Medical artificial intelligence （AI） is a new type of application formed by the combination of machine learning， computer vision， natural language processing， and other technologies with clinical medical treatment. With the continuous iteration and development of relevant technologies， medical AI has shown great potential in improving the efficiency of diagnosis and treatment， and service quality， but it also increases the possibility of triggering ethical issues. Ethical issues resulting from the clinical application of medical AI were analyzed， including the lack of algorithmic interpretability and transparency of medical AI， leading to information asymmetry and cognitive discrepancies； the concerning status of security and privacy protection of medical data； and the complex and unclear division of responsibilities due to the collaborative participation of multiple subjects in the clinical application of medical AI， resulting in increased difficulty in the identification of medical accidents and clarification of responsibilities. The paper proposed the principles of not harming patients’ interests， physician’s subjectivity， fairness and inclusiveness， and rapid response. It also explored the strategies and implementation paths for responding to the ethical issues of medical AI from multiple perspectives， including standardizing the environment and processes， clarifying responsibility attribution， continuously assessing the impact of data protection， guaranteeing data security， ensuring model transparency and interpretability， carrying out multi-subject collaboration， as well as the principles of being driven by ethical values and adhering to the “human health-centeredness.” It aimed to provide guidance for the healthy development of medical AI， ensuring technological progress while effectively managing and mitigating accompanying ethical risks， thereby promoting the benign development of medical AI technology and better serving the healthcare industry and patients.

OBJECTIVE: To explore the role of CDGSH iron-sulfur domain 2 (CISD2) in patients with sepsis-related myocardial injury (SMI) and its predictive value for 28-day prognosis and myocardial damage through clinical studies and cell experiments. METHODS: A retrospective study was conducted. Adult patients diagnosed with sepsis admitted to the critical care medicine of Third Affiliated Hospital of Qiqihar Medical University from January 2023 to January 2024 were enrolled. The clinical data, laboratory indicators, expression level of CISD2 mRNA in peripheral blood mononuclear cells (PBMC) 24 hours after admission, and 28 days prognosis were collected. Patients were divided into SMI group [left ventricular ejection fraction (LVEF) < 0.50 or LVEF decreased by ≥ 10% from baseline] and sepsis non-myocardial injury group based on LVEF. The expression levels of CISD2 mRNA were compared between the two groups, and the correlation between CISD2 and myocardial injury was analyzed. Patients were divided into the low-expression group (CISD2 mRNA < 0.5 copy/μL) and the high-expression group (CISD2 mRNA ≥ 0.5 copy/μL) based on the expression of CISD2 mRNA, and into the survival group and the death group based on the prognosis at 28 days. The clinical characteristics were analyzed between the groups. Multivariate Logistic regression was used to analyze the independent predictors of 28-day mortality in patients with sepsis. The predictive value of CISD2 for myocardial damage and 28-day prognosis in patients with sepsis were evaluated by using the receiver operator characteristic curve (ROC curve). In addition, in vitro experiments using human AC16 cardiomyocytes was conducted. The cells were divided into control group, lipopolysaccharide (LPS) group, the LPS+transfection group with overexpression of CISD2 plasmid (LPS+p-CISD2 group), and the LPS + transfection group with negative control plasmid (LPS+p-NC group). The mRNA expression of CISD2 in cells were detected by real-time quantitative polymerase chain reaction (RT-qPCR), the protein expression of CISD2 in cells were detected by Western blotting, and the cell viability was determined by cell counting kit-8 (CCK-8). RESULTS: A total of 85 sepsis patients were included, with 32 developing myocardial injury and 53 without myocardial injury. There were 40 cases of low expression of CISD2 and 45 cases of high expression of CISD2. At 28 days, 60 cases survived and 25 cases died. The mRNA expression of CISD2 in the SMI group was significantly lower than that in the sepsis non-myocardial injury group (copy/μL: 0.41±0.09 vs. 0.92±0.13, P < 0.05). CISD2 was significantly correlated with myocardial injury in patients with sepsis (r = 0.729, P < 0.05). The proportion of LVEF < 0.50 (67.50% vs. 11.11%), sequential organ failure score (SOFA: 15.63±2.15 vs. 11.12±1.52), and acute physiology and chronic health evaluation II (APACHEII: 29.49±3.51 vs. 22.41±2.61) in the CISD2 low-expression group were significantly higher than those in the CISD2 high-expression group (all P < 0.05), while there were no significantly differences in other indicators. The Kaplan-Meier survival curve showed that the 28-day survival time of sepsis patients with in the CISD2 low-expression group was significantly shorter than that in the CISD2 high-expression group (Log-rank test: χ ² = 5.601, P < 0.05). The proportion of CISD2 low-expression and the proportion of LVEF < 0.50 in the survival group were both higher than those in the death group (80.00% vs. 33.33%, 64.00% vs. 26.67%, both P < 0.05), while there were no significantly differences in other indicators. Multivariate Logistic regression analysis showed that CIDS2 and LVEF were independent predictive factors for 28-day mortality in patients with sepsis [CIDS2: odds ratio (OR) = 3.400, 95% confidence interval (95%CI) was 1.026-11.264, P = 0.045; LVEF: OR = 2.905, 95%CI was 1.029-8.199, P = 0.044]. ROC curve analysis showed that when CISD2 was expressed at a low level, patients with sepsis were at high risk of death within 28 days and myocardial injury. The sensitivity of CISD2 in predicting the 28-day mortality of patients with sepsis was 80.00%, and the specificity was 66.67%, and the area under the curve (AUC) was 0.733 (95%CI was 0.626-0.823). The sensitivity of CISD2 in predicting myocardial injury in patients with sepsis was 83.87%, the specificity was 74.07%, and the AUC was 0.790 (95%CI was 0.688-0.871). In addition, compared with the control group, the mRNA and protein expressions of CISD2 as well as the cell activity in the LPS group were significantly decreased. The mRNA and protein expressions of CISD2 and the activity of cardiomyocytes transfected with p-CISD2 were significantly increased. CONCLUSIONS CISD2 plays a protective role in sepsis-associated myocardial injury and has good predictive value for 28-day prognosis and myocardial injury.
Humans ; Sepsis/metabolism* ; Prognosis ; Retrospective Studies ; Male ; Female ; Middle Aged ; RNA, Messenger/genetics* ; Aged ; Myocardium/metabolism*

