Traditional Chinese medicine （TCM） diagnostics is a discipline that studies the basic theories and fundamental skills of diagnostic methods， disease diagnosis， and differentiation in accordance with the theories of TCM. The artificial intelligence （AI） technology has gained remarkable achievements in the intelligentization of the four diagnostic methods in TCM and the standardization of differentiation and diagnosis. However， it still faces many challenges. The standardization of clinical data collection is difficult， and the data quality is uneven， which affects the usability of the data. The integration of the four diagnostic information is insufficient. Most instruments can only collect data from a single diagnostic method， lacking overall integrity. The scientific nature of the diagnostic model needs to be improved. The existing models lack dynamics and the reasoning logic of TCM differentiation. The accuracy of intelligent methods needs to be improved， and the existing evaluation indicators cannot fully reflect the practical application effect of the model. Furthermore， the relevant laws and regulations are still not perfect， and data security and patient privacy lack guarantees. The cultivation of compound talents is insufficient， and there is a lack of interdisciplinary talents who are proficient in both TCM and AI. On this basis， this paper expounded on the current development status， difficulties， and bottlenecks of AI in TCM diagnosis and then explored the development trend of AI in the field of TCM diagnosis. It proposed solutions such as optimizing the data collection process， constructing multimodal diagnostic models， facilitating multi-disciplinary exchanges and cooperation， improving laws and regulations， and cultivating compound talents. It is hoped that modern， standardized， normalized， and intelligent TCM diagnosis can be further promoted， thereby providing new impetus and methods for the inheritance and innovation of TCM.

Objective: To develop a RHCE genotyping assay based on kompetitive allele-specific PCR (KASP) and assess its clinical accuracy for RhCE blood group determination. Methods: KASP primers were designed to interrogate three RHCE loci: the 109 bp insertion/deletion in intron 2, c. 307T>C, and c. 676C>G. A total of 1 194 RhD-positive inpatients from Chengdu were typed by both KASP genotyping and manual tube serology. Discordant samples (n=10) were retested by both methods and further resolved by Sanger sequencing. An additional 377 cases were tested for the c. 48C>G locus to evaluate the predictive accuracy of individual loci and combined locus testing for RhC antigen. Results: Genotyping concordance with serology was 100.0% for both the c. 676C>G locus (RhE/Rhe) and the c. 307T>C locus (Rhc). For RhC prediction using the 109 bp insertion, overall accuracy was 99.7% (1 191/1 194); the 3 discordant cases were confirmed by Sanger sequencing to be false negatives attributable to 109 bp deletion in intron 2. Testing the c. 48C>G allele for RhC prediction yielded 7 false positives, with an accuracy of 98.1% (370/377). RhC antigen status was determined by combining the 109 bp insertion and the c. 48C allele. After excluding 10 samples with inconsistent results between the two loci, the accuracy reached 100% in the remaining 367 samples. When both loci were applied in combination, accuracy reached 100% in the 367 cases with concordant results. Among the 1 194 patients, CCee (45.8%) and CcEe (31.7%) were the most common RhCE phenotypes. The e antigen had the highest positivity rate (92.2%), and the Ce haplotype was the most frequent (66.9%). Conclusion: The KASP-based RHCE genotyping method achieves high accuracy for clinical RhCE typing. Combining the 109 bp insertion/deletion with the c. 48C allele significantly improves RhC antigen prediction compared with either locus alone. This method was applied to RhCE genotyping of 1 194 RhD-positive inpatients in Chengdu, providing local RhCE phenotype and haplotype distribution data to support RhCE-matched transfusion practice.

