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.

Effect of Linked-color Imaging on the Detection of Adenomas in Screening Colonoscopies.

OBJECTIVES: Linked-color imaging (LCI) is a new image-enhancement option that emphasizes mucosal surface contrast, facilitating the differentiation between colorectal lesions and normal mucosa. This study aimed to evaluate the potential of LCI to increase the detection of colorectal adenomas in screening colonoscopies. METHODS: A prospective randomized study was conducted using white-light imaging (WLI), blue-laser imaging (BLI)-bright and LCI. The outcome measures were adenoma detection rate (ADR), mean number of adenomas per patient, and withdrawal time. Lesion characteristics such as size, morphology, location, and histology were also evaluated. RESULTS: A total of 205 patients were randomized, and 251 adenomas were detected. The overall ADR was 62%. The ADR was 52.9% for WLI, 62.1% for BLI-bright, and 71% for LCI, and was significantly higher in the LCI group than in the WLI group (P=0.04). No significant difference was observed between LCI and BLI-bright (P=0.28) or BLI-bright and WLI (P=0.30). The mean number of adenomas per patient was 1.01, 1.03, and 1.62 for WLI, BLI-bright, and LCI, respectively, with a significant difference (P=0.02). Withdrawal time did not differ among the groups. A total of 71 adenomas were detected by WLI, 68 by BLI-bright, and 112 by LCI. There was no difference in the size and morphology of the adenomas detected, nor in the diagnosis of sessile serrated adenomas/polyps. CONCLUSION: LCI significantly increased the detection of adenomas in screening colonoscopies.

Blue laser imaging: a new image-enhanced endoscopy for the diagnosis of colorectal lesions.

BACKGROUND: Image-enhanced endoscopy enables real-time differential diagnosis of colorectal lesions through the observation of microvascular architecture. PURPOSE: To evaluate the efficacy of using blue laser imaging (BLI) for capillary pattern analysis in the differential diagnosis of neoplastic and non-neoplastic lesions. PATIENTS AND METHODS: This prospective study included 920 consecutive superficial lesions diagnosed in 457 patients. The capillary pattern was analysed using BLI-bright magnification on the basis of the Teixeira classification. Histopathology was used as the reference standard. RESULTS: The adenoma detection rate was 42.3%, with a mean of 0.95 adenomas per patient. Neoplastic lesions were predominant (70.3%), of which 33 (5.1%) had advanced histology. Neoplastic progression was significantly increased in patients aged at least 50 years, in lesions at least 10 mm and in lesions located in the right colon (P<0.01). BLI-based capillary pattern analysis showed 95.5% accuracy, 95.7% sensitivity, 95.2% specificity, 97.9% positive predictive value and 90.3% negative predictive value in the diagnosis of neoplastic lesions. For 672 diminutive lesions (≤5 mm), BLI-based capillary pattern analysis showed 95.7% accuracy, 96.6% sensitivity, 93.6% specificity, 97.2% positive predictive value and 92.2% negative predictive value. Analysing only lesions up to 5 mm in the rectum and sigmoid colon, the values were 95.2, 93.9, 96.5, 95.8 and 94.8%, respectively. CONCLUSION: BLI associated with magnification yielded excellent results for the real-time predictive histological diagnosis of colorectal lesions.