Can a New AI Tool Have an Impact in MRI Screening for Prostate Cancer?
In addition to segmentation of the prostate, the artificial intelligence (AI)-enabled advance reportedly facilitates PI-RADS scoring by assessing the size and intensity of possible lesions.
An emerging advance in segmenting and analyzing prostate lesions on magnetic resonance imaging (MRI) may help reduce the rate of misdiagnosis and enhance the efficiency of screening for prostate cancer.
Incorporating artificial intelligence (AI) technology, the new PI-RADS assistant (RSIP Vision) provides a baseline for scoring through the established Prostate Imaging Reporting and Data System (PI-RADS) with automated segmentation of the prostate as well as detection and analysis of lesions, according to the company.
RSIP Vision said the PI-RADS assistant assesses a variety of lesion parameters, including intensity, dimensions, and restriction, and enables objective follow-up comparison to lesions from previous MRI exams.
“Reviewing these scans accounts for a large portion of the radiologist’s workload. The PI-RADS assistant will significantly reduce examination time, improve scoring accuracy and precision, and ultimately lead to better clinical outcomes,” emphasized Rabeeh Fares, MD, a senior radiologist at the Tel-Aviv Sourasky Medical Center in Israel.
Dr. Fares, who was involved with the development of the PI-RADS assistant, noted that MRI screening has become increasingly popular for stratifying risk for prostate cancer due to the non-invasive nature of the imaging modality. In addition to a “huge influx of examinations” to report on, Dr. Fares said consistent reporting has been challenging in the past. He pointed out that one of the challenges was the potential of a patient to have scans from multiple MRI machines as the same imaging sequence may appear differently on different MRI devices.
“(This) has a direct effect on the image quality and image appearance,” added Dr. Fares. “This can make it challenging to report accurately, especially in follow-up exams.”
Dr. Fares also noted that variability in PI-RADS reporting has been an issue. He cited an example of two senior radiologists reporting the same lesion in an exam but assigning different PI-RADS scores. The automated objective measurements with the PI-RADS assistant should improve accuracy and consistency with PI-RADS scoring, according to Dr. Fares.
“This way, the whole management ‘wheel’ of prostate cancer patients can move faster and steadier,” noted Dr. Fares.
Considering Breast- and Lesion-Level Assessments with Mammography AI: What New Research Reveals
June 27th 2025While there was a decline of AUC for mammography AI software from breast-level assessments to lesion-level evaluation, the authors of a new study, involving 1,200 women, found that AI offered over a seven percent higher AUC for lesion-level interpretation in comparison to unassisted expert readers.
Can CT-Based Deep Learning Bolster Prognostic Assessments of Ground-Glass Nodules?
June 19th 2025Emerging research shows that a multiple time-series deep learning model assessment of CT images provides 20 percent higher sensitivity than a delta radiomic model and 56 percent higher sensitivity than a clinical model for prognostic evaluation of ground-glass nodules.
