Metabolic dysfunction-associated fatty liver disease （MAFLD） has become one of the most prevalent chronic liver diseases worldwide， posing a serious challenge to public health. In this context， the integration of artificial intelligence （AI）， especially intelligent diagnosis and treatment and digital health interventions based on machine learning， can break through the limitations of traditional methods， realize efficient screening of multi-dimensional data such as key genes， biomarkers， and biochemical metabolites， and achieve revolutionary breakthroughs in risk prediction， subtype identification， and therapeutic effect assessment for MAFLD. This article systematically reviews the ground-breaking application of machine learning models in driving the innovation of clinical diagnosis and precise risk prediction of MAFLD， conducts a comprehensive comparative analysis of digital health practice cases of MAFLD in China and globally， and deeply analyzes their advantages and limitations in terms of research subjects， interventions， and management team. Studies have shown that the deep integration of digital health and long-term management of MAFLD is becoming the key engine driving the transformation of disease management modes towards an intelligent， individualized， and precise era， but there are various ethical and technical issues that need to be addressed urgently.

Under the current legal framework， living will， as an important legal tool for safeguarding patients’ autonomy and dignity， have been widely recognized and implemented in many countries and regions. In China， the promotion of living will also has a solid legal foundation， with their legitimacy reflected in several provisions of the Civil Code of the People’s Republic of China. One of the highlights of the Medical Regulations of the Shenzhen Special Economic Zone （revised in 2022） is the clarification of the legal effect of living will. To ensure that patients’ living will can be accurately implemented at critical moments， the rights and obligations of patients， family members， and healthcare professionals should be clearly defined within the legal framework， and clear guidance should be provided at every stage of implementation.

There is a substantial unmet need for treatments in the field of pediatric rare diseases, and investigator initiated trial(IIT) provide a critical pathway for testing and developing new drugs or treatment strategies. However, healthcare institutions, when conducting such research, must address compliance risks related to project approval, contract management, data protection, and conflict of interest management. This study aims to analyze the particularities and challenges of IIT in pediatric rare diseases, review relevant regulations and regulatory requirements, and provide healthcare institutions with a reference framework for compliance risk management to maximize the benefits of IIT. Based on literature review, analysis of laws and regulations, practical work experience, and frameworks from other institutions, we summarize the unique aspects of pediatric rare disease IIT in terms of participant characteristics, innovative technologies, and organizational structures.On this basis, targeted compliance management recommendations are proposed, which include establishing a risk rating and full-cycle risk monitoring mechanism, a consent and ethical review mechanism tailored to pediatric participants, a robust contract management mechanism, a comprehensive data security management mechanism, and a multidisciplinary team and multi-channel compensation mechanism. The study concludes that healthcare institutions, funders, and other collaborating entities should implement compliance management in line with the characteristics of IIT to ensure the safety and effectiveness of research and facilitate innovation and development in the treatment of pediatric rare diseases.

Objective To explore the diagnosis and treatment experience of malignant peritoneal mesothelioma by analyzing the clinical characteristics of patients with single center peritoneal mesothelioma.Methods Retrospec-tively analyze the clinical data,including age and treatment regimens,of 99 patients with malignant peritoneal mes-othelioma admitted to Peking Union Medical College Hospital from June 2012 to July 2023.Kaplan-Meier for surviv-al analysis and Logistic and Cox regression were applied to assess the impact of factors like age and treatment regi-mens on survival time.Results The median age of diagnosis for 99 patients with peritoneal mesothelioma was 56 years,of which 82 were pathologically diagnosed.Among them,38 cases(51.4%)underwent tumor cell reduction surgery,and 57 cases(77.0%)underwent systemic or local treatment.The median survival time for all patients was 43 years.The age of diagnosis may affect the prognosis of patients with malignant mesothelioma,and those aged≥65 at the time of diagnosis have a worse prognosis.Conclusions Malignant peritoneal mesothelioma is a rare but invasive malignant tumor with a poor prognosis.There is still a need to develop novo drugs and comprehensive treat-ment strategies.

