OBJECTIVE: The recent surge in large language models (LLMs) across various fields has yet to be fully realized in traditional Chinese medicine (TCM). This study aims to bridge this gap by developing a large language model tailored to TCM knowledge, enhancing its performance and accuracy in clinical reasoning tasks such as diagnosis, treatment, and prescription recommendations.
METHODS: This study harnessed a wide array of TCM data resources, including TCM ancient books, textbooks, and clinical data, to create 3 key datasets: the TCM Pre-trained Dataset, the Traditional Chinese Patent Medicine (TCPM) Question Answering Dataset, and the Spleen and Stomach Herbal Prescription Recommendation Dataset. These datasets underpinned the development of the Lingdan Pre-trained LLM and 2 specialized models: the Lingdan-TCPM-Chat Model, which uses a Chain-of-Thought process for symptom analysis and TCPM recommendation, and a Lingdan Prescription Recommendation model (Lingdan-PR) that proposes herbal prescriptions based on electronic medical records.
RESULTS: The Lingdan-TCPM-Chat and the Lingdan-PR Model, fine-tuned on the Lingdan Pre-trained LLM, demonstrated state-of-the art performances for the tasks of TCM clinical knowledge answering and herbal prescription recommendation. Notably, Lingdan-PR outperformed all state-of-the-art baseline models, achieving an improvement of 18.39% in the Top@20 F1-score compared with the best baseline.
CONCLUSIONS: This study marks a pivotal step in merging advanced LLMs with TCM, showcasing the potential of artificial intelligence to help improve clinical decision-making of medical diagnostics and treatment strategies. The success of the Lingdan Pre-trained LLM and its derivative models, Lingdan-TCPM-Chat and Lingdan-PR, not only revolutionizes TCM practices but also opens new avenues for the application of artificial intelligence in other specialized medical fields. Our project is available at https://github.com/TCMAI-BJTU/LingdanLLM.

It is essential for modern medical students to continuously enhance their clinical thinking abilities. This study aims to evaluate the efficacy of the combined World Café discussion and case-based learning (CBL) approach within the clinical thinking training course. The clinical thinking training course incorporated the combined World Café discussion and CBL approach. The assessment of the accuracy and rationality of clinical symptoms, medical examination, pathological processes, diagnostic results, diagnostic basis, and drug use was conducted through case-related queries. Feedback from students and instructors regarding the teaching content, teaching process, and teaching effect was gathered through questionnaires. The findings indicate that the students achieved high marks in all assessed areas, including clinical symptoms, medical examination, pathological processes, diagnostic results, diagnostic basis, and drug use. The feedback from students and instructors on the teaching content, teaching process, and teaching effect was positive. Medical educators can use our findings to implement the combined World Café discussion and CBL mode to enhance student engagement.NEW & NOTEWORTHY The combined World Café discussion and case-based learning approach was implemented in the clinical thinking training course. Students\' scores for clinical symptoms, medical examination, pathological process, diagnostic results, diagnostic basis, and drug use were all excellent. Feedback from both students and teachers on the teaching content, teaching process, and teaching effect was positive.

ChatGPT (OpenAI), a cutting-edge natural language processing model, holds immense promise for revolutionizing medical education. With its remarkable performance in language-related tasks, ChatGPT offers personalized and efficient learning experiences for medical students and doctors. Through training, it enhances clinical reasoning and decision-making skills, leading to improved case analysis and diagnosis. The model facilitates simulated dialogues, intelligent tutoring, and automated question-answering, enabling the practical application of medical knowledge. However, integrating ChatGPT into medical education raises ethical and legal concerns. Safeguarding patient data and adhering to data protection regulations are critical. Transparent communication with students, physicians, and patients is essential to ensure their understanding of the technology\'s purpose and implications, as well as the potential risks and benefits. Maintaining a balance between personalized learning and face-to-face interactions is crucial to avoid hindering critical thinking and communication skills. Despite challenges, ChatGPT offers transformative opportunities. Integrating it with problem-based learning, team-based learning, and case-based learning methodologies can further enhance medical education. With proper regulation and supervision, ChatGPT can contribute to a well-rounded learning environment, nurturing skilled and knowledgeable medical professionals ready to tackle health care challenges. By emphasizing ethical considerations and human-centric approaches, ChatGPT\'s potential can be fully harnessed in medical education, benefiting both students and patients alike.