OBJECTIVE To explore the prevalence of Ureaplasma urealyticum and Mycoplasma hominis among the patients with chronic cervicitis(CC)and observe their drug resistance to commonly used antibiotics.METHODS A total of 91 patients with CC who were treated in gynecology department of Women and Children's Hospital Affiliated to Ningbo University from Jan.2022 to Jun.2024 were assigned as the CC group,meanwhile,91 healthy women who received physical examination were chosen as the control group.The genital tract secretions were collected from all of the research subjects for the culture of U.urealyticum and M.hominis and drug suscep-tibility testing.The isolation rates of U.urealyticum,M.hominis and U.urealyticum plus M.hominis were com-pared between the two groups.The isolation rates of U.urealyticum and M.hominis were compared among the different age groups of CC patients.The drug susceptibility testing of U.urealyticu m and M.hominis for doxycyc-line(DOX),josamycin(JOS),ofloxacin(OFL),clarithromycin(CLA),erythromycin(ERY),tetracycline(TET),azithromycin(AZI)and pristinamycin(PTN)were observed.RESULTS Totally 75(82.41％)genital tract secretion samples tested positive for Mycoplasma among the 91 samples,37 detected with U.urealyticum,25 were M.hominis,and 13 were U.urealyticum plus M.hominis.The isolation rates of U.urealyticum,M.hominis and U.urealyticum plus M.hominis of the CC group were 40.66％,24.47％and 14.29％,respective-ly,higher than 8.79％,4.40％and 5.49％of the control group(P＜0.05).The total detection rate of U.urealyti-cum,M.hominis and U.urealyticum plus M.hominis was higher among the CC patients aged between 20 and 40 years old than among the CC patients aged more than 40 years old(P＜0.05).The U.urealyticum strains from the positive specimens of the CC patients were highly sensitive to CL A and DOS but were resistant to OFL,CIP and PTN;the M.honinis and U.urealyticum plus M.hominis strains were sensitive to JOS and DOX but were resistant to OFL and CIP.CONCLUSIONS The detection rates of U.urealyticum plus M.hominis are higher a-mong the CC patients than among the normal population.The isolated U.urealyticum and M.hominis strains are highly resistant to quinolones and aminoglycosides.It is necessary for the hospital to empirically choose sensitive antibiotics based on the result of drug susceptibility testing.

Objective:To explore the interrelationship between brain functional networks and features in functional magnetic resonance imaging(fMRI)of patients with Alzheimer's disease(AD),and to construct mixed-function networks(MFN),and apply them in machine learning classification models to improve the accuracy of AD classification.Methods:102 AD patients and 227 healthy subjects in the Alzheimer's Neuroimaging Initiative(ADNI)dataset were retrospectively analyzed.The partial correlation brain network of the blood oxygen level dependent(BOLD)signal was calculated and fused with low-frequency wave amplitude(ALFF),fractional low-frequency wave amplitude(fALFF)and local consistency(ReHo)features to construct MFN.Network topology parameters were extracted,and a variety of machine learning classification models were constructed based on MFN topological parameters,accuracy,precision,recall and area under the curve(AUC)were used to evaluate the predictive efficiency of the models.Results:By constructed MFN and calculated intra group to inter group ratio(IIGR),35 features could be obtained from ALFF,fALFF and ReHo feature topological parameter analysis,after rank sum test and FDR correction,there were statistical differences among 28 features(P＜0.05).The classification results show that,all the five classifiers have high classification performance on the test data set.The accuracy,precision and recall rates of random forest(RF),adaptive lifting algorithm(AdaBoost),guided aggregation algorithm(Bagging)and support vector machine(SVM)were all 99.7％,and the AUC values were up to 100％,99.5％,99.1％and 99.5％,respectively.The accuracy(98.5％),precision(98.5％),recall(98.5％),and AUC(99.1％)of the multi-layer perceptron(MLP)were slightly lower than other models,but remained excellent.It was worth noting that RF has the highest AUC value of all models at 100.0％,while Bagging has the lowest AUC value(99.1％)in the integrated approach.The results of performance comparison show that,MFN classification model can significantly improve the recognition and classification of AD disease,and greatly improve the performance of various indicators of the classifier.The results showed that,MFN classification model was superior to intelligent classification based fusion,DBN-based multitask learning,PVT-TSVM,unsupervised learning and clustering,SVM and SVM of degree 3 polynomial kernel function in key indicators such as accuracy(99.13％),AUC(99.42％),recall rate(99.46％)and specificity(99.42％)with plasma proteins,machine learning algorithms.It was further proved that MFN classification model has good generalization ability and robustness in AD disease classification.Conclusion:The AD classification model constructed based on brain mixed function network topology parameters and machine learning can improve the accuracy of AD classification.

