With the rapid development of artificial intelligence, AI tools have demonstrated significant value in medical research and academic writing, enhancing efficiency in data processing, literature retrieval, manuscript drafting, visualization, and interdisciplinary collaboration. Beyond serving as auxiliary tools, AI is increasingly becoming a research partner, contributing to hypothesis generation, experimental design, and multimodal data analysis, thereby fostering a paradigm shift toward“ human–AI coresearch”. Typical applications include rapid drafting of medical manuscripts, research integrity checks, and automated generation of imaging reports and scientific figures. However, the widespread adoption of AI also raises challenges concerning authorship, data traceability, content reliability, and privacy protection. International guidelines such as the ICMJE Recommendations and public statements from Science and Nature explicitly emphasize that AI tools cannot be listed as authors, that their use must be transparently disclosed, and that ultimate responsibility lies with human researchers. Therefore, it is urgent to establish a normative framework centered on transparency, accountability, verifiability, and compliant openness, ensuring that AI delivers both efficiency and innovation while laying the foundation for a sustainable research ecosystem in the digital era.

Key Words: artificial intelligence; medical research paradigm; data-driven science; generative models; research ethics

Primary hospitals play a crucial role in early screening and diagnosis of lung cancer. This not only significantly improves the survival and cure rates of patients, but also greatly enhances their quality of life, reduces the burden on the healthcare system, optimizes resource allocation, and promotes the synchronous development of related medical industries and the economy. However, primary hospitals still face multiple challenges in lung cancer screening and early diagnosis. Issues such as outdated equipment and technology, uneven levels of personnel, unequal distribution of resources, lack of patient awareness, and insufficient policy support are particularly prominent. To overcome these challenges, we need to address them from multiple dimensions, including updating medical equipment, enhancing personnel training, optimizing resource allocation, improving patient education levels, and seeking more policy support.

Key Words: pulmonary nodules; lung cancer; primary hospital; screening; diagnosis; artificial intelligence

Employing virtual reality and augmented reality technologies in medical education can provide immersive learning experiences, interactive collaboration, personalized and adaptive learning, seamless integration of the real and virtual worlds, as well as medical humanities, not only revolutionizing the content and methods of traditional medical education but also offering medical students an even more realistic and enriching learning experience. The prospects for the development of medical education in the metaverse are immense, not only enhancing the quality of medical education but also driving innovation and transformation across the entire educational system.

Key Words: metaverse; medicine; education

Research on digital human GPT in medicine mainly focuses on its applications in healthcare. This technology can help doctors make diagnoses faster and more accurately by automatically interpreting medical images and electronic medical records, thereby improving diagnostic accuracy and efficiency. At the same time, it can provide personalized health education and patient care, which can improve the patient experience and increase patient satisfaction and compliance. In addition, GPT can automate the processing of large amounts of textual data, significantly reducing the workload of medical staff and reducing medical costs. Its prediagnosis and health management functions can also help detect and prevent diseases early, reducing the cost of later treatment. When applied in scientific research, GPT can identify anomalies in medical data and help researchers discover new treatments or disease prediction models. It can also automatically generate new hypotheses and protocols based on existing medical knowledge, providing practical recommendations for researchers. In addition, GPT can help solve medical problems and promote the progress of scientific research through reasoning and logical thinking. Looking forward to the future, digital human GPT in medicine has broad development prospects. With the continuous advancement of technology and the increasing demand for medical care, the application of GPT in the medical and health fields will be deeper and more extensive. It can not only improve the quality and efficiency of medical services but also promote innovation in and the development of medical research. At the same time, with the increasing demand for privacy and data security, we must understand how we can ensure the safe storage and processing of sensitive medical data, avoid the risk of data leakage, and maintain patient privacy and data compliance to ensure the future development of digital human GPT in medicine.

Key Words: digital human in medicine; GPT; natural language processing; medical; healthcare

The treatment of prostate cancer (PCa) is shifting towards the use of highly accurate image-guided procedures in order to achieve better oncologic results. A common strategy consists of using both pre-operative multiparametric magnetic resonance imaging and intra-operative transrectal ultrasound during a prostate biopsy or focal ablation procedure, thus offering enhanced localisation information through high spatial resolution and dynamic response, respectively. However, reliable non-rigid registration is still technically challenging owing to differences in cross-modal imaging physics as well as large deformations between the two modalities caused by rectal probe compression; this paper reviews how deep learning has evolved, focusing on convolutional neural networks, generative models, including generative adversarial networks and diffusion models, and transformer-based architectures. We discuss the extent to which they utilise biomechanical priors to inform the solution of registration problems, against standardized challenges such as µ-RegPro. State-of-the-art approaches achieve sub-millimetre target registration errors with real-time inference times for intra-operative deployment. Addressing outstanding challenges related to interpretation and generalization, this review provides an outlook of the road map to develop "physics-aware" smart interventional systems. All these developments represent important steps toward a fully automated, precise, and minimally invasive PCa management pipeline.