A 30-year-old woman presented with rapidly progressive dementia 1 month after the coronavirus disease 2019 infection. Repeated CSF analysis showed extreme hypoglycorrhachia, while cultures, metagenomic next-generation sequencing, and cytopathology testing of CSF were negative. Laboratory investigations for possible etiologies revealed elevated blood ammonia and cancer antigen 125. Brain MRI demonstrated bilateral symmetric diffuse cortical lesions with mild hyperintensity on T1-weighted image and postcontrast enhancement. A more thorough history and specific examinations subsequently indicated an underlying etiology. This case provides an approach for evaluating young patients with rapidly progressive dementia, extreme hypoglycorrhachia, and diffuse CNS lesions, highlighting the importance of considering a broad differential diagnosis.

BACKGROUND: Virtual simulation and face-to-face simulation are effective for clinical judgment training. Rare studies have tried to improve clinical judgment ability by applying virtual simulation and face-to-face simulation together. This study aimed to evaluate the effect of an integrated non-immersive virtual simulation and high-fidelity face-to-face simulation program on enhancing nursing students\' clinical judgment ability and understanding of nursing students\' experiences of the combined simulation.
METHODS: A sequential exploratory mixed-methods study was conducted in a nursing simulation center of a university in Central China. Third-year nursing students (n = 122) taking clinical training in ICUs were subsequentially assigned to the integrated non-immersive virtual simulation and high-fidelity face-to-face simulation program arm (n = 61) or the face-to-face simulation-only arm (n = 61) according to the order in which they entered in ICU training. Clinical judgment ability was measured by the Lasater Clinical Judgment Rubric (LCJR). Focus group interviews were conducted to gather qualitative data.
RESULTS: Students in both arms demonstrated significant improvement in clinical judgment ability scores after simulation, and students in the integrated arm reported more improvement than students in the face-to-face simulation-only arm. The qualitative quotes provided a context for the quantitative improvement measured by the LJCR in the integrated arm. Most of the quantitative findings were confirmed by qualitative findings, including the domains and items in the LJCR. The findings verified and favored the effect of the combination of non-immersive virtual simulation and high-fidelity face-to-face simulation integrated program on enhancing nursing students\' clinical judgment ability.
CONCLUSIONS: The integrated virtual simulation and face-to-face simulation program was feasible and enhanced nursing students\' self-reported clinical judgment ability. This integrated non-immersive virtual simulation and high-fidelity face-to-face simulation program may benefit nursing students and newly graduated nurses in the ICU more than face-to-face simulation only.

A 74-year-old woman presented with acute-onset right ptosis and binocular diplopia. CT scan showed low-density lesions in the bilateral basal ganglia and adjacent to lateral ventricles. Intracranial aneurysm was not detected. This case highlights the importance of neurologic localization of ophthalmoplegia based on physical examination and the microanatomy of the oculomotor nerve.

