Search

This article discusses the management of splenic arterial bleeding as an intraprocedural complication during pseudocyst drainage using LAMS (lumen-apposing metal stents). The case study involves a 56-year-old male with a pancreatic pseudocyst who experienced abdominal pain. During the LAMS release, significant hemorrhagic content and actively bleeding vascular structure were observed. An urgent angio-CT scan revealed active arterial bleeding from the splenic artery, leading to radioembolization. The article emphasizes the rarity and life-threatening nature of this complication, highlighting the importance of performing this technique exclusively in reference centers. [Extracted from the article]

This article discusses the development of an artificial intelligence (AI) algorithm for the evaluation of pancreatic cystic lesions (PCLs) during endoscopic ultrasound (EUS). PCLs are increasingly common, and distinguishing between mucinous PCLs (M-PCLs) and non-mucinous PCLs (NM-PCLs) is important for patient management. The study used a multicenter approach, involving two tertiary centers in Portugal and Spain, and developed a convolutional neural network (CNN) based on 42 EUS exams. The CNN achieved high levels of accuracy, sensitivity, and specificity in distinguishing between M-PCLs and NM-PCLs. The authors emphasize the need for further multicenter studies to expand datasets and improve the clinical application of AI algorithms in real-time. [Extracted from the article]

This article discusses the use of artificial intelligence (AI) algorithms applied to digital single-operator cholangioscopy (D-SOC) for the diagnosis and characterization of biliary strictures. The study involved D-SOC exams from three centers in Portugal, Spain, and the United States. The AI model achieved an overall accuracy of 94.1% in detecting malignant biliary strictures, with a sensitivity of 93.5% and a specificity of 94.8%. The model also showed good performance in detecting morphological features associated with malignancy. The authors suggest that AI technologies have the potential to improve the clinical outcome of patients with suspected biliary malignancy. [Extracted from the article]

This article discusses the development of a deep learning model for the automatic differentiation of solid pancreatic lesions using endoscopic ultrasound (EUS). The model was trained using a dataset of 27,756 images from 107 EUS exams conducted in two specialized centers. The model achieved high sensitivity, specificity, positive and negative predictive values, and accuracy in identifying pancreatic ductal adenocarcinoma (PDAC) and pancreatic neuroendocrine tumors (pNETs). The authors conclude that deep learning models can help improve the accuracy of diagnosing pancreatic adenocarcinoma. [Extracted from the article]

Copyright of Revista Ingeniería e Investigación is the property of Universidad Nacional de Colombia, Facultad de Ingenieraia and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this article we describe a domain-independent model to manage the conversation strategies between users and computers. We apply the model in COBBER, a Conversational case-based reasoning (CCBR) framework based on reusable ontologies. The overall goal of the model is to support users following an affective approach in order to keep them in the adequate mood to interact with the computer. We describe the main tasks of the framework and the methodology to instantiate the conversation strategies for a specific domain. We design the dynamic behavior of conversation strategies with causal loops from system dynamics theory. We adapt this theoretical framework to work with the CCBR approach. [ABSTRACT FROM AUTHOR]

Background: Several artificial intelligence systems based on large language models (LLMs) have been commercially developed, with recent interest in integrating them for clinical questions. Recent versions now include image analysis capacity, but their performance in gastroenterology remains untested. This study assesses ChatGPT-4's performance in interpreting gastroenterology images. Methods: A total of 740 images from five procedures-capsule endoscopy (CE), device-assisted enteroscopy (DAE), endoscopic ultrasound (EUS), digital single-operator cholangioscopy (DSOC), and high-resolution anoscopy (HRA)-were included and analyzed by ChatGPT-4 using a predefined prompt for each. ChatGPT-4 predictions were compared to gold standard diagnoses. Statistical analyses included accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the curve (AUC). Results: For CE, ChatGPT-4 demonstrated accuracies ranging from 50.0% to 90.0%, with AUCs of 0.50-0.90. For DAE, the model demonstrated an accuracy of 67.0% (AUC 0.670). For EUS, the system showed AUCs of 0.488 and 0.550 for the differentiation between pancreatic cystic and solid lesions, respectively. The LLM differentiated benign from malignant biliary strictures with an AUC of 0.550. For HRA, ChatGPT-4 showed an overall accuracy between 47.5% and 67.5%. Conclusions: ChatGPT-4 demonstrated suboptimal diagnostic accuracies for image interpretation across several gastroenterology techniques, highlighting the need for continuous improvement before clinical adoption.

Introduction: Endoscopic ultrasound (EUS) allows for characterization and biopsy of pancreatic lesions. Pancreatic cystic neoplasms (PCN) include mucinous (M-PCN) and nonmucinous lesions (NM-PCN). Pancreatic ductal adenocarcinoma (P-DAC) is the commonest pancreatic solid lesion (PSL), followed by pancreatic neuroendocrine tumor (P-NET). Although EUS is preferred for pancreatic lesion evaluation, its diagnostic accuracy is suboptimal. This multicentric study aims to develop a convolutional neural network (CNN) for detecting and distinguishing PCN (namely M-PCN and NM-PCN) and PSL (particularly P-DAC and P-NET)., Methods: A CNN was developed with 378 EUS examinations from 4 international reference centers (Centro Hospitalar Universitário São João, Hospital Universitario Puerta de Hierro Majadahonda, New York University Hospitals, Hospital das Clínicas Faculdade de Medicina da Universidade de São Paulo). About 126.000 images were obtained-19.528 M-PCN, 8.175 NM-PCN, 64.286 P-DAC, 29.153 P-NET, and 4.858 normal pancreas images. A trinary CNN differentiated normal pancreas tissue from M-PCN and NM-PCN. A binary CNN distinguished P-DAC from P-NET. The total data set was divided into a training and testing data set (used for model's evaluation) in a 90/10% ratio. The model was evaluated through its sensitivity, specificity, positive and negative predictive values, and accuracy., Results: The CNN had 99.1% accuracy for identifying normal pancreatic tissue, 99.0% and 99.8% for M-PCN and NM-PCN, respectively. P-DAC and P-NET were distinguished with 94.0% accuracy., Discussion: Our group developed the first worldwide CNN capable of detecting and differentiating the commonest PCN and PSL in EUS images, using examinations from 4 centers in 2 continents, minimizing the impact of the demographic bias. Larger multicentric studies are needed for technology implementation., (Copyright © 2024 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of The American College of Gastroenterology.)

