Artificial intelligence applications in prostate cancer.

Artificial intelligence (AI) applications have enabled remarkable advancements in healthcare delivery. These AI tools are often aimed to improve accuracy and efficiency of histopathology assessment and diagnostic imaging interpretation, risk stratification (i.e., prognostication), and prediction of therapeutic benefit for personalized treatment recommendations. To date, multiple AI algorithms have been explored for prostate cancer to address automation of clinical workflow, integration of data from multiple domains in the decision-making process, and the generation of diagnostic, prognostic, and predictive biomarkers. While many studies remain within the pre-clinical space or lack validation, the last few years have witnessed the emergence of robust AI-based biomarkers validated on thousands of patients, and the prospective deployment of clinically-integrated workflows for automated radiation therapy design. To advance the field forward, multi-institutional and multi-disciplinary collaborations are needed in order to prospectively implement interoperable and accountable AI technology routinely in clinic.

Dose uncertainty due to needle-tip localization error in prostate seed implantation.

Purpose: This study quantified the dosimetric uncertainty caused by needle-tip detection errors in ultrasound images due to bevel-tip orientation differences, with respect to the location on template grid. Material and methods: Trans-rectal ultrasound (TRUS) system with physical template grid and 18-gauge bevel-tip brachytherapy needles were used. TRUS was set at 6.5 MHz in water phantom, and measurements were taken with 50% and 100% B-mode TRUS gains. Needle-tip localization errors were then retrospectively applied back to 45 prostate seed implant plans to evaluate the important planning parameters for the prostate (D90, V100, V150, and V200), urethra (D10 and D30), and rectum (V100, D2cc, and D0.1cc), following the ABS and AAPM TG-137 guidelines. Results: The needle-tip detection errors for 50% and 100% TRUS gains were 3.7 mm (max) and 5.2 mm (max), respectively. The observed significant decrease in prostate coverage (mean D90 lower by 12.8%, and V100 lower by 3.9% for smaller prostates) after seed placements were corrected by compensating the needle-tip detection errors. Apex of the prostate was hotter, and the base was cooler. Dosimetric difference for urethral and rectal parameters were not statistically significant. Conclusions: This study revealed that the beveled needle-tip orientation could considerably impact the needle tips detection accuracy, based on which the seeds might be delivered. These errors can lead to significant dosimetric uncertainty in prostate seed implantation.