High-fidelity patient simulation (HFPS) is widely used in professional training to enhance students\' competence in clinical management. A guideline for HFPS provides a systematic approach to direct students to learning during the simulation process. Problem-solving (PS) and clinical reasoning (CR) skills are essential to developing students\' professional competence in safe and effective care. These two skills should be initiated in the early training. A structured guideline was developed for HFPS. This study aimed to investigate the effects of the structured HFPS guideline on the development of PS and CR skills in junior nursing students. The students were required to go through four sessions, pre-briefing, simulation design, facilitation, and debriefing, for the HFPS; the study utilized the Problem-Solving Inventory (PSI) and the Nurses\' Clinical Reasoning Scale (NCRS) to measure PS and CR abilities before and after HFPS. Bivariate analysis, a one-sample t-test, and an independent t-test were performed to evaluate the performance of the PS and CR skills during the two study periods. A total of 189 students were recruited, with 92 in the intervention group and 97 in the control group. The research assistant was responsible for student recruitment through email invitations and allocating the students into the control group or the intervention group. A Wilcoxon analysis was performed and revealed significant differences in PS and CR between the two groups (p < 0.001). The analytic results showed that the PSI, particularly in domains of Problem-Solving Confidence (PSC) (p < 0.001) and overall PS (p < 0.001), and the CR (p < 0.001) had significant improvement after HFPS, particularly in the intervention group. The study concluded that the structured HFPS guideline significantly improved the students\' problem-solving and clinical reasoning abilities. Nurse educators play an important role in providing explicit learning instructions in a simulation guideline that directs and guides students to learn at each stage of HFPS. The students can be directed to be engaged in their learning through HFPS to enhance their competence in knowledge and skill development (PS and CR) for their personal and professional development.

A 17-year-old girl presented with a long history of cognitive impairment, personality and behavioral changes, dysarthria, and paroxysmal lower-extremity weakness. She was initially suspected of having mitochondrial encephalomyopathy with lactic acidosis and stroke-like episodes because of stroke-like symptoms, such as episodic lower-extremity weakness, as well as abnormal brain MRI findings of generalized cerebral atrophy, extensive high-intensity lesions in the cortex and subcortical white matter on fluid-attenuated inversion recovery images, decreased N-acetyl aspartate/creatine ratio, and a lactate peak in the focal area on spectrum images. However, there were no relatives with similar presentations in the family of the patient. The whole mitochondrial genome and whole-exome sequencing did not suggest pathogenic mutations, and no abnormalities were found in the blood or CSF lactate levels. In this case, we detail the clinical manifestations, diagnostic workup, and imaging findings. This case highlights the importance of assessing cognitive function and the relevant differential diagnoses in an adolescent with cognitive impairment.

A 14-year-old girl presented with acute ascending, symmetric numbness, and flaccid paralysis 3 weeks after a suspected gastrointestinal infection. She had experienced anorexia since this gastrointestinal episode. EMG showed a sensorimotor axonal polyneuropathy. Routine CSF analysis and serum-specific antibodies (antiganglioside and node of Ranvier-associated antibodies) were all negative. Laboratory investigations for possible etiologies revealed only mild metabolic perturbations. During her hospitalization, she developed mild cognitive deficits. Brain MRI showed bilateral symmetric basal ganglia lesions with hyperintensity on T2 fluid-attenuated inversion recovery, diffusion-weighted imaging hyperintensity, and corresponding apparent diffusion coefficient hypointensity, but without contrast enhancement. A more thorough and detailed history indicated exercise intolerance, and specific examinations subsequently revealed an underlying etiology. This case presentation discusses specific etiology of an acute-onset diffuse and symmetric neuropathy after an acquired injury in a teenager, emphasizing the need of a broad differential diagnosis in this condition.

Peripheral neuropathies, especially those with atypical features, remain a diagnostic challenge. In this case, a 60-year-old patient presented with acute-onset weakness starting in the right hand then sequentially involving the left leg, left hand, and right leg over 5 days. The asymmetric weakness was accompanied by persistent fever and elevated inflammatory markers. Subsequent development of rashes combined with careful review of the history led us to the final diagnosis and targeted treatment. This case highlights clinical pattern recognition with the help of electrophysiologic studies in peripheral neuropathies, which provide shortcuts to narrow the differential diagnosis. We also illustrate the important pitfalls from history taking to ancillary testing in diagnosing the rare but treatable cause of peripheral neuropathy (eFigure 1, links.lww.com/WNL/C541).