Explore the vast range of titles available.

Artificial Intelligence ( ai ) systems are precious support for decision-making, with many applications also in the medical domain. The interaction between md s and ai enjoys a renewed interest following the increased possibilities of deep learning devices. However, we still have limited evidence-based knowledge of the context, design, and psychological mechanisms that craft an optimal human– ai collaboration. In this multicentric study, 21 endoscopists reviewed 504 videos of lesions prospectively acquired from real colonoscopies. They were asked to provide an optical diagnosis with and without the assistance of an ai support system. Endoscopists were influenced by ai ( O R = 3.05 ), but not erratically: they followed the ai advice more when it was correct ( O R = 3.48 ) than incorrect ( O R = 1.85 ). Endoscopists achieved this outcome through a weighted integration of their and the ai opinions, considering the case-by-case estimations of the two reliabilities. This Bayesian-like rational behavior allowed the human– ai hybrid team to outperform both agents taken alone. We discuss the features of the human– ai interaction that determined this favorable outcome.

ObjectiveArtificial intelligence (AI) may reduce underdiagnosed or overlooked upper GI (UGI) neoplastic and preneoplastic conditions, due to subtle appearance and low disease prevalence. Only disease-specific AI performances have been reported, generating uncertainty on its clinical value.DesignWe searched PubMed, Embase and Scopus until July 2020, for studies on the diagnostic performance of AI in detection and characterisation of UGI lesions. Primary outcomes were pooled diagnostic accuracy, sensitivity and specificity of AI. Secondary outcomes were pooled positive (PPV) and negative (NPV) predictive values. We calculated pooled proportion rates (%), designed summary receiving operating characteristic curves with respective area under the curves (AUCs) and performed metaregression and sensitivity analysis.ResultsOverall, 19 studies on detection of oesophageal squamous cell neoplasia (ESCN) or Barrett's esophagus-related neoplasia (BERN) or gastric adenocarcinoma (GCA) were included with 218, 445, 453 patients and 7976, 2340, 13 562 images, respectively. AI-sensitivity/specificity/PPV/NPV/positive likelihood ratio/negative likelihood ratio for UGI neoplasia detection were 90% (CI 85% to 94%)/89% (CI 85% to 92%)/87% (CI 83% to 91%)/91% (CI 87% to 94%)/8.2 (CI 5.7 to 11.7)/0.111 (CI 0.071 to 0.175), respectively, with an overall AUC of 0.95 (CI 0.93 to 0.97). No difference in AI performance across ESCN, BERN and GCA was found, AUC being 0.94 (CI 0.52 to 0.99), 0.96 (CI 0.95 to 0.98), 0.93 (CI 0.83 to 0.99), respectively. Overall, study quality was low, with high risk of selection bias. No significant publication bias was found.ConclusionWe found a high overall AI accuracy for the diagnosis of any neoplastic lesion of the UGI tract that was independent of the underlying condition. This may be expected to substantially reduce the miss rate of precancerous lesions and early cancer when implemented in clinical practice.

Detection and diagnosis of colon polyps are key to preventing colorectal cancer. Recent evidence suggests that AI-based computer-aided detection (CADe) and computer-aided diagnosis (CADx) systems can enhance endoscopists' performance and boost colonoscopy effectiveness. However, most available public datasets primarily consist of still images or video clips, often at a down-sampled resolution, and do not accurately represent real-world colonoscopy procedures. We introduce the REAL-Colon (Real-world multi-center Endoscopy Annotated video Library) dataset: a compilation of 2.7 M native video frames from sixty full-resolution, real-world colonoscopy recordings across multiple centers. The dataset contains 350k bounding-box annotations, each created under the supervision of expert gastroenterologists. Comprehensive patient clinical data, colonoscopy acquisition information, and polyp histopathological information are also included in each video. With its unprecedented size, quality, and heterogeneity, the REAL-Colon dataset is a unique resource for researchers and developers aiming to advance AI research in colonoscopy. Its openness and transparency facilitate rigorous and reproducible research, fostering the development and benchmarking of more accurate and reliable colonoscopy-related algorithms and models.