Objective:To develop an intelligent recognition and precise segmentation technique using ultrasonic images，and to enhance diagnostic efficiency and accuracy.Methods:A total of 1，430 patients diagnosed with uterine fibroids through transvaginal ultrasonography at the Maternal and Child Health Hospital of Guangming from November 2020 to October 2024 were retrospectively included. Ultrasonic images were manually annotated by two experienced physicians and reviewed by a senior expert. The Mask DINO deep learning model was used for lesion segmentation，and the segmentation results were optimized using ellipse fitting technology. Model performance was evaluated using the Dice coefficient，intraclass correlation coefficient（ICC），mean absolute error（MAE），and measurement accuracy.Results:In the test set of 286 cases，the average Dice coefficient of model prediction was 0.992，indicating extremely high segmentation accuracy. The average accuracy of lesion identification by the model was 0.909，with 241 correctly identified samples，19 basically correct samples，and 26 incorrect samples. In terms of long and short axis measurements，the ICC of the model's direct predictions were 0.871（short axis）and 0.784（long axis），with MAE of 0.436 cm（short axis）and 0.508 cm（long axis）. After optimization with ellipse fitting，the ICC increased to 0.893（short axis）and 0.866（long axis），and the MAE decreased to 0.191 cm（short axis）and 0.274 cm（long axis），the measurement accuracy improved significantly.Conclusions:The intelligent recognition and precise segmentation technique for uterine fibroids based on ultrasonic images constructed in this study performed excellently in lesion segmentation and measurement，it can significantly improve the efficiency and accuracy of diagnosis.

Objective To evaluate the recognition capability of AI-enabled Cellsee CS-BM1 automatic cell morphology analyzer for pe-ripheral blood smears and its roles in assisting manual classification,and explore the application value of AI system in the diagnosis network of tiered primary medical units.Methods The blood samples which triggered the re-examination rules were collected from six primary medical units,including the Laboratory Department of Shanghai Jiahui International Hospital,and so on,from March to No-vember 2023.The smears of peripheral blood were prepared and AI analyzer was used for pre-classification to evaluate its recognition performance in identifying the samples with abnormal WBC and RBC.The sensitivity,specificity,and accuracy of WBC classification by six junior and intermediate technicians,both with and without AI assistance,were analyzed.Additionally,the roles of the AI system in tiered diagnosis of primary medical units were also evaluated.Results The sensitivity,specificity,and accuracy of AI system in recognizing malignant primitive cells were 92.86％,95.16％,and 95.10％,respectively.The sensitivities of AI system in recognizing immature granulocytes,reactive lymphocytes,and nucleated RBCs were all greater than 90％.The sensitivity of AI system in identif-ying abnormal morphology of RBCs reached 99.59％,along with rapid quantitative analysis for various anomalous types of RBCs.In AI-assisted mode,the sensitivity of recognition for all cell types was improved to varying degrees by junior and intermediate technicians,and the sensitivity for recognizing malignant primitive cells,reactive lymphocytes,and immature granulocytes increased to 58.24％,53.39％,and 62.37％for junior technicians,and to 92.06％,83.24％,and 83.12％for intermediate technicians,respectively.The improvements for junior technicians were particularly significant,with increases of 12.46％,10.61％,and 3.71％for each cell type,respectively.Both groups achieved higher specificity and accuracy.Through AI pre-classification and manual review,a variety of pe-ripheral blood cell-related diseases were accurately diagnosed in the tiered healthcare practice of primary medical units,including 339 cases(11.13％)of red blood cell diseases,5 cases(0.16％)of platelet diseases,2 343 cases(76.90％)of infection-related disea-ses,and 28 cases(0.92％)of malignant hematological diseases.In addition,332 cases(10.90％)which lacked an obvious related cause or required further examinations were identified as well.Conclusion AI pre-classification has demonstrated strong cell recogni-tion capabilities and may assist technicians in improving the sensitivity,specificity,and accuracy of blood cell classification.AI could en-hance the disease-screening capabilities in the tiered diagnosis network of primary medical units,presenting a broad application prospect.

