Intravitreal anti-vascular endothelial growth factor(anti-VEGF)therapy has been widely used, but the variability in its therapeutic efficacy limits individualized treatment. In recent years, the application of artificial intelligence(AI)has opened up new avenues for personalized treatment response prediction, and its core branches include machine learning(ML)and deep learning(DL). This review systematically retrieved and analyzed 41 relevant studies published up to April 2025. Comprehensive analysis reveals that AI predictive models are evolving from forecasting single endpoints(such as visual acuity or central retinal thickness)to integrating multi-dimensional endpoints(encompassing anatomical, functional, and treatment demand parameters)and generating predictive imaging outputs. In terms of technical approaches, DL models(28 studies, accounting for 68.3%)dominate this field due to their robust image interpretation capabilities, while ML models(10 studies, 24.4%)retain significant value in the analysis of structured clinical data. Cross-disease comparisons indicate that research efforts are most concentrated on age-related macular degeneration(ARMD)and diabetic macular edema(DME), with shared conceptual frameworks for model construction, yet distinct anatomical and functional indicators are prioritized for each disease. Currently, the field confronts several key challenges, including insufficient prospective clinical validation, limited model interpretability(the “black box problem”), and a scarcity of high-quality multi-center datasets. Moving forward, it is imperative to advance real-world validation and develop explainable AI techniques to expedite the clinical translation of these predictive models.

This research study focuses on addressing the limitations of current neuropathic pain(NP)treatments by developing a novel dual-target modulator,E0199,targeting both Nav1.7,Nay1.8,and Nay1.9 and Kv7 channels,a crucial regulator in controlling NP symptoms.The objective of the study was to synthesize a compound capable of modulating these channels to alleviate NP.Through an experimental design involving both in vitro and in vivo methods,E0199 was tested for its efficacy on ion channels and its therapeutic potential in a chronic constriction injury(CCI)mouse model.The results demonstrated that E0199 significantly inhibited Nav1.7,Nav1.8,and Nav1.9 channels with a particularly low half maximal inhibitory concentration(ICs0)for Nay1.9 by promoting sodium channel inactivation,and also effectively increased Kv7.2/73,Kv7.2,and Kv7.5 channels,excluding Kv7.1 by promoting potassium channel acti-vation.This dual action significantly reduced the excitability of dorsal root ganglion neurons and alle-viated pain hypersensitivity in mice at low doses,indicating a potent analgesic effect without affecting heart and skeletal muscle ion channels critically.The safety of E0199 was supported by neurobehavioral evaluations.Conclusively,E0199 represents a ground-breaking approach in NP treatment,showcasing the potential of dual-target small-molecule compounds in providing a more effective and safe thera-peutic option for NP.This study introduces a promising direction for the future development of NP therapeutics.

This research study focuses on addressing the limitations of current neuropathic pain (NP) treatments by developing a novel dual-target modulator, E0199, targeting both Na_V1.7, Na_V1.8, and Na_V1.9 and K_V7 channels, a crucial regulator in controlling NP symptoms. The objective of the study was to synthesize a compound capable of modulating these channels to alleviate NP. Through an experimental design involving both in vitro and in vivo methods, E0199 was tested for its efficacy on ion channels and its therapeutic potential in a chronic constriction injury (CCI) mouse model. The results demonstrated that E0199 significantly inhibited Na_V1.7, Na_V1.8, and Na_V1.9 channels with a particularly low half maximal inhibitory concentration (IC₅₀) for Na_V1.9 by promoting sodium channel inactivation, and also effectively increased K_V7.2/7.3, K_V7.2, and K_V7.5 channels, excluding K_V7.1 by promoting potassium channel activation. This dual action significantly reduced the excitability of dorsal root ganglion neurons and alleviated pain hypersensitivity in mice at low doses, indicating a potent analgesic effect without affecting heart and skeletal muscle ion channels critically. The safety of E0199 was supported by neurobehavioral evaluations. Conclusively, E0199 represents a ground-breaking approach in NP treatment, showcasing the potential of dual-target small-molecule compounds in providing a more effective and safe therapeutic option for NP. This study introduces a promising direction for the future development of NP therapeutics.

