Artificial intelligence algorithms aimed at characterizing or detecting prostate cancer on MRI: How accurate are they when tested on independent cohorts? - A systematic review.

PURPOSE: The purpose of this study was to perform a systematic review of the literature on the diagnostic performance, in independent test cohorts, of artificial intelligence (AI)-based algorithms aimed at characterizing/detecting prostate cancer on magnetic resonance imaging (MRI). MATERIALS AND METHODS: Medline, Embase and Web of Science were searched for studies published between January 2018 and September 2022, using a histological reference standard, and assessing prostate cancer characterization/detection by AI-based MRI algorithms in test cohorts composed of more than 40 patients and with at least one of the following independency criteria as compared to the training cohort: different institution, different population type, different MRI vendor, different magnetic field strength or strict temporal splitting. RESULTS: Thirty-five studies were selected. The overall risk of bias was low. However, 23 studies did not use predefined diagnostic thresholds, which may have optimistically biased the results. Test cohorts fulfilled one to three of the five independency criteria. The diagnostic performance of the algorithms used as standalones was good, challenging that of human reading. In the 12 studies with predefined diagnostic thresholds, radiomics-based computer-aided diagnosis systems (assessing regions-of-interest drawn by the radiologist) tended to provide more robust results than deep learning-based computer-aided detection systems (providing probability maps). Two of the six studies comparing unassisted and assisted reading showed significant improvement due to the algorithm, mostly by reducing false positive findings. CONCLUSION: Prostate MRI AI-based algorithms showed promising results, especially for the relatively simple task of characterizing predefined lesions. The best management of discrepancies between human reading and algorithm findings still needs to be defined.

Vascular hyperintensities on baseline FLAIR images are associated with functional outcome in stroke patients with successful recanalization after mechanical thrombectomy.

PURPOSE: The purpose of this study was to assess the prognostic value of vascular hyperintensities on FLAIR images (VHF) at admission MRI in patients with acute ischemic stroke (AIS) achieving successful recanalization after mechanical thrombectomy. MATERIALS AND METHODS: Patients with AIS treated by mechanical thrombectomy following admission MRI from the single-center HIBISCUS-STROKE cohort were assessed for eligibility. VHF were categorized using a four-scale classification and were considered poor when grade < 3 (i.e., absence of distal VHF). Recanalization was considered successful when modified thrombolysis in cerebral infarction score was ≥ 2B Functional outcome was considered poor if modified Rankin scale (mRS) at three months was > 2. Univariable and multiple variable logistic regressions were performed to identify factors associated with poor functional outcome despite successful recanalization. RESULTS: A total of 108 patients were included. There were 65 men and 43 women with a median age of 70.5 years (interquartile range: 55.0, 81.0; age range: 22.0-93.0 years). Among them, 39 subjects (36.1%) had poor functional outcome at three months. Univariable logistic regressions indicated that poorly extended VHF (VHF grade < 3) were associated with a poor functional outcome (P = 0.008) as well as age, hypertension and diabetes, baseline National Institute of Health Stroke Scale (NIHSS) score, pre-stroke mRS, lack of intravenous thrombolysis, cerebral microangiopathy and the presence of microbleeds. Multivariable analysis confirmed that poor VHF status was independently associated with a poor functional outcome (odds ratio [OR], 4.26; 95% confidence interval [CI]: 1.55-12.99; P = 0.007) in combination with hypertension (OR, 1.25; 95% CI: 0.87-1.85; P = 0.02), baseline NIHSS score (OR, 1.09; 95% CI: 1.04-1.20; P = 0.03), pre-stroke mRS (OR, 2.05; 95% CI: 1.07-4.61; P = 0.05) and lack of intravenous thrombolysis (OR, 0.23; 95% CI: 0.08-0.61; P = 0.004). CONCLUSION: Poorly extended VHF (grade <3) at admission MRI are associated with a poor functional outcome at three months despite successful recanalization by mechanical thrombectomy.