The Becalm project is an open and low-cost solution for the remote monitoring of respiratory support therapies like the ones used in COVID-19 patients. Becalm combines a decision-making system based on Case-Based Reasoning with a low-cost, non-invasive mask that enables the remote monitoring, detection, and explanation of risk situations for respiratory patients. This paper first describes the mask and the sensors that allow remote monitoring. Then, it describes the intelligent decision-making system that detects anomalies and raises early warnings. This detection is based on the comparison of cases that represent patients using a set of static variables plus the dynamic vector of the patient time series from sensors. Finally, personalized visual reports are created to explain the causes of the warning, data patterns, and patient context to the healthcare professional. To evaluate the case-based early-warning system, we use a synthetic data generator that simulates patients' clinical evolution from the physiological features and factors described in healthcare literature. This generation process has been verified with a real dataset and allows the validation of the reasoning system with noisy and incomplete data, threshold values, and life/death situations. The evaluation demonstrates promising results and good accuracy (0.91) for the proposed low-cost solution to monitor respiratory patients.

Competing Interests: The authors declare that they have no conflict of interest.

Background: Due to the heterogeneous and systemic nature of the chronic obstructive pulmonary disease (COPD), the new guidelines are oriented toward individualized attention. Multidimensional scales could facilitate its proper clinical and prognostic assessment, but not all of them were validated in an international primary care cohort, different from the original ones used for model development. Therefore, our main aim is to assess the prognostic capacity of the ADO, BODEx and DOSE indices in primary care for predicting mortality in COPD patients and to validate the models obtained in subgroups of patients, classified by revised Global Initiative for Chronic Obstructive Lung Disease (2011) and updated Spanish Guideline (2014). Besides, we want to confirm that the prognostic capacity of all indices increases if the number of exacerbations is substituted by the interval between them and to assess the impact on health of the patient's lifestyle, social network and adherence to treatment., Methods: Design: External validation of scales, open and prospective cohort study in primary care., Setting: 36 health centres in 6 European high, medium and low income countries., Subjects: 477 patients diagnosed with COPD, captured in clinical visit by their General Practitioner/Nurse., Predictors: Detailed patient history, exacerbations, lung function test and questionnaires at baseline., Outcomes: Exacerbations, all-cause mortality and specific mortality, within 5 years of recruitment., Analysis: Multivariate logistic regression and Cox regression will be used. Possible non-linear effect of the indices will be studied by using Structured Additive Regression models with penalised splines. Subsequently, we will assess different aspects of the regression models: discrimination, calibration and diagnostic precision. Clinical variables modulated in primary care and the interval between exacerbations will be considered and incorporated into the analysis., Discussion: The Research Agenda for General Practice/Family Medicine highlights that the evidence on predictive values of prognostic indices in primary care is scarce. A prospective cohort like that of PROEPOC/COPD provides good opportunities for research into COPD and make communication easier between family practitioners, nursing staff, pneumologists and other professionals, supporting a multi-disciplinary approach to the treatment of these patients., Trial Registration: ISRCTN52402811 . Date: 15/01/2015. Prospectively registered.

Introduction: Early stages of Crohn's disease [CD] are predominantly inflammatory and early treatment could be useful to change the natural history of CD. We aimed to evaluate the impact of early treatment in our cohort of CD patients., Methods: We retrospectively reviewed clinical records of all CD patients at our centre who have received immunomodulators. Time from diagnosis to first CD-related major abdominal surgery or end of follow-up was considered. Dates of diagnosis, of starting immunomodulators (thiopurines / anti-tumour necrosis factor [TNF]), and of the first CD-related surgery when appropriate were collected., Results: Of 422 patients who received thiopurines, 189 operated patients started thiopurines after a median of 117 months (interquartile range [IQR] 44-196) since diagnosis; non-operated patients, after a median of 30 months [IQR 6-128], p < 0,005. Odds ratio [OR] for surgery was 1.006 (95% confidence interval [CI]1.004-1008) for each month of delay in starting thiopurines. Among 272 patients who received anti-TNFs, 137 operated patients started anti-TNFs after a median of 166 months [IQR 90-233] since diagnosis; non-operated patients after a median of 59 months [IQR 14-162]; p < 0,005. OR for surgery was 1.008 [95% CI 1.005-1.010] for each month of delay in starting anti-TNFs. Among 467 patients who received thiopurines and/or anti-TNF, 210 operated patients started any immunomodulator after a median of 120 months [IQR 48-197] since diagnosis and non-operated patients after a median of 30 months [IQR 6-126], p < 0,005. OR for surgery was 1.008 [95% CI 1.005-1.010] for each month of delay in starting immunomodulators., Conclusions: In our experience, time between diagnosis and thiopurine or anti-TNF initiation was associated with the risk of major abdominal surgery in Crohn's disease., (Copyright © 2015 European Crohn’s and Colitis Organisation (ECCO). Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com.)