Kidney transplantation is the best renal replacement therapy for patients with end-stage renal disease. However, it faces significant challenges due to a critical shortage of donor organs and the underutilization of expanded standard donor (ESD) kidneys.Dual kidney transplantation can increase the utilization of expanded standard donor kidneys and enlarge the donor pool, which is an effective solution to deal with kidney shortage. This review provides a systematic presentation of the current status of research on allocation and recipient selection, surgical technique, complications, postoperative efficacy and immunosuppression protocols for adult dual kidney transplantation, with the aim of providing assistance in clinical practice.

Objective:To summarize the current situation and shortcomings of China's health technology innovation policies,and provide reference for policy formulation and improvement.Methods:Text mining was used to sort out 24 policy documents related to health technology innovation issued by the national and provincial levels since the 13th Five Year Plan period.A PMC index evaluation model for health technology innovation policies was established,and a quantitative analysis of health technology innovation policies was conducted through 9 primary indicators and 43 secondary indicators.Results:Among the 24 policies,2 were rated as perfect,8 were rated as excellent,and 14 were rated as acceptable.Conclusions and Suggestions:China's policies on health and medical science and technology innovation have been basically improved.They can be further refined by focusing on core and key technologies,emphasizing clinical research and transformation,and advancing digital and intelligent strategies.

OBJECTIVE To explore the prevalence of Ureaplasma urealyticum and Mycoplasma hominis among the patients with chronic cervicitis(CC)and observe their drug resistance to commonly used antibiotics.METHODS A total of 91 patients with CC who were treated in gynecology department of Women and Children's Hospital Affiliated to Ningbo University from Jan.2022 to Jun.2024 were assigned as the CC group,meanwhile,91 healthy women who received physical examination were chosen as the control group.The genital tract secretions were collected from all of the research subjects for the culture of U.urealyticum and M.hominis and drug suscep-tibility testing.The isolation rates of U.urealyticum,M.hominis and U.urealyticum plus M.hominis were com-pared between the two groups.The isolation rates of U.urealyticum and M.hominis were compared among the different age groups of CC patients.The drug susceptibility testing of U.urealyticu m and M.hominis for doxycyc-line(DOX),josamycin(JOS),ofloxacin(OFL),clarithromycin(CLA),erythromycin(ERY),tetracycline(TET),azithromycin(AZI)and pristinamycin(PTN)were observed.RESULTS Totally 75(82.41％)genital tract secretion samples tested positive for Mycoplasma among the 91 samples,37 detected with U.urealyticum,25 were M.hominis,and 13 were U.urealyticum plus M.hominis.The isolation rates of U.urealyticum,M.hominis and U.urealyticum plus M.hominis of the CC group were 40.66％,24.47％and 14.29％,respective-ly,higher than 8.79％,4.40％and 5.49％of the control group(P＜0.05).The total detection rate of U.urealyti-cum,M.hominis and U.urealyticum plus M.hominis was higher among the CC patients aged between 20 and 40 years old than among the CC patients aged more than 40 years old(P＜0.05).The U.urealyticum strains from the positive specimens of the CC patients were highly sensitive to CL A and DOS but were resistant to OFL,CIP and PTN;the M.honinis and U.urealyticum plus M.hominis strains were sensitive to JOS and DOX but were resistant to OFL and CIP.CONCLUSIONS The detection rates of U.urealyticum plus M.hominis are higher a-mong the CC patients than among the normal population.The isolated U.urealyticum and M.hominis strains are highly resistant to quinolones and aminoglycosides.It is necessary for the hospital to empirically choose sensitive antibiotics based on the result of drug susceptibility testing.