Objective:To develop an intelligent recognition and precise segmentation technique using ultrasonic images，and to enhance diagnostic efficiency and accuracy.Methods:A total of 1，430 patients diagnosed with uterine fibroids through transvaginal ultrasonography at the Maternal and Child Health Hospital of Guangming from November 2020 to October 2024 were retrospectively included. Ultrasonic images were manually annotated by two experienced physicians and reviewed by a senior expert. The Mask DINO deep learning model was used for lesion segmentation，and the segmentation results were optimized using ellipse fitting technology. Model performance was evaluated using the Dice coefficient，intraclass correlation coefficient（ICC），mean absolute error（MAE），and measurement accuracy.Results:In the test set of 286 cases，the average Dice coefficient of model prediction was 0.992，indicating extremely high segmentation accuracy. The average accuracy of lesion identification by the model was 0.909，with 241 correctly identified samples，19 basically correct samples，and 26 incorrect samples. In terms of long and short axis measurements，the ICC of the model's direct predictions were 0.871（short axis）and 0.784（long axis），with MAE of 0.436 cm（short axis）and 0.508 cm（long axis）. After optimization with ellipse fitting，the ICC increased to 0.893（short axis）and 0.866（long axis），and the MAE decreased to 0.191 cm（short axis）and 0.274 cm（long axis），the measurement accuracy improved significantly.Conclusions:The intelligent recognition and precise segmentation technique for uterine fibroids based on ultrasonic images constructed in this study performed excellently in lesion segmentation and measurement，it can significantly improve the efficiency and accuracy of diagnosis.

Objective To evaluate the recognition capability of AI-enabled Cellsee CS-BM1 automatic cell morphology analyzer for pe-ripheral blood smears and its roles in assisting manual classification,and explore the application value of AI system in the diagnosis network of tiered primary medical units.Methods The blood samples which triggered the re-examination rules were collected from six primary medical units,including the Laboratory Department of Shanghai Jiahui International Hospital,and so on,from March to No-vember 2023.The smears of peripheral blood were prepared and AI analyzer was used for pre-classification to evaluate its recognition performance in identifying the samples with abnormal WBC and RBC.The sensitivity,specificity,and accuracy of WBC classification by six junior and intermediate technicians,both with and without AI assistance,were analyzed.Additionally,the roles of the AI system in tiered diagnosis of primary medical units were also evaluated.Results The sensitivity,specificity,and accuracy of AI system in recognizing malignant primitive cells were 92.86％,95.16％,and 95.10％,respectively.The sensitivities of AI system in recognizing immature granulocytes,reactive lymphocytes,and nucleated RBCs were all greater than 90％.The sensitivity of AI system in identif-ying abnormal morphology of RBCs reached 99.59％,along with rapid quantitative analysis for various anomalous types of RBCs.In AI-assisted mode,the sensitivity of recognition for all cell types was improved to varying degrees by junior and intermediate technicians,and the sensitivity for recognizing malignant primitive cells,reactive lymphocytes,and immature granulocytes increased to 58.24％,53.39％,and 62.37％for junior technicians,and to 92.06％,83.24％,and 83.12％for intermediate technicians,respectively.The improvements for junior technicians were particularly significant,with increases of 12.46％,10.61％,and 3.71％for each cell type,respectively.Both groups achieved higher specificity and accuracy.Through AI pre-classification and manual review,a variety of pe-ripheral blood cell-related diseases were accurately diagnosed in the tiered healthcare practice of primary medical units,including 339 cases(11.13％)of red blood cell diseases,5 cases(0.16％)of platelet diseases,2 343 cases(76.90％)of infection-related disea-ses,and 28 cases(0.92％)of malignant hematological diseases.In addition,332 cases(10.90％)which lacked an obvious related cause or required further examinations were identified as well.Conclusion AI pre-classification has demonstrated strong cell recogni-tion capabilities and may assist technicians in improving the sensitivity,specificity,and accuracy of blood cell classification.AI could en-hance the disease-screening capabilities in the tiered diagnosis network of primary medical units,presenting a broad application prospect.

