Niroop Rajashekar

Randomized clinical trials (RCTs) of computer-aided detection (CADe) system-enhanced colonoscopy compared with conventional colonoscopy suggest increased adenoma detection rate (ADR) and decreased adenoma miss rate (AMR), but the effect on detection of advanced colorectal neoplasia (ACN) is unclear. To conduct a systematic review to compare performance of CADe-enhanced and conventional colonoscopy. Cochrane Library, Google Scholar, Ovid EMBASE, Ovid MEDLINE, PubMed, Scopus, and Web of Science Core Collection databases were searched through February 2024. Published RCTs comparing CADe-enhanced and conventional colonoscopy. Average adenoma per colonoscopy (APC) and ACN per colonoscopy were primary outcomes. Adenoma detection rate, AMR, and ACN detection rate (ACN DR) were secondary outcomes. Balancing outcomes included withdrawal time and resection of nonneoplastic polyps (NNPs). Subgroup analyses were done by neural network architecture. Forty-four RCTs with 36 201 cases were included. Computer-aided detection-enhanced colonoscopies have higher average APC (12 090 of 12 279 [0.98] vs. 9690 of 12 292 [0.78], incidence rate difference [IRD] = 0.22 [95% CI, 0.16 to 0.28]) and higher ADR (7098 of 16 253 [44.7%] vs. 5825 of 15 855 [36.7%], rate ratio [RR] = 1.21 [CI, 1.15 to 1.28]). Average ACN per colonoscopy was similar (1512 of 9296 [0.16] vs. 1392 of 9121 [0.15], IRD = 0.01 [CI, -0.01 to 0.02]), but ACN DR was higher with CADe system use (1260 of 9899 [12.7%] vs. 1119 of 9746 [11.5%], RR = 1.16 [CI, 1.02 to 1.32]). Using CADe systems resulted in resection of almost 2 extra NNPs per 10 colonoscopies and longer total withdrawal time (0.53 minutes [CI, 0.30 to 0.77]). Statistically significant heterogeneity in quality and sample size and inability to blind endoscopists to the intervention in included studies may affect the performance estimates. Computer-aided detection-enhanced colonoscopies have increased APC and detection rate but no difference in ACN per colonoscopy and a small increase in ACN DR. There is minimal increase in procedure time and no difference in performance across neural network architectures. None. (PROSPERO: CRD42023422835).

Integration of artificial intelligence (AI) into clinical decision support systems (CDSS) poses a socio-technological challenge that is impacted by usability, trust, and human-computer interaction (HCI). AI-CDSS interventions have shown limited benefit in clinical outcomes, which may be due to insufficient understanding of how health-care providers interact with AI systems. Large language models (LLMs) have the potential to enhance AI-CDSS, but haven’t been studied in either simulated or real-world clinical scenarios. We present findings from a randomized controlled trial deploying AI-CDSS for the management of upper gastrointestinal bleeding (UGIB) with and without an LLM interface within realistic clinical simulations for physician and medical student participants. We find evidence that LLM augmentation improves ease-of-use, that LLM-generated responses with citations improve trust, and HCI varies based on clinical expertise. Qualitative themes from interviews suggest the perception of LLM-augmented AI-CDSS as a team-member used to confirm initial clinical intuitions and help evaluate borderline decisions.

A man in his 50s with a history of sleeve gastrectomy and rivaroxaban use for multiple venous thromboembolic events presented with subacute progressive cough. Axial images of a computerized tomography (CT) scan of the chest showed ground glass and consolidations within the right middle and right lower lobes, and he was treated for presumed community-acquired pneumonia. Given continued symptoms despite 7 days of antibiotics, his CT images were re-reviewed. Wedge shaped airspace disease more recognizable in sagittal views prompted concern for pulmonary infarction, and acquisition of a contrast-enhanced chest CT revealed interval increase of a prior pulmonary embolism and findings of acute on chronic thrombosis. Given the patient was adherent to rivaroxaban and had a negative hypercoagulable work-up, rivaroxaban malabsorption related to the patient’s sleeve gastrectomy was invoked as the cause of anticoagulant failure, and the patient was successfully transitioned to warfarin therapy, highlighting malabsorption as a cause of recurrent VTE in patients who have received bariatric surgery and who take direct-acting oral anticoagulants.

Applications of large language models (LLMs) like ChatGPT have potential to enhance clinical decision support through conversational interfaces. However, challenges of human-algorithmic interaction and clinician trust are poorly understood. GutGPT, a LLM for gastrointestinal (GI) bleeding risk prediction and management guidance, was deployed in clinical simulation scenarios alongside the electronic health record (EHR) with emergency medicine physicians, internal medicine physicians, and medical students to evaluate its effect on physician acceptance and trust in AI clinical decision support systems (AI-CDSS). GutGPT provides risk predictions from a validated machine learning model and evidence-based answers by querying extracted clinical guidelines. Participants were randomized to GutGPT and an interactive dashboard, or the interactive dashboard and a search engine. Surveys and educational assessments taken before and after measured technology acceptance and content mastery. Preliminary results showed mixed effects on acceptance after using GutGPT compared to the dashboard or search engine but appeared to improve content mastery based on simulation performance. Overall, this study demonstrates LLMs like GutGPT could enhance effective AI-CDSS if implemented optimally and paired with interactive interfaces.

The myocyte enhancer factor 2 (MEF2) family of transcription factors plays important roles in developmental processes and adaptive responses. Although MEF2 proteins are known to bind DNA in the nucleus to regulate specific gene expression, there are reports that show that MEF2 also functions in the cytoplasm. Previous structural studies of MEF2 focused exclusively on DNA-bound MEF2 with and without various corepressors or coactivators. While these studies have established a comprehensive structural model of DNA recognition and cofactor recruitment by MEF2, the structure of MEF2 not bound to DNA, which include cytoplasmic MEF2 and free MEF2 in the nucleus, is unknown. Here we determined the structure of the MADS-box/MEF2 domain of MEF2B without DNA nor cofactor. The Apo structure of MEF2B reveals a largely preformed DNA binding interface that may be important for recognizing the shape of DNA from the minor groove side. In addition, our structure also reveals that the C-terminal helix of the MEF2-specific domain could flip up to bind to the hydrophobic groove that serves as the binding sites of MEF2 transcription cofactors. These observations shed new insights into DNA binding and cofactor interaction by MEF2 proteins.