Objective To evaluate the clinical application of CT-guided coaxial needle biopsy in pancreatic puncture-cutting biopsy via adipose tissue space.Methods The clinical data of 30 patients,who received CT-guided pancreatic coaxial needle biopsy via adipose tissue space at the Pingdu Hospital Area of Affiliated Hospital of Qingdao University of China from June 2021 to April 2024,were retrospectively analyzed.The patients included 20 males and 10 females with a mean age of(65.1±8.7)years.The lesions were located in the pancreatic head(n=7),pancreatic neck(n=4),pancreatic body(n=7),pancreatic tail(n=10)and uncinate process(n=2).Under CT guidance,a 17 G co-axial guiding needle was inserted to establish a puncture pass through adipose tissue space,then an 18 G biopsy needle was used to make tissue sampling.A total of 32 procedures of pancreatic puncture-cutting biopsy were performed,two of these patients received two times of biopsy.After sampling,a self-made blocking agent(a mixed liquid of gelatin sponge particles and snake venom hemagglutinin),was used to occlude the puncture channel so as to prevent complications such as bleeding,pancreatic leakage,needle tract implantation metastasis,etc.Results A total of 32 procedures of pancreatic puncture-cutting biopsy with coaxial guiding needle via adipose tissue space were performed,including two patients who received two times of biopsy.The success rate of puncturing was 100％(32/32),the success rate of sampling was 93.8％(30/32),and the accuracy of sampling was 93.8％(30/32).No serious puncture-related complications occurred in all patients.Conclusion CT-guided pancreatic coaxial needle biopsy via adipose tissue space is clinically safe,accurate in sampling and minimally-invasive.This interventional diagnostic technique is worth applying in clinical practice.

Traditional Chinese medicine(TCM)features complex compatibility mechanisms involving multi-component,multi-target,and multi-pathway interactions.This study presents an interpretable graph artificial intelligence(GraphAI)framework to quantify such mechanisms in Chinese herbal formulas(CHFs).A multidimensional TCM knowledge graph(TCM-MKG;https://zenodo.org/records/13763953)was constructed,integrating seven standardized modules:TCM terminology,Chinese patent medicines(CPMs),Chinese herbal pieces(CHPs),pharmacognostic origins(POs),chemical compounds,biological targets,and diseases.A neighbor-diffusion strategy was used to address the sparsity of compound-target associations,increasing target coverage from 12.0％to 98.7％.Graph neural networks(GNNs)with attention mechanisms were applied to 6,080 CHFs,modeled as graphs with CHPs as nodes.To embed domain-specific semantics,virtual nodes medicinal properties,i.e.,therapeutic nature,flavor,and me-ridian tropism,were introduced,enabling interpretable modeling of inter-CHP relationships.The model quantitatively captured classical compatibility roles such as"monarch-minister-assistant-guide",and uncovered TCM etiological types derived from diagnostic and efficacy patterns.Model validation using 215 CHFs used for coronavirus disease 2019(COVID-19)management highlighted Radix Astragali-Rhi-zoma Phragmitis as a high-attention herb pair.Mass spectrometry(MS)and target prediction identified three active compounds,i.e.,methylinissolin-3-O-glucoside,corydalin,and pingbeinine,which converge on pathways such as neuroactive ligand-receptor interaction,xenobiotic response,and neuronal func-tion,supporting their neuroimmune and detoxification potential.Given their high safety and dietary compatibility,this herb pair may offer therapeutic value for managing long COVID-19.All data and code are openly available(https://github.com/ZENGJingqi/GraphAI-for-TCM),providing a scalable and inter-pretable platform for TCM mechanism research and discovery of bioactive herbal constituents.

To address the limitations of traditional minimally invasive thermal ablation technology such as poor conformability, carbonization and electromagnetic radiation, this paper proposes a steam thermal ablation technology that uses saturated steam internal energy to replace the traditional electromagnetic radiation energy. Through the steam thermal ablation system and the steam thermal ablation needle designed based on simulation, the ex vivo pig liver experiments were carried out. The results have the characteristics of the maximum ablation axis ratio (short diameter / long diameter) and non-carbonization with the same type of thermal ablation technology. Based on the near-infrared light, in this paper the curative effect of the reduced scattering coefficient of the steam thermal ablation results was evaluated. The reduced scattering coefficients of the coagulation area all exceeded 16, reaching the completely damaged state, which verified that the steam thermal ablation can effectively inactivate the tumor cells.
Steam ; Animals ; Swine ; Liver Neoplasms/surgery* ; Ablation Techniques/methods* ; Liver/surgery* ; Equipment Design

