Jacobi Medical Center (NYC Health + Hospitals/Jacobi) is a municipal hospital operated by NYC Health + Hospitals in affiliation with the Albert Einstein College of Medicine. The facility is located in the Morris Park neighborhood of the Bronx, New York City. It is named in honor of German physician Abraham Jacobi, who is regarded as the father of American pediatrics.Founded in 1955 as Bronx Municipal Hospital Center, the hospital opened concurrent with the opening of the Albert Einstein College of Medicine. This was the first time a medical school and municipal hospital entered into a formal affiliation agreement at the same time they were both built—and their relationship continues to this day. Jacobi is a primary clerkship site for 3rd- and 4th-year medical students from the Albert Einstein College of Medicine. Jacobi offers residency training programs in Internal Medicine, Pediatrics and Radiology. It also offers many joint residency programs with Montefiore Medical Center.Jacobi provides health care for some 1.2 million Bronx and New York City area residents. It is one of the 11 acute care hospitals of NYC Health + Hospitals and a partner in the North Bronx Healthcare Network with the North Central Bronx Hospital. As one of the largest medical facilities of NYC, Jacobi houses the Bronx's only burn unit and Level I trauma center. The hospital also houses a Level III neonatal intensive care unit and FDNY EMS Station 20 (formerly NYC*EMS Station 23). Jacobi had over 320,000 clinical visits and over 100,000 emergency department visits in 2016. Jacobi also houses the only Snakebite Treatment Center in the tri-state area.The Jacobi Hyperbaric Medicine Service provides care for both inpatient and ambulatory patients in a large facility featuring a spacious, 23-foot long chamber. This room-like chamber is one which patients and medical support staff can actually walk into. It can comfortably accommodate up to nine patients at a time. Operating around the clock, the Hyperbaric Medicine Service provides life-saving care for burn patients and smoke inhalation injuries. In addition, Jacobi is the designated New York City Referral Center for the Diver's Alert Network, providing emergency care for decompression sickness. It also treats patients with gas gangrene and acute air embolism, and is widely used on an elective basis to promote wound healing from radiation injuries, compromised skin grafts, diabetic ulcers, and osteomyelitis.
Clinical diagnosis is a complex reasoning process in which clinicians gather evidence, form hypotheses, and test them against alternative explanations. In medical training, this reasoning is explicitly developed through counterfactual questioning–e.g., asking how a diagnosis would change if a key symptom were absent or altered–to strengthen differential diagnosis skills. As large language model (LLM)-based systems are increasingly used for diagnostic support, ensuring the interpretability of their recommendations becomes critical. However, most existing LLM-based diagnostic agents reason over fixed clinical evidence without explicitly testing how individual findings support or weaken competing diagnoses. In this work, we propose a counterfactual multi-agent diagnostic framework inspired by clinician training that makes hypothesis testing explicit and evidence-grounded. Our framework introduces counterfactual case editing to modify clinical findings and evaluate how these changes affect competing diagnoses. We further define the Counterfactual Probability Gap, a method that quantifies how strongly individual findings support a diagnosis by measuring confidence shifts under these edits. These counterfactual signals guide multi-round specialist discussions, enabling agents to challenge unsupported hypotheses, refine differential diagnoses, and produce more interpretable reasoning trajectories. Across three diagnostic benchmarks and seven LLMs, our method consistently improves diagnostic accuracy over prompting and prior multi-agent baselines, with the largest gains observed in complex and ambiguous cases. Human evaluation further indicates that our framework produces more clinically useful, reliable, and coherent reasoning. These results suggest that incorporating counterfactual evidence verification is an important step toward building reliable AI systems for clinical decision support.
Clinical expertise improves not only by acquiring medical knowledge, but by accumulating experience that yields reusable diagnostic patterns. Recent LLMs-based diagnostic agents have shown promising progress in clinical reasoning for decision support. However, most approaches treat cases independently, limiting experience reuse and continual adaptation. We propose SEA, a self-learning diagnostic agent with cognitively inspired dual-memory module. We design a reinforcement training framework tailored to our designed agent for joint optimization of reasoning and memory management. We evaluate SEA in two complementary settings. On standard evaluation with MedCaseReasoning dataset, SEA achieves 92.46% accuracy, outperforming the strongest baseline by +19.6%, demonstrating the benefit of jointly optimizing reasoning and memory. On the long-horizon with ER-Reason dataset, SEA attains the best final accuracy (0.7214) and the largest improvement (+0.35 $\Delta$Acc@100), while baseline methods show limited or unstable gains. Expert evaluation further indicates that rules consolidated from SEA show strong clinical correctness, usefulness and trust, suggesting that the induced rules in dual-memory module are reliable and practically meaningful. Overall, SEA improves both diagnostic reasoning ability and continual learning by effectively transforming experience into reusable knowledge.
BACKGROUND:The association between OSA and sudden cardiac death (SCD) risk in patients with hypertrophic cardiomyopathy (HCM) is poorly defined. We hypothesized that OSA may be associated with the risk of cardiac arrest and ventricular fibrillation (CA-VF) in HCM patients. METHODS:We conducted a retrospective analysis of the National Inpatient Sample (2016-2020), including adults with HCM without prior history of SCD. Patients were stratified by the presence of OSA. The primary outcome was a composite of CA-VF. Multivariate logistic regression and propensity score matching (PSM) assessed the association between OSA and outcomes, adjusting for age, gender, family history of SCD, CIHD, AF, use of ICD, syncope, and LVEF<50%. Mediation analysis was performed to assess whether AF, a known outcome of OSA and predictor of SCD in HCM, mediates the association between OSA and the composite endpoint. RESULTS:Of 52,106 HCM admissions, 8848 (16.98%) had OSA. In a multivariate model excluding AF, OSA was associated with CA-VF (OR 1.16, 95% CI 1.10-1.27, p = 0.002). However, after adjusting for AF, this association was no longer significant (OR 1.001, 95% CI 0.91-1.10, p = 0.97). Mediation analysis showed a significant average causal mediation effect (ACME, p < 0.001), with AF accounting for approximately 98.5% of the total effect of OSA on CA-VF. CONCLUSION:In this large cohort of HCM patients, the association between OSA and CA-VF appeared to be predominantly mediated by AF, with no significant independent direct effect observed. These findings suggest that OSA may contribute to arrhythmic risk in HCM primarily through its relationship with AF and should be considered hypothesis-generating.
Obesity and type 2 diabetes are the primary drivers of atherosclerotic cardiovascular disease (ASCVD), the leading cause of death worldwide. Injectable GLP-1 receptor agonists reduce major adverse cardiovascular events. Yet for many individuals, injection hesitancy remains a significant barrier to long-term adherence. Orforglipron is a novel once-daily oral non-peptide GLP-1 receptor agonist designed to provide comprehensive cardiometabolic risk reduction. PRISMA-compliant systematic review and meta-analysis (PROSPERO CRD420251229397) of placebo-controlled phase 2 and phase 3 trials of orforglipron. Random-effects models were used to pool mean differences (MD) and risk ratios (RR) with 95