Objective To explore an artificial intelligence (AI)-based method for automated hand hygiene monitoring and to compare the effectiveness of three algorithms (UniFormerV2, TDN, C3D) in recognizing hand hygiene steps in surgical settings, thereby aiding hospital infection control. Methods From April to October 2024, we non-invasively collected 641 video recordings of healthcare staff performing hand hygiene at four-bay scrub sinks in two tertiary hospitals using overhead HD cameras. The dataset was annotated by five trained experts for model training and validation. Results Following training on 385 samples, internal validation (n=119) showed the C3D model achieved 81% accuracy, 87% recall, and an 83% F1-score. The TDN model achieved 93%, 91%, and 92% for the same metrics. The UniFormerV2 model outperformed both, with an accuracy, recall, and F1-score of 93%—an improvement of over 10 percentage points compared to traditional CNNs (TDN, C3D). It also achieved an 84% accuracy in external validation, demonstrating strong generalization. Conclusion The UniFormerV2 model is more accurate than CNN-based models for hand hygiene step recognition and shows robust performance in external validation. It presents a viable tool for healthcare facilities to enhance hand hygiene management, ultimately improving medical quality and patient safety.

Houpo Sanwutang， included in the Catalogue of Ancient Classical Prescriptions （Second Batch）， was first recorded in the Synopsis of Golden Chamber written by ZHANG Zhongjing from the Eastern Han dynasty and was modified by successive generations of medical experts. A total of 37 pieces of effective data involving 37 ancient Chinese medical books were retrieved from different databases. Through literature mining， statistical analysis， and data processing， combined with modern articles， this study employed bibliometrics to investigate the historical origin， composition， decoction methods， clinical application， and other key information. The results showed that the medicinal origin of Houpo Sanwutang was clearly documented in classic books. Based on the conversion of the measurements from the Han Dynasty， it is recommended that 110.4 g Magnolia Officinalis Cortex， 55.2 g Rhei Radix et Rhizoma， and 72 g Aurantii Fructus Immaturus should be taken. Magnolia Officinalis Cortex and Aurantii Fructus Immaturus should be decocted with 2 400 mL water first， and 1 000 mL should be taken from the decocted liquid. Following this， Rhei Radix et Rhizoma should be added for further decoction， and then 600 mL should be taken from the decocted liquid. A single dose of administration is 200 mL， and the medication can be stopped when patients restore smooth bowel movement. Houpo Sanwutang has the effect of moving Qi， relieving stuffiness and fullness， removing food stagnation， and regulating bowels. It can be used in treating abdominal distending pain， guarding， constipation， and other diseases with the pathogenesis of stagnated heat and stagnated Qi in the stomach. The above results provide reference for the future development and research of Houpo Sanwutang.

OBJECTIVE To investigate the induction of apoptosis in hepatocellular carcinoma cells by polyphyllin 9 （PP9） through the regulation of the Fas/Fas ligand （FasL） signaling pathway， and its inhibitory effect on the growth of transplanted tumor in nude mice. METHODS Based on the screening of cell lines and intervention conditions， HepG2 cells were selected as the experimental subject to investigate the effects of 2 μmol/L and 4 μmol/L PP9 treatment on cell colony formation activity， apoptosis rate， as well as the protein expressions of Fas， FasL， cleaved caspase-8 and cleaved caspase-3. Additionally， Fas inhibitor KR- 33493 was introduced to investigate the underlying mechanism of PP9’s anti-hepatocellular carcinoma activity. Using HepG2 cell tumor-bearing nude mice model as the object， and 5-fluorouracil （20 mg/kg） as the positive control， the effects of 10 mg/kg PP9 on tumor volume， tumor mass， and the protein expressions of the nuclear proliferation-associated antigen Ki-67 and cleaved caspase-3 in tumor-bearing nude mice were investigated. RESULTS Compared with the control group， 2， 4 μmol/L PP9 significantly decreased the number of clones and the clone formation rate of cells， but significantly increased the apoptosis rate， the protein expressions of Fas， FasL， cleaved caspase-8 and cleaved caspase-3 （P＜0.05 or P＜0.01）. However， the combination of Fas inhibitor KR-33493 could significantly reverse the effect of PP9 on the up-regulation of proteins related to the Fas/FasL signaling pathway （P＜0.01）. Compared with the control group， the tumor volume （on day 27）， mass and protein expression of Ki- 67 in nude mice of the PP9 group were significantly decreased， while the protein expression of cleaved caspase-3 was significantly increased （P＜0.01）. CONCLUSIONS PP9 can induce apoptosis of HepG2 cells by activating the Fas/FasL signaling pathway. Meanwhile， PP9 can also effectively inhibit the growth of transplanted tumors in nude mice.