The minimally invasive thermal ablation technology differs from traditional surgical operations, which requires auxiliary equipment to evaluate ablation results. However, the ultrasound and CT currently used in clinical practice have shortcomings such as artifacts and radiation. Therefore, this paper proposes a design for a minimally invasive thermal ablation evaluation system based on the principle of electrical impedance tomography technology to monitor the ablation range. At the same time, the innovative introduction of a programmable gain feedforward signal as the parameter signal of the multiplier demodulator in the electrical impedance tomography system design can effectively solve the problem of weak signals being submerged in noise and improve imaging accuracy. The system controls the amplitude of the excitation current signal and the acquisition / processing of boundary voltages via an STM32, uploads the collected data to an upper computer, and reconstructs the conductivity distribution using the Newton-Raphson algorithm to map the size of the ablation area. Experimental results show that the system can effectively reflect the size of the microwave ablation area. Under the same minimally invasive ablation parameters, the average imaging errors are 0.6 mm for the long diameter, 0.8 mm for the short diameter, and 1.75% for the axial ratio (long diameter / short diameter), demonstrating high consistency. This verifies the technical potential of electrical impedance tomography in minimally invasive thermal ablation.
Electric Impedance ; Tomography/instrumentation* ; Equipment Design

Traditional Chinese medicine (TCM) features complex compatibility mechanisms involving multi-component, multi-target, and multi-pathway interactions. This study presents an interpretable graph artificial intelligence (GraphAI) framework to quantify such mechanisms in Chinese herbal formulas (CHFs). A multidimensional TCM knowledge graph (TCM-MKG; https://zenodo.org/records/13763953) was constructed, integrating seven standardized modules: TCM terminology, Chinese patent medicines (CPMs), Chinese herbal pieces (CHPs), pharmacognostic origins (POs), chemical compounds, biological targets, and diseases. A neighbor-diffusion strategy was used to address the sparsity of compound-target associations, increasing target coverage from 12.0% to 98.7%. Graph neural networks (GNNs) with attention mechanisms were applied to 6,080 CHFs, modeled as graphs with CHPs as nodes. To embed domain-specific semantics, virtual nodes medicinal properties, i.e., therapeutic nature, flavor, and meridian tropism, were introduced, enabling interpretable modeling of inter-CHP relationships. The model quantitatively captured classical compatibility roles such as "monarch-minister-assistant-guide," and uncovered TCM etiological types derived from diagnostic and efficacy patterns. Model validation using 215 CHFs used for coronavirus disease 2019 (COVID-19) management highlighted Radix Astragali-Rhizoma Phragmitis as a high-attention herb pair. Mass spectrometry (MS) and target prediction identified three active compounds, i.e., methylinissolin-3-O-glucoside, corydalin, and pingbeinine, which converge on pathways such as neuroactive ligand-receptor interaction, xenobiotic response, and neuronal function, supporting their neuroimmune and detoxification potential. Given their high safety and dietary compatibility, this herb pair may offer therapeutic value for managing long COVID-19. All data and code are openly available (https://github.com/ZENGJingqi/GraphAI-for-TCM), providing a scalable and interpretable platform for TCM mechanism research and discovery of bioactive herbal constituents.

OBJECTIVE: To propose a machine learning-based method for predicting the trajectories during manual acupuncture manipulation (MAM), aiming to improve the precision and consistency of acupuncture practitioner' operation and provide the real-time suggestions on MAM error correction. METHODS: Computer vision technology was used to analyze the hand micromotion when holding needle during acupuncture, and provide a three-dimensional coordinate description method of the index finger joints of the holding hand. Focusing on the 4 typical motions of MAM, a machine learning-based MAM trajectory predictive model was designed. By integrating the changes of phalangeal joint angle and hand skeletal information of acupuncture practitioner, the motion trajectory of the index finger joint was predicted accurately. Besides, the roles of machine learning-based MAM trajectory predictive model in the skill transmission of acupuncture manipulation were verified by stratified randomized controlled trial. RESULTS: The performance of MAM trajectory predictive model, based on the long short-term memory network (LSTM), obtained the highest stability and precision, up to 98%. The learning effect was improved when the model applied to the skill transmission of acupuncture manipulation. CONCLUSION The machine learning-based MAM predictive model provides acupuncture practitioner with precise action prediction and feedback. It is valuable and significant for the inheritance and error correction of manual operation of acupuncture.
Humans ; Acupuncture Therapy/instrumentation* ; Machine Learning ; Adult ; Male ; Female