BACKGROUND: Changes in N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels may not fully translate into patient-reported health status in patients with heart failure (HF). We aimed to evaluate the correlation between NT-proBNP levels and patient-reported health status changes at one month after discharge of patients, and their associations with risk of death and rehospitalization in patients with acute HF. METHODS: We used data from the China Patient-centered Evaluative Assessment of Cardiac Events Prospective Heart Failure Study (PEACE 5p-HF Study). Patient-reported health status was measured by the 12-item Kansas City Cardiomyopathy Questionnaire (KCCQ-12). Patients who were hospitalized for HF and completed the KCCQ-12 and NT-proBNP tests before and one month after discharge were eligible in our study. We stratified patients into different groups based on NT-proBNP levels (i.e., improved, stable, and deteriorated) and KCCQ-12 scores (i.e., not deteriorated and deteriorated). We also examined the associations of the joint NT-proBNP and KCCQ-12 change with the risk of one-year and four-year clinical outcomes. RESULTS: A total of 2461 patients were included in the analysis. The mean age was 64.06 ± 13.51 years, and 36.37% (895/2461) of the study population were female. Among patients with improved NT-proBNP levels, 115 (10.95%) patients had deteriorated KCCQ-12 scores. The correlation between the change in the KCCQ-12 score and NT-proBNP level was weak ( r2 = 0.002, P = 0.013). Stratification by changes in the KCCQ-12 score revealed subgroups with distinctive risks, such that patients with deteriorated KCCQ-12 scores in any of the NT-proBNP change groups exhibited an increased risk of one-year all-cause death than participants with not deteriorated KCCQ-12 scores in any of the NT-proBNP change groups. Patients with improved NT-proBNP levels and deteriorated KCCQ-12 scores presented greater risks of one-year all-cause death (hazard ratio [HR]: 2.45, 95% confidence interval [CI]: 1.34-4.48) than patients with stable NT-proBNP levels and not deteriorated KCCQ-12 scores (HR [95% CI], 1.77 [1.25-2.53]). CONCLUSIONS: A discrepancy between changes in NT-proBNP levels and KCCQ-12 scores was common. The change in NT-proBNP levels was not sufficient to characterize critical aspects related to HF during one month after discharge of patients. Changes in the KCCQ-12 score exhibit complementary information to NT-proBNP levels for the prediction of clinical outcomes in patients with acute HF. REGISTRATION www.clinicaltrials.gov (No. NCT02878811).
Aged ; Female ; Humans ; Male ; Middle Aged ; Health Status ; Heart Failure/metabolism* ; Natriuretic Peptide, Brain/metabolism* ; Peptide Fragments/metabolism* ; Prospective Studies

Background::Computer-aided detection (CAD) software has been introduced to automatically interpret digital chest X-rays. This study aimed to evaluate the performance of CAD software (JF CXR-1 v3.0, which was developed by a domestic Hi-tech enterprise) in tuberculosis (TB) case finding in China.Methods::In 2019, we conducted an internal evaluation of the performance of JF CXR-1 v3.0 by reading standard images annotated by a panel of experts. In 2020, using the reading results of chest X-rays by a panel of experts as the reference standard, we conducted an on-site prospective study to evaluate the performance of JF CXR-1 v3.0 and local radiologists in TB case finding in 13 township health centers in Zhongmu County, Henan Province.Results::Internal assessment results based on 277 standard images showed that JF CXR-1 v3.0 had a sensitivity of 85.94% (95% confidence interval [CI]: 77.42%, 94.45%) and a specificity of 74.65% (95% CI: 68.81%, 80.49%) to distinguish active TB from other imaging conditions. In the on-site evaluation phase, images from 3705 outpatients who underwent chest X-ray detection were read by JF CXR-1 v3.0 and local radiologists in parallel. The imaging diagnosis of local radiologists for active TB had a sensitivity of 32.89% (95% CI: 22.33%, 43.46%) and a specificity of 99.28% (95% CI: 99.01%, 99.56%), while JF CXR-1 v3.0 showed a significantly higher sensitivity of 92.11% (95% CI: 86.04%, 98.17%) ( p < 0.05) and maintained high specificity at 94.54% (95% CI: 93.81%, 95.28%). Conclusions::CAD software could play a positive role in improving the TB case finding capability of township health centers.

Background::Computer-aided detection (CAD) software has been introduced to automatically interpret digital chest X-rays. This study aimed to evaluate the performance of CAD software (JF CXR-1 v3.0, which was developed by a domestic Hi-tech enterprise) in tuberculosis (TB) case finding in China.Methods::In 2019, we conducted an internal evaluation of the performance of JF CXR-1 v3.0 by reading standard images annotated by a panel of experts. In 2020, using the reading results of chest X-rays by a panel of experts as the reference standard, we conducted an on-site prospective study to evaluate the performance of JF CXR-1 v3.0 and local radiologists in TB case finding in 13 township health centers in Zhongmu County, Henan Province.Results::Internal assessment results based on 277 standard images showed that JF CXR-1 v3.0 had a sensitivity of 85.94% (95% confidence interval [CI]: 77.42%, 94.45%) and a specificity of 74.65% (95% CI: 68.81%, 80.49%) to distinguish active TB from other imaging conditions. In the on-site evaluation phase, images from 3705 outpatients who underwent chest X-ray detection were read by JF CXR-1 v3.0 and local radiologists in parallel. The imaging diagnosis of local radiologists for active TB had a sensitivity of 32.89% (95% CI: 22.33%, 43.46%) and a specificity of 99.28% (95% CI: 99.01%, 99.56%), while JF CXR-1 v3.0 showed a significantly higher sensitivity of 92.11% (95% CI: 86.04%, 98.17%) ( p < 0.05) and maintained high specificity at 94.54% (95% CI: 93.81%, 95.28%). Conclusions::CAD software could play a positive role in improving the TB case finding capability of township health centers.