OBJECTIVE To investigate the effectiveness of resident clinical pharmacist-led medication therapy management （MTM） model in geriatric cardiology departments， and provide reference for optimizing resident pharmaceutical services. METHODS A retrospective cohort study was conducted， incorporating data from inpatients admitted to the Department of Cardiovascular Medicine in the Geriatric Medical Center of our hospital during March to August 2023 （conventional group， n＝ 903） and the same period in 2024 （MTM group， n＝963）. The conventional group received only standard pharmaceutical services （including prospective prescription review and retrospective order evaluation）， while the MTM group received additional resident clinical pharmacist-led interventions-such as medication reconciliation， personalized therapeutic drug monitoring （TDM）， standardized intravenous infusion management， and a four-stage closed-loop monitoring process-based on conventional care. The effectiveness of the MTM model was evaluated by comparing the primary outcome measures （e.g.， intravenous infusion rate， TDM target attainment rate） and secondary outcome measures [e.g.， incidence of drug-drug interactions （DDIs）， incidence of grade 3 or higher acute kidney injury， average length of hospital stay， cholesterol， and medication cost per capita] between the two groups. RESULTS Compared with the conventional group， in terms of primary outcome indexes： both the overall intravenous infusion rate and the use rate of acid-suppressive injection were significantly lowered in the MTM group （P＜0.05）； serum concentration target attainment rates for digoxin and vancomycin were increased significantly （P＜0.05）. For secondary outcome indexes， the MTM group exhibited significant decreases in the work incidence of grade 3 or higher acute kidney injury， the incidence of DDIs， the rate of patients leaving the hospital against medical advice， alanine amino-transferase， aspartate transferase and the per capita total medication cost （P＜0.05）. Additionally， there was a notable increase in the creatinine， estimated glomerular filtration rate and a significant shortening of the per capita length of hospital stay （P＜0.05）. CONCLUSIONS The resident clinical pharmacist-led MTM model can significantly optimize medication therapy processes， enhance medication safety and cost-effectiveness， thus playing a positive role in promoting rational drug use and improving patient outcomes.

OBJECTIVE To construct ensemble learning models for predicting high-use days of budesonide based on meteorological factors， thereby providing reference for hospital pharmacy management. METHODS Meteorological data for 2024 and outpatient budesonide usage data from the jurisdiction of Sanming Hospital of Integrated Traditional Chinese and Western Medicine were collected. High-use days were defined as the 75th percentile of outpatient budesonide usage， and a corresponding dataset was established. The prediction task was formulated as a classification problem， and three ensemble learning models were developed： Random Forest， Extreme Gradient Boosting （XGBoost）， and Histogram-based Gradient Boosting Classifier. Model performance was evaluated using accuracy， precision， recall， F1-score， and log-loss. Model interpretability was analyzed using Shapley Additive Explanations （SHAP）. RESULTS The Histogram-based Gradient Boosting Classifier achieved the best performance （accuracy＝0.75， F1-score＝0.48）， followed by XGBoost （accuracy＝0.74， F1-score＝0.43） and Random Forest （accuracy＝0.72， F1-score＝0.22）. SHAP results suggested that the prediction results of the last two models have the highest correction. CONCLUSIONS Ensemble learning models can effectively predict high-use days of budesonide， with the Histogram- based Gradient Boosting Classifier demonstrating the best predictive performance. Low temperature， high humidity， and low atmospheric pressure show significant positive impacts on the prediction of daily budesonide usage.

Objective To assess the health risks of gaseous pollutants in the indoor air of hair and beauty salons in Jinan, and to provide technical support for strengthening the hygiene management of hair and beauty salons in Jinan and promoting the improvement of conditions. Methods Every year, indoor air samples were collected from 10-16 selected hair salons and beauty salons in Jinan, and relevant information on practitioners was also collected. According to the ^“Technical Guidelines for Environmental Health Risk Assessment of Chemicals^”, an assessment was conducted on the carcinogenic and non-carcinogenic risks of inhalation pathways of gaseous pollutants in the indoor air of hair salons and beauty salons. Results Benzene, toluene, xylene, formaldehyde, and ammonia were detected in the indoor air of hair salons and beauty salons. Formaldehyde, benzene, and ammonia all exceeded the standard in hair salons and beauty salons. The median risk values of formaldehyde and benzene for carcinogenesis in hair salons and beauty salons were both greater than 10^-6, with maximum values higher than 10^-4. The median chronic non-carcinogenic risk value of formaldehyde in the indoor air of hair salons and beauty salons was greater than 1. The median chronic non-carcinogenic risk values for benzene and ammonia were both less than 1, but the maximum risk value was greater than 1. Conclusion Benzene and formaldehyde in the indoor air of hair salons and beauty salons in Jinan City have carcinogenic and non-carcinogenic risks, while ammonia has non-carcinogenic risks, which should be paid attention to.