With the increasing adoption of low-dose computed tomography (LDCT) screening and the widespread use of chest CT in health examinations, chronic disease management, and multidisciplinary care, the detection rate of pulmonary nodules has risen substantially. Accordingly, the focus of pulmonary nodule management has shifted from simple lesion detection to refined evaluation, risk stratification, and longitudinal follow-up. Traditional radiologic assessment mainly relies on nodule diameter, density, margin characteristics, and interval growth. Although these approaches remain clinically valuable, they are limited by interobserver variability, suboptimal reproducibility, and difficulty in longitudinal comparison, especially in small nodules, juxta-vascular nodules, pleural-based nodules, subsolid nodules, and multiple nodules. Deep learning can automatically extract multi-scale imaging representations from two-dimensional, three-dimensional, and longitudinal CT data. In recent years, it has been widely applied to pulmonary nodule detection, precise segmentation, phenotypic characterization, malignancy risk prediction, dynamic follow-up, and progression forecasting. This review summarizes the clinical basis of refined pulmonary nodule imaging assessment and the current management framework, and systematically discusses recent advances in deep learning for nodule detection and segmentation, radiologic phenotype characterization, malignancy risk stratification, longitudinal follow-up, and clinical translation. In addition, based on the Fleischner Society guidelines, ACR Lung-RADS, BTS guideline, and recent consensus statements on subsolid nodules, this review analyzes the major barriers to real-world implementation, including insufficient external validation, heterogeneity of reference standards, limited interpretability, poor probability calibration, and incomplete workflow integration.

Key Words: pulmonary nodule; deep learning; low-dose computed tomography; ground-glass nodule; refined imaging assessment; risk stratification

Objective: To evaluate the effectiveness of integrating medical dispute case analysis into doctor-patient communication training for undergraduate medical students. Methods: During the 2025 academic year, the Medical Humanities teaching team at Songjiang Hospital, Affiliated to Shanghai Jiao Tong University School of Medicine, redesigned the Doctor-Patient Communication course for undergraduates at Kangda College, Nanjing Medical University. The reform group (n=42, 2025 cohort) received instruction integrating medical dispute case analysis, while the traditional group (n=42, 2024 cohort) received conventional teaching. Outcomes were evaluated using satisfaction surveys, course experience questionnaires, regular assessments, final exams, and video-based analyses of classroom engagement. Results: Compared with the traditional group, the reform group demonstrated significantly higher rates of students reporting "very satisfied" with learning interest, teaching method evaluation, and perceived self-improvement (P<0.05 or P<0.01). The reform group also had a higher overall positive classroom experience rate [100.0% (210/210) vs. 97.1% (204/210), P<0.05], greater course attractiveness (97.6% vs. 92.8%, P<0.05), and significantly higher scores in both regular assessments and final examinations (P<0.01). Conclusion: Incorporating medical dispute case analysis into doctor-patient communication teaching may enhance educational outcomes for medical students. It provides a reference model for promoting high-quality medical teaching in China.

As the number one killer of cancer in the world, lung cancer causes about 2.1 million cases and more than 1.8 million deaths every year. AI technologies such as GPT have brought hope to the prevention and treatment of lung cancer, which can efficiently and accurately empower consultation and diagnosis and treatment suggestions through NLP, promote knowledge sharing and balanced improvement of medical standards, and provide patients with personalized health guidance and psychological comfort. The core technical principle is “four changes”. (1) Change the cleaning data to data selection; (2) Change simple consultation to face the digital expert; (3) Change blind favor to quality control and verification; (4) Change to simply evidence-based and increase medical experience. A controlled study between BAIMGPT and DeepSeek has shown significant advantages in terms of intimacy, security, question understanding, and answer accuracy. Its unique technical architecture and “four innovations” ensure professionalism and accuracy and provide an important reference for the future development of AI-assisted diagnosis systems. BAIMGPT can show significant value in key links such as lung cancer screening, pulmonary nodule consultation and management, diagnosis and staging, treatment plan formulation and postoperative management. Through intelligent image analysis, the diagnosis and treatment process can be optimized, medical costs can be reduced, and resource utilization efficiency can be improved. The successful implementation of BAIMGPT relies on the collaborative work of medical experts, information engineers, data analysis experts and other talents in multiple fields to jointly build a highly applicable and easy-to-operate knowledge system. Through the integration of interdisciplinary knowledge, the efficiency and accuracy of lung cancer screening and evaluation are improved, and the patient consultation experience is optimized. User experience optimization: BAIMGPT has been comprehensively optimized in terms of interface design, intimacy, sense of security, visual empowerment, voice interaction, accessibility and convenience, etc., to ensure that users can quickly get started and generate high-quality reports, enhance patient confidence, and optimize the diagnosis and treatment experience. BAIMGPT has been granted a registered trademark by the State Intellectual Property Office and approved by the Ethics Committee of Zhongshan Hospital. Cooperate with the International Metaverse Medical Association and the China Lung Cancer Prevention and Control Alliance to ensure the legitimacy and professionalism of the technology. At the same time, BAIMGPT strictly adheres to ethical requirements, reduces human bias, and ensures the fairness and transparency of the operation of the system. With the rapid development of AI technology, BAIMGPT will show potential in the field of lung cancer screening, evaluation and diagnosis and treatment. In the future, the efficiency and accuracy of the system can be further improved by combining other imaging technologies, optimizing the accuracy of disease-specific models, and expanding the functions of disease progression prediction. BAIMGPT is expected to become an important force in promoting the development of public health and helping to achieve the important goal of “Healthy China”.

Key Words: artificial intelligence; generative pre-trained transformer; natural language processing; lung cancer screening

Artificial intelligence (AI) is increasingly utilized in precision medicine, with notable applications observed in neuropathology. In glioma diagnostics, histological classification, molecular subtyping, and WHO grading are automated by AI-based platforms, enhancing diagnostic consistency and operational efficiency. Critically, AI predicts prognosis, assesses survival and recurrence risks, and guides personalized treatment strategies. As issues like data silos and “black-box” algorithms are resolved, AI is poised to support decision-making by pathologists and clinicians throughout the clinical workflow of glioma management.

Key Words: glioma; artificial intelligence; neural networks; prognosis