Plastic surgery is characterized by high degree of specialization, a broadscope of diseases inclusion and rapid advancements in knowledge. It is closely related to many disciplines, and there is great heterogeneity among different patients, which requires comprehensive abilities of doctors. However, plastic surgery teaching in China is currently marked by a short training cycle, a uniform training mode, and students’ time constraints and heavy tasks. Chat generative pre-trained transformer (ChatGPT), a large-scale language model introduced by the artificial intelligence company OpenAI, can use deep learning technology to generate natural language texts, understand the context of a conversation and then generate responses similar to those of humans, and is widely used in various industries, including the medical field. This review began by identifying the current challenges in plastic surgery teaching, introduced potential applications of ChatGPT in the field, and outlined its advantages. It also discussed the limitations and potential future directions for its development.

Plastic surgery is characterized by high degree of specialization, a broadscope of diseases inclusion and rapid advancements in knowledge. It is closely related to many disciplines, and there is great heterogeneity among different patients, which requires comprehensive abilities of doctors. However, plastic surgery teaching in China is currently marked by a short training cycle, a uniform training mode, and students’ time constraints and heavy tasks. Chat generative pre-trained transformer (ChatGPT), a large-scale language model introduced by the artificial intelligence company OpenAI, can use deep learning technology to generate natural language texts, understand the context of a conversation and then generate responses similar to those of humans, and is widely used in various industries, including the medical field. This review began by identifying the current challenges in plastic surgery teaching, introduced potential applications of ChatGPT in the field, and outlined its advantages. It also discussed the limitations and potential future directions for its development.

As the special subject of the applicant for registration of medical device, the research and development institutions have insufficient conditions and abilities to become medical device registrants, and there are certain difficulties in the actual registration application process, such as not clearing the certification path for the research and development institutions to hold the certificate. In view of the existing problems, by comparing the path of medicine research and development institutions to become medical device registrants and combining with the actual medical device industry to give relevant suggestions, including improving quality management over the whole life cycle of medical devices, quality and safety responsibility ability of research and development institutions, establishing the registration and certification path of research and development institutions, supporting laws and regulations, etc., so as to ensure that the research and development institutions become medical device registrants successfully.
Research ; Certification

ObjectiveTo investigate the effect of intermittent theta burst stimulation (iTBS) of the multi-target cerebral cortex after stroke on functional recovery of the upper limb of the hemiplegic side. MethodsFrom November, 2019 to August, 2020, 40 stroke patients in Gansu Provine Hospital Rehabilitation Center were included and randomly divided into single-target stimulation group (n = 20) and multiple-target stimulation group (n = 20). Both groups underwent basic neurorehabilitation drug therapy and conventional rehabilitation exercises. The single-target stimulation group received repetitive transcranial magnetic stimulation (rTMS) (iTBS mode) only in the primary motor cortex (M1) of the affected side. The multi-target stimulation group received rTMS (iTBS mode) in the cerebellar cortex of the healthy brain and M1 of the affected side, once a day, six days a week, for four weeks. Before and after treatment, the scores of Fugl-Meyer Assessment-Upper Extremities (FMA-UE), Action Research Arm Test (ARAT) and modified Barthel Index (MBI), and the latency and amplitude of somatosensory-evoked potentials N20 were compared. ResultsNo serious adverse reaction occurred during treatment. After treatment, the scores of FMA-UE, MBI and ARAT, and the amplitude and latency of N20 improved in both groups (|t| > 3.478, |Z| > 2.243, P < 0.05); and the scores of FMA-UE and ARAT, and the amplitude of N20 were better in the multiple-target stimulation group than in the single-target stimulation group (t > 2.939, Z = -2.697, P < 0.01). ConclusionMulti-target stimulation is superior to single-target stimulation for improving upper limb motor function and N20 amplitude in the hemiplegics after stroke.