Objective:To summarize the current situation and shortcomings of China's health technology innovation policies,and provide reference for policy formulation and improvement.Methods:Text mining was used to sort out 24 policy documents related to health technology innovation issued by the national and provincial levels since the 13th Five Year Plan period.A PMC index evaluation model for health technology innovation policies was established,and a quantitative analysis of health technology innovation policies was conducted through 9 primary indicators and 43 secondary indicators.Results:Among the 24 policies,2 were rated as perfect,8 were rated as excellent,and 14 were rated as acceptable.Conclusions and Suggestions:China's policies on health and medical science and technology innovation have been basically improved.They can be further refined by focusing on core and key technologies,emphasizing clinical research and transformation,and advancing digital and intelligent strategies.

Objective:To explore the interrelationship between brain functional networks and features in functional magnetic resonance imaging(fMRI)of patients with Alzheimer's disease(AD),and to construct mixed-function networks(MFN),and apply them in machine learning classification models to improve the accuracy of AD classification.Methods:102 AD patients and 227 healthy subjects in the Alzheimer's Neuroimaging Initiative(ADNI)dataset were retrospectively analyzed.The partial correlation brain network of the blood oxygen level dependent(BOLD)signal was calculated and fused with low-frequency wave amplitude(ALFF),fractional low-frequency wave amplitude(fALFF)and local consistency(ReHo)features to construct MFN.Network topology parameters were extracted,and a variety of machine learning classification models were constructed based on MFN topological parameters,accuracy,precision,recall and area under the curve(AUC)were used to evaluate the predictive efficiency of the models.Results:By constructed MFN and calculated intra group to inter group ratio(IIGR),35 features could be obtained from ALFF,fALFF and ReHo feature topological parameter analysis,after rank sum test and FDR correction,there were statistical differences among 28 features(P＜0.05).The classification results show that,all the five classifiers have high classification performance on the test data set.The accuracy,precision and recall rates of random forest(RF),adaptive lifting algorithm(AdaBoost),guided aggregation algorithm(Bagging)and support vector machine(SVM)were all 99.7％,and the AUC values were up to 100％,99.5％,99.1％and 99.5％,respectively.The accuracy(98.5％),precision(98.5％),recall(98.5％),and AUC(99.1％)of the multi-layer perceptron(MLP)were slightly lower than other models,but remained excellent.It was worth noting that RF has the highest AUC value of all models at 100.0％,while Bagging has the lowest AUC value(99.1％)in the integrated approach.The results of performance comparison show that,MFN classification model can significantly improve the recognition and classification of AD disease,and greatly improve the performance of various indicators of the classifier.The results showed that,MFN classification model was superior to intelligent classification based fusion,DBN-based multitask learning,PVT-TSVM,unsupervised learning and clustering,SVM and SVM of degree 3 polynomial kernel function in key indicators such as accuracy(99.13％),AUC(99.42％),recall rate(99.46％)and specificity(99.42％)with plasma proteins,machine learning algorithms.It was further proved that MFN classification model has good generalization ability and robustness in AD disease classification.Conclusion:The AD classification model constructed based on brain mixed function network topology parameters and machine learning can improve the accuracy of AD classification.

The exoskeleton robot for lower limb medical rehabilitation in foreign countries and China was introduced in terms of the research status,structure and working principle,and analysis was carried out over its key technologies.It's pointed out the exoskeleton robot for lower limb medical rehabilitation would be enhanced in energy endurance,safety and comfort,individualized and intelligent control,modularity and lightweight design.[Chinese Medical Equipment Journal,2025,46(1):88-100]