Risk Stratification of prostate imaging reporting and data system (PI-RADS 3) lesions based on mean ADC values.
Motivation: PI-RADS 3 lesions have low cancer-positive rates, resulting in many unnecessary biopsies. Goal(s): This study investigates whether quantitative ADC measurements improve the classification of PI-RADS 3 lesions as cancerous or non-cancerous. Approach: Prostate MRI scans from 01/2017 to 10/2024 were analyzed. Of 1074 PI-RADS 3 lesions, final analysis included 152 PZ lesions and 75 TZ lesions. Results: Final analysis showed significant differences in mADC in PZ (cut-off: 920×10⁻⁶ mm²/sec) and TZ (cut-off: 610 mm2/sec) and TZ PSA density (P <0.001, AUC: 0.723). Combined analysis of mADC values and PSA density gives an AUC of 0.81. Impact: The results show that mean ADC values could help differentiate between cancers and non-cancers in PZ and TZ PI-RADS 3 lesions.
Authors
- Hero K. Hussain (ORCID: https://orcid.org/0000-0003-3236-0652)
- Tejinder Kaur (ORCID: https://orcid.org/0000-0002-2721-4581)
- Juliana Ramirez
- Nicole Seiberlich (ORCID: https://orcid.org/0000-0002-8356-5407)
- Vikas Gulani (ORCID: https://orcid.org/0000-0003-0889-5999)
- Shane A. Wells (ORCID: https://orcid.org/0000-0002-1067-6624)
- Yun Jiang (ORCID: https://orcid.org/0000-0002-0713-9641)
- Julian Neshewat
Publication Details
- Journal
- Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition
- Published
- 2025-09-16
- DOI
- https://doi.org/10.58530/2025/2278
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00