Performance of artificial intelligence in the characterization of colorectal lesions.

Background: Image-enhanced endoscopy (IEE) has been used in the differentiation between neoplastic and non-neoplastic colorectal lesions through microvasculature analysis. This study aimed to evaluate the computer-aided diagnosis (CADx) mode of the CAD EYE system for the optical diagnosis of colorectal lesions and compare it with the performance of an expert, in addition to evaluating the computer-aided detection (CADe) mode in terms of polyp detection rate (PDR) and adenoma detection rate (ADR). Methods: A prospective study was conducted to evaluate the performance of CAD EYE using blue light imaging (BLI), dichotomizing lesions into hyperplastic and neoplastic, and of an expert based on the Japan Narrow-Band Imaging Expert Team (JNET) classification for the characterization of lesions. After white light imaging (WLI) diagnosis, magnification was used on all lesions, which were removed and examined histologically. Diagnostic criteria were evaluated, and PDR and ADR were calculated. Results: A total of 110 lesions (80 (72.7%) dysplastic lesions and 30 (27.3%) nondysplastic lesions) were evaluated in 52 patients, with a mean lesion size of 4.3 mm. Artificial intelligence (AI) analysis showed 81.8% accuracy, 76.3% sensitivity, 96.7% specificity, 98.5% positive predictive value (PPV), and 60.4% negative predictive value (NPV). The kappa value was 0.61, and the area under the receiver operating characteristic curve (AUC) was 0.87. Expert analysis showed 93.6% accuracy, 92.5% sensitivity, 96.7% specificity, 98.7% PPV, and 82.9% NPV. The kappa value was 0.85, and the AUC was 0.95. Overall, PDR was 67.6% and ADR was 45.9%. Conclusions: The CADx mode showed good accuracy in characterizing colorectal lesions, but the expert assessment was superior in almost all diagnostic criteria. PDR and ADR were high.

The prognostic impact of different stages of acute kidney injury in patients with decompensated cirrhosis: a prospective cohort study.

AIM: To assess the impact of the different stages of acute kidney injury (AKI) on the prognosis of patients hospitalized with decompensated cirrhosis. METHODS: This was a prospective cohort study of consecutive patients admitted in two tertiary hospitals in southern Brazil. Participants were considered eligible if they were admitted for acute decompensation of cirrhosis. The main exposure factor was the onset of AKI. AKI stages were defined according the European recommendations. The outcomes evaluated were survival time and death rates at 28 and 90 days from hospital admission. A χ2 test was used to compare mortality between groups. Kaplan-Meier survival analyses were undertaken assessing time to event as days from AKI diagnosis to death or liver transplant. RESULTS: Two hundred and five patients were included in the study, and 121 met the criteria for AKI. Patients with AKI 1b, AKI 2 and AKI 3 had higher 90-day mortality than patients without AKI (P = 0.008, P < 0.001 and P < 0.001, respectively). However, there was no difference in 90-day mortality when patients with AKI 1a were compared with those without AKI (P = 0.742). The mean survival of patients without AKI was higher than that of patients with AKI 1b (591.4 and 305.4 days, respectively, P = 0.015), while there was no significant difference between the mean survival of patients without AKI and that of patients with AKI 1a (591.4 and 373.6 days, respectively, P = 0.198). CONCLUSION: Only AKI ≥1b seems to substantially impact mortality of patients hospitalized for acute decompensation of cirrhosis.

Acute-on-chronic liver failure: A comparison of three different diagnostic criteria.

INTRODUCTION AND AIM: Different criteria are applied for the diagnosis of acute-on-chronic liver failure (ACLF). Our aim was to compare the performance of different ACLF diagnostic criteria for predicting mortality. MATERIALS AND METHODS: This was a prospective cohort study of adult cirrhotic patients admitted to a tertiary hospital for acute decompensation (AD) of cirrhosis. The evaluated outcome was mortality at 28 and 90 days, according to the different ACLF diagnostic criteria: Chronic Liver Failure Consortium (CLIF-C), Asian Pacific Association for the Study of the Liver-ACLF Research Consortium (AARC) and North American Consortium for the Study of End-Stage Liver Disease (NACSELD). Prognostic performance was evaluated using receiver operating characteristic (ROC) curves. RESULTS: 146 patients were included. 43 (29.5%) with ACLF according to CLIF-C definition, 14 (9.6%) with ACLF by AARC definition, and 6 (4.1%) by NACSELD definition. According to Kaplan-Meier survival analyses median survival of patients with ACLF by CLIF-C definition was 27.0 days, median survival of patients with ACLF by AARC definition was 27.0 days, and median survival of patients with ACLF by NACSELD definition was 4.0 days. The areas under the ROC curves for performance evaluation in predicting mortality at 28 days for CLIF-C, AARC and NACSELD criteria were, respectively, 0.710, 0.560 and 0.561 (p=0.002). Regarding 90-day mortality, the areas under the ROC curves were 0.760, 0.554 and 0.555 respectively (p<0.001). CONCLUSION: ACLF definition proposed by CLIF-C had better performance in predicting mortality at 28 and 90 days when compared to criteria proposed by AARC and NACSELD.