AIM: To analyze the reasons and clinical outcomes of intraocular lens(IOL)exchange.METHODS:This retrospective case series study included 47 patients(53 eyes)who underwent IOL exchange surgery at the Department of Ophthalmology, the First Affiliated Hospital of Zhengzhou University, between April 2020 and May 2024, aged from 5 to 87(44.8±3.6)years old. Patients' demographics, surgical indications, surgical techniques, preoperative and postoperative uncorrected visual acuity(UCVA)and best-corrected visual acuity(BCVA), as well as postoperative complications were recorded.RESULTS:The reasons for IOL exchange included refractive error(18 cases, 23 eyes, 43%), IOL dislocation(12 cases, 13 eyes, 25%), IOL opacification(12 cases, 12 eyes, 23%), neuroadaptation failure(3 cases, 3 eyes, 6%), and patient dissatisfaction with visual quality(2 cases, 2 eyes, 4%). The surgical techniques for IOL exchange included in-the-bag IOL fixation(16 eyes, 30%), ciliary sulcus fixation(27 eyes, 51%), and scleral suture fixation(10 eyes, 19%). There was statistical significant difference between preoperative UCVA(LogMAR)and UCVA(LogMAR)at 1 d postoperatively(1.03±0.64 vs 0.50±0.46, P<0.05), and there was statistical significant difference between preoperative BCVA(LogMAR)and BCVA(LogMAR)at 1 mo postoperatively(0.41±0.37 vs 0.17±0.21, P<0.05). Postoperative complications included posterior capsule opacification in 2 eyes and IOL dislocation in 1 eye.CONCLUSION:Refractive error, IOL dislocation, and IOL opacification were the three most common reasons for IOL exchange. Although less frequent, factors such as neuroadaptation issues associated with multifocal IOLs and patient-reported visual quality dissatisfaction reflect growing expectations for improved visual outcomes. IOL exchange surgery, though technically challenging, demonstrates favorable clinical efficacy and a low complication rate, representing an effective intervention for managing postoperative IOL-related issues following cataract surgery.

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

OBJECTIVE: To investigate the effect of manual acupuncture manipulations (MAMs) on subcutaneous muscle tissue, by developing quantitative models of "lifting and thrusting" and "twisting and rotating", based on machine learning techniques. METHODS: A depth camera was used to capture the acupuncture operator's hand movements during "lifting and thrusting" and "twisting and rotating" of needle. Simultaneously, the ultrasound imaging was employed to record the muscle tissue responses of the participants. Amplitude and angular features were extracted from the movement data of operators, and muscle fascicle slope features were derived from the data of ultrasound images. The dynamic time warping barycenter averaging algorithm was adopted to align the dual-source data. Various machine learning techniques were applied to build quantitative models, and the performance of each model was compared. The most optimal model was further analyzed for its interpretability. RESULTS: Among the quantitative models built for the two types of MAMs, the random forest model demonstrated the best performance. For the quantitative model of the "lifting and thrusting" technique, the coefficient of determination (R²) was 0.825. For the "twisting and rotating" technique, R² reached 0.872. CONCLUSION Machine learning can be used to effectively develop the models and quantify the effects of MAMs on subcutaneous muscle tissue. It provides a new perspective to understand the mechanism of acupuncture therapy and lays a foundation for optimizing acupuncture technology and designing personalized treatment regimen in the future.
Humans ; Acupuncture Therapy/methods* ; Machine Learning ; Male ; Adult ; Female ; Subcutaneous Tissue/diagnostic imaging* ; Young Adult

Retrieving keyframes most relevant to text from small intestine videos with given labels can efficiently and accurately locate pathological regions. However, training directly on raw video data is extremely slow, while learning visual representations from image-text datasets leads to computational inconsistency. To tackle this challenge, a small bowel video keyframe retrieval based on multi-modal contrastive learning (KRCL) is proposed. This framework fully utilizes textual information from video category labels to learn video features closely related to text, while modeling temporal information within a pretrained image-text model. It transfers knowledge learned from image-text multimodal models to the video domain, enabling interaction among medical videos, images, and text data. Experimental results on the hyper-spectral and Kvasir dataset for gastrointestinal disease detection (Hyper-Kvasir) and the Microsoft Research video-to-text (MSR-VTT) retrieval dataset demonstrate the effectiveness and robustness of KRCL, with the proposed method achieving state-of-the-art performance across nearly all evaluation metrics.
Humans ; Video Recording ; Intestine, Small/diagnostic imaging* ; Machine Learning ; Image Processing, Computer-Assisted/methods* ; Algorithms