADC (gADC) statistics for the whole skeleton and for individual body regions in a structured report with color-coded visualizations.
  • Automated Computation: In a retrospective single-center study of 360 patients with metastatic castration-resistant prostate cancer, image processing from import of the complete whole-body MRI study to generation of the segmentation masks was performed with a reported median computation time of 90 seconds per examination [3].
  • Radiologist Review and Approval: Segmentation results are presented in the reading environment for review, editing and approval by the radiologist before quantitative values such as TDV are recorded. In the study, radiologists reviewed the segmentation alongside additional sequences such as T1-weighted images and relative fat fraction maps; where manual refinement was required, this took an additional 4 to 6 minutes per case [3].
  • Therapy Evaluation Across Time Points: mint Lesion compares measurements from multiple time points in a structured framework that supports longitudinal assessment.

mint Lesion implements this workflow within its reading and reporting environment.


Practical Outcomes for Clinical Workflows

Supported by the AI-powered mint Lesion software, radiologists can incorporate full tumor load quantification into standard clinical workflows and maintain reproducible procedures across different scans and time points. Ultimately, this equips them to supply the tumor board with structured, objective data to evaluate therapy effectiveness, providing a factual basis for decisions regarding ongoing patient care.

Interested in AI-supported quantification in other oncologic applications? Discover how mint Lesion supports structured assessment in multiple myeloma.

 

*based on research funded by the National Institute for Health and Care Research (NIHR)

 

Interested in how AI-supported structured reporting can support your clinical workflows? Get in touch with our team to learn more about mint Lesion.

References

[1] Padhani AR, Lecouvet FE, Tunariu N, et al. (2017). METastasis Reporting and Data System for Prostate Cancer: Practical Guidelines for Acquisition, Interpretation, and Reporting of Whole-body Magnetic Resonance Imaging-based Evaluations of Multiorgan Involvement in Advanced Prostate Cancer. European Urology, 71(1), 81-92. DOI: 10.1016/j.eururo.2016.05.033. PMID 27317091.

[2] Candito A, Blackledge MD, Holbrey R, et al. AI-driven software for automated quantification of skeletal metastases and treatment response evaluation using whole-body diffusion-weighted MRI (WB-DWI) in advanced prostate cancer. Phys Med Biol. 2025 Nov 14;70(22). DOI: 10.1088/1361-6560/ae19c5. PMID: 41166885.

[3] D’Erme L, Avesani G, Russo L, et al. Total diffusion volume on whole-body diffusion-weighted imaging is a strong prognostic marker of disease survival in metastatic castration-resistant prostate cancer. Eur Radiol. 2026. DOI: 10.1007/s00330-026-12792-1.

Systems on FHIR: Driving Healthcare Innovation Through Interoperability
Interoperability is revolutionizing healthcare by enabling the seamless exchange of patient data across systems. This efficient data flow is critical…
Read more
Human-AI Collaboration in Prostate Cancer Diagnosis
As prostate cancer diagnosis becomes increasingly complex, the collaboration between human expertise and artificial intelligence (AI) is emerging as a…
Read more
Insights into the BZKF BORN-Project – Interview with Dr. Mandy Wahlbuhl-Becker
The Bavarian Oncological Radiology Network (BORN) is enhancing cancer diagnostics in Bavaria through standardized imaging and structured…
Read more
scroll-top