Search

Background/Objectives: To develop a deep learning model for esophageal motility disorder diagnosis using high-resolution manometry images with the aid of Gemini. Methods: Gemini assisted in developing this model by aiding in code writing, preprocessing, model optimization, and troubleshooting. Results: The model demonstrated an overall precision of 0.89 on the testing set, with an accuracy of 0.88, a recall of 0.88, and an F1-score of 0.885. It presented better results for multiple categories, particularly in the panesophageal pressurization category, with precision = 0.99 and recall = 0.99, yielding a balanced F1-score of 0.99. Conclusions: This study demonstrates the potential of artificial intelligence, particularly Gemini, in aiding the creation of robust deep learning models for medical image analysis, solving not just simple binary classification problems but more complex, multi-class image classification tasks.

The initial clinical manifestation of acute mesenteric ischemia poses a diagnostic challenge, often leading to delays in identification and subsequent surgical intervention, contributing to adverse outcomes. Serum biomarkers, offering insights into the underlying pathophysiology, hold promise as prognostic indicators for acute mesenteric ischemia. This systematic review comprehensively explores the role of blood biomarkers in predicting clinical outcomes during follow-up for patients with mesenteric ischemia. A thorough literature search across the PubMed, Cochrane Library, and EMBASE databases yielded 33 relevant publications investigating the efficacy of serum biomarkers in predicting outcomes for mesenteric ischemia. Numerous studies underscore the utility of blood biomarkers in swiftly and accurately differentiating between causes of mesenteric ischemia, facilitating a prompt diagnosis. Elevated levels of specific biomarkers, particularly D-dimers, consistently correlate with heightened mortality risk and poorer clinical outcomes. While certain serum indicators exhibit substantial potential in associating with mesenteric ischemia, further research through rigorous human trials is imperative to enhance their consistent predictive ability during the follow-up period. This study underscores the diagnostic and prognostic significance of specific biomarkers for mesenteric ischemia, emphasizing the necessity for standardized procedures in future investigations.

Background: Small bowel disorders present a diagnostic challenge due to the limited accessibility of the small intestine. Accurate diagnosis is made with the aid of specific procedures, like capsule endoscopy or double-ballon enteroscopy, but they are not usually solicited and not widely accessible. This study aims to assess and compare the diagnostic effectiveness of enteroscopy and video capsule endoscopy (VCE) when combined with artificial intelligence (AI) algorithms for the automatic detection of small bowel diseases. Materials and methods: We performed an extensive literature search for relevant studies about AI applications capable of identifying small bowel disorders using enteroscopy and VCE, published between 2012 and 2023, employing PubMed, Cochrane Library, Google Scholar, Embase, Scopus, and ClinicalTrials.gov databases. Results: Our investigation discovered a total of 27 publications, out of which 21 studies assessed the application of VCE, while the remaining 6 articles analyzed the enteroscopy procedure. The included studies portrayed that both investigations, enhanced by AI, exhibited a high level of diagnostic accuracy. Enteroscopy demonstrated superior diagnostic capability, providing precise identification of small bowel pathologies with the added advantage of enabling immediate therapeutic intervention. The choice between these modalities should be guided by clinical context, patient preference, and resource availability. Studies with larger sample sizes and prospective designs are warranted to validate these results and optimize the integration of AI in small bowel diagnostics. Conclusions: The current analysis demonstrates that both enteroscopy and VCE with AI augmentation exhibit comparable diagnostic performance for the automatic detection of small bowel disorders.