Background: Circumferential or near-circumferential colorectal lesions are challenging to remove endoscopically; therefore, they are often surgically managed. There are limited data on the outcomes of endoscopic submucosal dissection (ESD) for these lesions, usually from Eastern settings, where ESD is more well established. Objective: The objective of the study was to retrospectively analyze the outcomes of circumferential colorectal ESD in a Western center. Methods: Consecutive patients referred for endoscopic resection of colorectal lesions between January 2015 and April 2025 were included if they had undergone ESD for colorectal laterally spreading tumors with ≥90% involvement of the luminal circumference. Results: Overall, 53 patients were enrolled (26 females, 49.1%; 70.6 ± 9.3 years). Mean lesion size was 91.8 ± 25.3 mm. The most frequent lesion location was the rectum (n = 36, 67.9%). Thirty-three lesions (62.3%) were circumferential, and twenty (37.7%) were near-circumferential. Median procedural time was 160.0 min (IQR 112.0–200.0 min). Histological analysis revealed high-grade dysplasia in 25/53 cases (47.2%) and adenocarcinoma in 28 patients (52.8%). Resection was en bloc in 51 cases (96.2%) and R0 in all cases (100%). Curative resection was achieved in 21 out of 28 adenocarcinoma patients (75%). Adverse events were intra-procedural major bleeding (n = 19, 18.9%), post-procedural bleeding (n = 2, 3.8%), delayed bleeding (n = 1, 1.9%), and intra-procedural perforation (n = 3, 5.7%). Post-ESD stricture was observed in 18.9% of cases (10/53); three of these (30%) were asymptomatic. All were successfully managed endoscopically. Patients who developed strictures had a longer median procedural time (206 min vs. 145 min, p = 0.0061) and a larger mean lesion size (110 mm vs. 90 mm, p = 0.035). Conclusions: ESD for colorectal circumferential and near-circumferential lesions was safe and effective in a Western expert center, supporting the use of this technique in this subset of lesions. Strictures are a common consequence that can be effectively managed endoscopically.

Background and AimsEndoscopy is effective in management of bariatric surgery (BS) adverse events (AEs) but a comprehensive evaluation of long-term results is lacking. Our aim is to assess the effectiveness of a standardized algorithm for the treatment of BS-AE.Patients and MethodsWe retrospectively analyzed 1020 consecutive patients treated in our center from 2012 to 2020, collecting data on demographics, type of BS, complications, and endoscopic treatment. Clinical success (CS) was evaluated considering referral delay, healing time, surgery, and complications type. Logistic regression was performed to identify variables of CS.ResultsIn the study period, we treated 339 fistulae (33.2%), 324 leaks (31.8%), 198 post-sleeve gastrectomy twist/stenosis (19.4%), 95 post-RYGB stenosis (9.3 %), 37 collections (3.6%), 15 LAGB migrations (1.5%), 7 weight regains (0.7%), and 2 hemorrhages (0.2%). Main endoscopic treatments were as follows: pigtail-stent positioning under endoscopic view for both leaks (CS 86.1%) and fistulas (CS 77.2%), or under EUS-guidance for collections (CS 88.2%); dilations and/or stent positioning for sleeve twist/stenosis (CS 80.6%) and bypass stenosis (CS 81.5%). After a median (IQR) follow-up of 18.5 months (4.29–38.68), complications rate was 1.9%. We found a 1% increased risk of redo-surgery every 10 days of delay to the first endoscopic treatment. Endoscopically treated patients had a more frequent regular diet compared to re-operated patients.ConclusionsEndoscopic treatment of BS-AEs following a standardized algorithm is safe and effective. Early endoscopic treatment is associated with an increased CS rate.

We retrospectively collected a series of 82 endoscopically removed early colorectal cancers. Histological specimens were revised by two gastrointestinal pathologists, performing a re-evaluation of all risk factors for lymph node metastasis. The comparison between second opinion and first pathological report revealed that lymphovascular invasion and tumor grading showed a lower level of concordance than other parameters. Our results demonstrated that second opinion modified risk assessment in about 10% of cases. It was mainly due to a lack in reporting of some parameters at the first diagnosis and a different evaluation in second opinion for updated guidelines. Considering the subgroup of patients with modified risk assessment, clinical data revealed that tumors, re-classified as low risk, did not develop lymph node metastasis that, conversely, occurred in patients identified as high risk by second opinion. In conclusion, second opinion significantly alters risk perception of endoscopically removed early colorectal carcinomas representing a valuable tool for their appropriate clinical management.

