The integration of generative artificial intelligence (AI), particularly large language models (LLMs), into clinical reasoning heralds transformative potential for medical practice. However, their capacity to authentically replicate the complexity of human clinical decision-making remains uncertain—a challenge defined here as the reliability challenge. While studies demonstrate LLMs’ ability to pass medical licensing exams and achieve diagnostic accuracy comparable to physicians, critical limitations persist. Crucially, LLMs mimic reasoning patterns rather than executing genuine logical reasoning, and their reliance on outdated or non-regional data undermines clinical relevance. To bridge this gap, we advocate for a synergistic paradigm where physicians leverage advanced clinical expertise while AI evolves toward transparency and interpretability. This requires AI systems to integrate real-time, context-specific evidence, align with local healthcare constraints, and adopt explainable architectures (e.g. multi-step reasoning frameworks or clinical knowledge graphs) to demystify decision pathways. Ultimately, reliable AI for clinical reasoning hinges on harmonizing technological innovation with human oversight, ensuring ethical adherence to beneficence and non-maleficence while advancing evidence-based, patient-centered care.

Here we aimed to define the prevalence of imposter syndrome (IS) and identify associated characteristics in Chinese medical students and residents enrolled at Peking Union Medical College Hospital (PUMCH). This was a single-center, cross-sectional study of medical students and residents enrolled at PUMCH conducted in September and October 2022. Participants were recruited to complete a 37-question survey on demographics, a Chinese version of the Clance Imposter Phenomenon Scale (CIPS), and self-assessments of anxiety, depression, burnout, sleep quality, challenges of clinical learning, and time allocation. IS prevalence and its associated factors were analyzed. One hundred and forty-eight medical students and 89 residents completed the survey. IS was significant or severe in 62.8% of medical students and 57.2% of residents. Students enrolled in the eight-year program had significantly higher CIPS scores than those enrolled in the 4 + 4 program (66.4 vs. 60.7, p = .005). There were no gender differences in IS prevalence and severity. Participants with severe IS had significantly higher self-rated anxiety, depression, insomnia, and burnout than participants with mild/moderate IS. Participants significantly challenged by clinical learning had significantly higher CIPS scores. IS is both prevalent and severe in Chinese medical students and residents. Classroom learning, an eight-year program, and being challenged by clinical learning are potentially associated with IS.

Non-tuberculous mycobacterial (NTM) infection is usually observed in patients with immunosuppressive conditions. It may also cause unregulated immune responses. While there have been increasing numbers of reported tuberculosis-related HPS (haemophagocytic syndrome), HPS caused by NTM infection is still very rarely reported. We report a previously healthy 21-year-old Chinese female with fever, night sweats and fatigue, in whom HPS was diagnosed according to the HLH-2004 criteria. Mycobacterium intracellulare was cultured from her peripheral blood. After treatment with corticosteroid, clarithromycin, rifampicin, ethambutol and amikacin, the patient finally recovered. We also reviewed relevant publications on NTM infection complicated with HPS and found 11 cases, including ours. Clinical presentations, diagnoses and prognoses were analysed and summarized to deepen our understanding of this rare condition. Most reported NTM-related cases were caused by disseminated infection. The lack of localized symptoms might add to the difficulty involved in making the right diagnosis. While it usually takes time to obtain tissue or blood culture results, granuloma in a bone marrow biopsy might be an early indicator of possible mycobacterial infection. Although treatment varied, the overall prognosis of NTM-related HPS was promising.

Artificial intelligence (AI) has revolutionized many fields, and its potential in healthcare has been increasingly recognized. Based on diverse data sources such as imaging, laboratory tests, medical records, and electrophysiological data, diagnostic AI has witnessed rapid development in recent years. A comprehensive understanding of the development status, contributing factors, and their relationships in the application of AI to medical diagnostics is essential to further promote its use in clinical practice. In this study, we conducted a bibliometric analysis to explore the evolution of task-specific to general-purpose AI for medical diagnostics. We used the Web of Science database to search for relevant articles published between 2010 and 2023, and applied VOSviewer, the R package Bibliometrix, and CiteSpace to analyze collaborative networks and keywords. Our analysis revealed that the field of AI in medical diagnostics has experienced rapid growth in recent years, with a focus on tasks such as image analysis, disease prediction, and decision support. Collaborative networks were observed among researchers and institutions, indicating a trend of global cooperation in this field. Additionally, we identified several key factors contributing to the development of AI in medical diagnostics, including data quality, algorithm design, and computational power. Challenges to progress in the field include model explainability, robustness, and equality, which will require multi-stakeholder, interdisciplinary collaboration to tackle. Our study provides a holistic understanding of the path from task-specific, mono-modal AI toward general-purpose, multimodal AI for medical diagnostics. With the continuous improvement of AI technology and the accumulation of medical data, we believe that AI will play a greater role in medical diagnostics in the future.

Background: Caroli syndrome (CS) is a rare congenital disorder without pathognomonic clinical symptoms or laboratory findings; therefore, the diagnosis is often delayed. The objective of this study was to investigate the diagnostic delay and associated risk factors in CS patients. Methods: This was a retrospective analysis of 16 CS patients admitted to a single tertiary medical center on mainland China. The diagnostic timelines of CS patients were reviewed to demonstrate the initial findings of CS at diagnosis, the risk factors associated with diagnostic delay, and potential clues leading to early diagnosis. Results: The median diagnostic delay was 1.75 years (range: 1 month to 29 years, interquartile range: 6.2 years) in 16 enrolled CS patients. Sex, age, and initial symptoms were not associated with diagnostic delay. 87.5% of CS patients were diagnosed by imaging, and the accuracies of ultrasonography, computed tomography (CT), and magnetic resonance cholangiopancreatography were 25, 69.2, and 83.3%, respectively. The median diagnostic delays for patients with or without CT performed at the first hospital visited according to physician and radiologist suspicion of the diagnosis were 7.4 months and 6 years, respectively (p = 0.021). Hepatic cysts with splenomegaly were detected by ultrasound in over half of CS patients. Conclusions: The majority of CS patients were not diagnosed until complications of portal hypertension had already developed. Recognition and early suspicion of the disease were important factors influencing diagnostic delay of CS. Hepatic cysts plus splenomegaly detected by US might raise the clinical suspicion to include CS in the differential diagnosis.