Applications of artificial intelligence in prostate cancer imaging

Detecting MRI-Invisible Prostate Cancers Using a Weakly Supervised Deep Learning Model

Background

MRI is an important tool for accurate detection and targeted biopsy of prostate lesions. However, the imaging appearances of some prostate cancers are similar to those of the surrounding normal tissue on MRI, which are referred to as MRI-invisible prostate cancers (MIPCas). The detection of MIPCas remains challenging and requires extensive systematic biopsy for identification. In this study, we developed a weakly supervised UNet (WSUNet) to detect MIPCas.

Methods

The study included 777 patients (training set: 600; testing set: 177), all of them underwent comprehensive prostate biopsies using an MRI-ultrasound fusion system. MIPCas were identified in MRI based on the Gleason grade (≥7) from known systematic biopsy results.

Results

The WSUNet model underwent validation through systematic biopsy in the testing set with an AUC of 0.764 (95% CI: 0.728-0.798). Furthermore, WSUNet exhibited a statistically significant precision improvement of 91.3% (p < 0.01) over conventional systematic biopsy methods in the testing set. This improvement resulted in a substantial 47.6% (p < 0.01) decrease in unnecessary biopsy needles, while maintaining the same number of positively identified cores as in the original systematic biopsy.

Conclusions

In conclusion, the proposed WSUNet could effectively detect MIPCas, thereby reducing unnecessary biopsies.

Emerging Trends in Artificial Intelligence-Based Urological Imaging Technologies and Practical Applications

The integration of artificial intelligence (AI) into medical imaging has notably expanded its significance within urology. AI applications offer a broad spectrum of utilities in this domain, ranging from precise diagnosis achieved through image segmentation and anomaly detection to improved procedural assistance in biopsies and surgical interventions. Although challenges persist concerning data security, transparency, and integration into existing clinical workflows, extensive research has been conducted on AI-assisted imaging technologies while recognizing their potential to reshape urological practices. This review paper outlines current AI techniques employed for image analysis to offer an overview of the latest technological trends and applications in the field of urology.

MA-QC: Free online software for prostate MR quality control and PI-QUAL assessment

PURPOSE

To validate MRI Analyzer Quality Control (MA-QC), a free and open-source online software designed to facilitate MR data acquisition quality control and PI-QUAL score calculation.

MATERIAL AND METHODS

MA-QC is a web-based software, designed for analysing DICOM data related to MR acquisition parameters. The software allows automatic extraction of 18 technical criteria, and manual input of 12 visual criteria, to calculate the PI-QUAL score. We collected 100 prostate MRI datasets from four MR device manufacturers to test data compatibility, automatic sequence recognition, and robustness of technical criteria extraction from DICOM data. The main issue was to determine the spatial resolution in the phase and frequency directions, due to variable encoding of the DICOM datasets.

RESULTS

Acquisition data could be extracted from all sample examinations (100%), with a median analysis speed of 15.2 ± 4.4 images per second and mean processing time of 96 [11-326] seconds per examination. MA-CQ automatically detected the optimal T2-w, DWI and DCE sequences in 71 out of 100 (71%) cases, and required manual selection of at least one sequence in 29 out of 100 (29%) cases to get the best parameters. Display of technical criteria for the 3 sequences was instantaneous. PI-QUAL score could be calculated in all cases.

CONCLUSION

This software brings substantial help in the quality assessment of prostate MRI examinations, by providing fast extraction of series data and the 18 technical parameters of PI-QUAL. PI-QUAL scoring can be performed in less than two minutes, helping to focus on the visual criteria, allowing use of this software in the clinical workflow in the aim of improving overall image quality in prostate MR imaging.

Biparametric prostate MRI: impact of a deep learning-based software and of quantitative ADC values on the inter-reader agreement of experienced and inexperienced readers

Objective

To investigate the impact of an artificial intelligence (AI) software and quantitative ADC (qADC) on the inter-reader agreement, diagnostic performance, and reporting times of prostate biparametric MRI (bpMRI) for experienced and inexperienced readers.

Materials and methods

A total of 170 multiparametric MRI (mpMRI) of patients with suspicion of prostate cancer (PCa) were retrospectively reviewed by one experienced and one inexperienced reader three times, following a wash-out period. First, only the bpMRI sequences, including T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI) sequences, and apparent diffusion coefficient (ADC) maps, were used. Then, bpMRI and quantitative ADC values were used. Lastly, bpMRI and the AI software were used. Inter-reader agreement between the two readers and between each reader and the mpMRI original reports was calculated. Detection rates and reporting times were calculated for each group.

Results

Inter-reader agreement with respect to mpMRI was moderate for bpMRI, Quantib, and qADC for both the inexperienced (weighted k of 0.42, 0.45, and 0.41, respectively) and the experienced radiologists (weighted k of 0.44, 0.46, and 0.42, respectively). Detection rate of PCa was similar between the inexperienced (0.24, 0.26, and 0.23) and the experienced reader (0.26, 0.27 and 0.27), for bpMRI, Quantib, and qADC, respectively. Reporting times were lower for Quantib (8.23, 7.11, and 9.87 min for the inexperienced reader and 5.62, 5.07, and 6.21 min for the experienced reader, for bpMRI, Quantib, and qADC, respectively).

Conclusions

AI and qADC did not have a significant impact on the diagnostic performance of both readers. The use of Quantib was associated with lower reporting times.

Prostate Health Index and Multiparametric MRI: Partners in Crime Fighting Overdiagnosis and Overtreatment in Prostate Cancer

Simple Summary

In the last decades, the widespread use of PSA as the standard tool for prostate cancer diagnosis led to a high rate of overdiagnosis and overtreatment. More recently, multiparametric magnetic resonance imaging (mpMRI) became part of the diagnostic pathway, and several next-generation PSA-based tests (PHI, PHI density, 4Kscore, STHLM3) have been proposed. The multivariable approach promises to help with a better stratification of PCa patients at initial diagnosis. In this study, we evaluated the performance of the prostate health index (PHI) and mpMRI for the prediction of positive biopsy and of high-grade PCa at radical prostatectomy (RP). Our findings suggested that PHI had a better ability than mpMRI to predict positive biopsy, whereas a comparable performance in the identification of pathological aggressive PCa was pointed out. Notably, PHI and PHI density might represent useful biomarkers to recognize high-grade PCa in patients with low or uncertain PI-RADS scores on mpMRI.

Abstract

Widespread use of PSA as the standard tool for prostate cancer (PCa) diagnosis led to a high rate of overdiagnosis and overtreatment. In this study, we evaluated the performance of the prostate health index (PHI) and multiparametric magnetic resonance imaging (mpMRI) for the prediction of positive biopsy and of high-grade PCa at radical prostatectomy (RP). To this end, we prospectively enrolled 196 biopsy-naïve patients who underwent mpMRI. A subgroup of 116 subjects with biopsy-proven PCa underwent surgery. We found that PHI significantly outperformed both PI-RADS score (difference in AUC: 0.14; p < 0.001) and PHI density (difference in AUC: 0.08; p = 0.002) in the ability to predict positive biopsy with a cut-off value of 42.7 as the best threshold. Conversely, comparing the performance in the identification of clinically significant prostate cancer (csPCa) at RP, we found that PHI ≥ 61.68 and PI-RADS score ≥ 4 were able to identify csPCa (Gleason score ≥ 7 (3 + 4)) both alone and added to a base model including age, PSA, fPSA-to-tPSA ratio and prostate volume. In conclusion, PHI had a better ability than PI-RADS score to predict positive biopsy, whereas it had a comparable performance in the identification of pathological csPCa.