PRPH2 gene mutations are frequently found in inherited retinal dystrophies (IRD) and are associated with a wide spectrum of clinical phenotypes. We studied 28 subjects affected by IRD carrying pathogenic PRPH2 mutations, belonging to 11 unrelated families. Functional tests (best-corrected visual acuity measurement, chromatic test, visual field, full-field, 30 Hz flicker, and multifocal electroretinogram), morphological retino-choroidal imaging (optical coherence tomography, optical coherence tomography angiography, and fundus autofluorescence), and clinical data were collected and analyzed. Common primary complaints, with onset in their 40s, were visual acuity reduction and abnormal dark adaptation. Visual acuity ranged from light perception to 20/20 Snellen. Visual field peripheral constriction and central scotoma were found. Chromatic sense was reduced in one third of patients. Electrophysiological tests were abnormal in most of the patients. Choroidal neovascular lesions were detected in five patients. Three novel PRPH2 variants were found in four different families. Based on the present multimodal study, we identified seven distinct PRPH2 phenotypes in 11 unrelated families carrying either different mutations or the same mutation, both within the same family or among them. Fundus autofluorescence modality turned out to be the most adequate imaging method for early recognition of this dystrophy, and the optical coherence tomography angiography was highly informative to promptly detect choroidal neovascularization, even in the presence of the extensive chorioretinal atrophy phenotype.

Accurate in-vivo optical characterization of colorectal polyps is key to select the optimal treatment regimen during colonoscopy. However, reported accuracies vary widely among endoscopists. We developed a novel intelligent medical device able to seamlessly operate in real-time using conventional white light (WL) endoscopy video stream without virtual chromoendoscopy (blue light, BL). In this work, we evaluated the standalone performance of this computer-aided diagnosis device (CADx) on a prospectively acquired dataset of unaltered colonoscopy videos. An international group of endoscopists performed optical characterization of each polyp acquired in a prospective study, blinded to both histology and CADx result, by means of an online platform enabling careful video assessment. Colorectal polyps were categorized by reviewers, subdivided into 10 experts and 11 non-experts endoscopists, and by the CADx as either “adenoma” or “non-adenoma”. A total of 513 polyps from 165 patients were assessed. CADx accuracy in WL was found comparable to the accuracy of expert endoscopists (CADx WL /Exp; OR 1.211 [0.766–1.915]) using histopathology as the reference standard. Moreover, CADx accuracy in WL was found superior to the accuracy of non-expert endoscopists (CADx WL /NonExp; OR 1.875 [1.191–2.953]), and CADx accuracy in BL was found comparable to it (CADx BL /CADx WL ; OR 0.886 [0.612–1.282]). The proposed intelligent device shows the potential to support non-expert endoscopists in systematically reaching the performances of expert endoscopists in optical characterization.

Background: Colonoscopy is crucial for diagnosing and monitoring inflammatory bowel disease (IBD), assessing disease activity, and detecting dysplasia. However, patient adherence to surveillance remains suboptimal due to discomfort, anxiety, and concerns about bowel preparation. Methods: This multicenter cross-sectional study assessed patient satisfaction with colonoscopy in IBD patients across three Italian centers. Participants completed pre- and post-examination questionnaires, including the Endoscopy Customer Satisfaction Questionnaire (ECSQ) and Perceived Stress Scale (PSS-10). Clinical factors, bowel preparation methods, and healthcare provider expertise were analyzed. Results: Among 444 enrolled patients, overall satisfaction was high (98.8%) but varied across procedural phases. Higher satisfaction was predicted by expert endoscopists (β = 2.11, p = 0.012), disease remission (β = 1.70, p = 0.020), and frequent endoscopic procedures in the last 24 months (β = 0.46, p = 0.041). Conversely, severe disease activity (β = −3.87, p < 0.001) was associated with lower satisfaction. Deep sedation and high-volume bowel preparation negatively impacted satisfaction. Conclusions: Optimizing bowel preparation, enhancing healthcare provider expertise, and implementing stress-reducing strategies could improve patient adherence to surveillance guidelines in IBD care.

We describe the macular morpho-functional assessment of a 65-year-old man affected by stellate nonhereditary idiopathic foveomacular retinoschisis (SNIFR), studied by visual field, SD-OCT, autofluorescence, full-field electroretinogram (ffERG), multifocal electroretinogram (mfERG) and multifocal Photopic Negative Response (mfPhNR) recordings. The typical presentation consists of the foveal appearance of radial cartwheel pattern for the splitting of the retinal layers at the level of the Henle fiber layer (HFL) and the outer plexiform layer (OPL), perfectly seen by Spectral Domain-Optical Coherence Tomography (SD-OCT). Despite a normal function of the outer retina of the peripheral and central retina evaluated by ffERG and mfERG respectively, we observed a reduced function of the retinal elements involved in the retinoschisis by recording mfPhNR that assesses mainly inner retina function (retinal ganglion cells and their axons). Therefore, it is likely that the observed impaired mfPhNR responses reflect the signaling defects derived from the delaminated middle retina and transmitted to the innermost retinal layers.