Purpose/Significance To analyze the satisfaction,service effectiveness and willingness to participate in telemedicine services from the perspective of medical staff,and to identify the existing problems.Method/Process The research group conducts an e-lectronic questionnaire survey for medical institutions across the country,and collects a total of 1 524 valid questionnaires.Based on questionnaire survey data,logistic regression models are constructed to analyze the key factors that affect the evaluation and attitude of medical staff.Result/Conclusion The overall satisfaction and service effectiveness of medical staff to telemedicine services are at a high level,and their willingness to participate is strong.It is necessary to improve the telemedicine service system and promote the experience of medical staff by strengthening infrastructure construction,optimizing service process,and perfecting incentive mechanisms.

This study aimed to investigate the prevalence and clinical characteristics of epilepsy in patients with patent foramen ovale (PFO) and the effect of PFO closure on seizures. Patients diagnosed with PFO were recruited and underwent brain magnetic resonance imaging, electrocardiography, transesophageal echocardiography, and transthoracic echocardiography with right ventriculography. In patients with epilepsy, electroencephalography was performed. A total of 110 patients completed the assessment. A chief complaint of chest tightness or palpitations was proportionately higher in patients aged<18 years, whereas headaches and seizures were higher in patients aged≥18 years ( χ2=4.69 ,P<0.05). Comorbid epilepsy was observed in 20.9% of patients with PFO. The age at admission in the epileptic group (14-66(27±14)years) was significantly lower than that in the non-epileptic group (16-81(38±21)years) and that in patients with headache as the chief complaint (16-68(39±12)years) ( t=3.29, P<0.05). The multivariate analysis found no risk factors related to the prognosis of epilepsy. The incidence of epilepsy was significantly higher in patients with PFO than in the general population.

Objective:To analyze the clinical characteristics and medication options of patients with elderly-onset epilepsy and influencing factors for medication efficacy.Methods:A total of 213 patients with elderly-onset epilepsy (age of onset≥65 years) were selected from Epilepsy Outpatient, Department of Neurology, First Medical Center of Chinese PLA General Hospital from February 1999 to March 2023. General data, imaging findings and follow-up results of these patients were collected. Seizure frequencies and types, medication types, and medication efficacy were analyzed retrospectively. According to medication efficacy, these patients were divided into effective anti-seizure medications (ASMs) group and ineffective ASMs group (effective ASMs was defined as having no seizures or seizure reduction>50% at 6 months after medication, and ineffective ASMs as having seizure reduction≤50% or seizure increase. Univariate and multivariate Logistic regression analyses were used to identify the influencing factor for ASMs efficacy.Results:In these 213 patients with elderly-onset epilepsy, 143 (67.1%) were males and 70 (32.9%) were females. Onset age was 70.0 (67.0, 74.5) years, with duration of 12 (4, 32) months. Time from first onset to treatment was 2.0 (1.0, 10.5) months, with that<2 months enjoying the largest proportion ( n=101). MRI/CT in 102 patients indicated potential epileptogenic abnormal structures, such as post-stroke gliosis/encephalomalacia ( n=67) and post-traumatic gliosis/encephalomalacia ( n=13). MRI/CT in 78 patients indicated non-epileptogenic abnormal structures, such as ischemic changes of small and medium vessels ( n=51) and brain atrophy ( n=15). Structural change was the most common cause ( n=160). Sixty-nine patients (32.4%) did not take medicine and 144 (67.6%) took medicine at the visiting; sodium valproate was mostly used ( n=74), followed by levetiracetam ( n=35) and carbamazepine ( n=24). Five patients had sodium valproate combined with levetiracetam, and 4 patients had sodium valproate combined with carbamazepine. Multivariate Logistic regression analysis showed that disease duration and medication combination were independent influencing factors for ASMs efficacy. Conclusion:Structural change is the main cause for elderly-onset epilepsy; medication efficacy is worse in patients with longer disease course and medication combination therapy.

Recently, sponsored by the Science Popularization Department of the China Anti Cancer Association, jointly organized by the Rehabilitation Branch of the China Anti Cancer Association and the Mijian Digital Cancer Patient Course Management Platform, and co-organized by the Science Popularization Special Committee of the China Anti Cancer Association, The "2022 White Paper on the Quality of Life of Chinese Lung Cancer Patients" has been officially released (herein after referred to as the "White Paper"), which mainly elaborates on the basic situation of Chinese lung cancer patients and the medical, social, and economic impacts caused by the disease. This article interprets the White Paper in order to help the public understand the real situation of lung cancer patients and provide important empirical evidence and valuable insights for the diagnosis, treatment, and rehabilitation of lung cancer in China.