Time points of outcome are often neglected in acupuncture meta-analyses: a methodological survey.

OBJECTIVES: To systematically understand the transparency of outcome measurement time point reporting in meta-analyses of acupuncture. STUDY DESIGN AND SETTING: We searched for meta-analyses of acupuncture published between 2013 and 2022 in PubMed, Embase, and Cochrane Library. A team of method-trained investigators screened studies for eligibility and collected data using pilot-tested standardized questionnaires. We documented in detail the reporting of outcome measurement time points in acupuncture meta-analyses. RESULTS: A total of 224 acupuncture meta-analyses were included. Of these, 98 (43.8%) studies did not specify the time points of primary outcome. Among 126 (56.3%) meta-analyses which reported the time points of primary outcome, only 22 (17.5%) meta-analyses specified time points in corresponding protocol. Among 48 (38.1%) meta-analyses that estimated treatment effects of multiple time points, 11 (22.9%) meta-analyses used inappropriate meta-analysis method (subgroup analysis) to pool effect size, and none of the meta-analyses used advanced methods for pooling effect sizes at different time points. CONCLUSION: Transparency in reporting outcome time points for acupuncture meta-analyses and appropriate methods to pool the effect size of multiple time points were lacking. For future systematic reviews, the transparency of outcome measurement time points should be emphasized in the protocols and final reports. Furthermore, advanced methods should be considered for pooling effect sizes at multiple time points.

ResNet for recognition of Qi-deficiency constitution and balanced constitution based on voice.

Background: According to traditional Chinese medicine theory, a Qi-deficiency constitution is characterized by a lower voice frequency, shortness of breath, reluctance to speak, an introverted personality, emotional instability, and timidity. People with Qi-deficiency constitution are prone to repeated colds and have a higher probability of chronic diseases and depression. However, a person with a Balanced constitution is relatively healthy in all physical and psychological aspects. At present, the determination of whether one has a Qi-deficiency constitution or a Balanced constitution are mostly based on a scale, which is easily affected by subjective factors. As an objective method of diagnosis, the human voice is worthy of research. Therefore, the purpose of this study is to improve the objectivity of determining Qi-deficiency constitution and Balanced constitution through one's voice and to explore the feasibility of deep learning in TCM constitution recognition. Methods: The voices of 48 subjects were collected, and the constitution classification results were obtained from the classification and determination of TCM constitutions. Then, the constitution was classified according to the ResNet residual neural network model. Results: A total of 720 voice data points were collected from 48 subjects. The classification accuracy rate of the Qi-deficiency constitution and Balanced constitution was 81.5% according to ResNet. The loss values of the model training and test sets gradually decreased to 0, while the ACC values of the training and test sets tended to increase, and the ACC values of the training set approached 1. The ROC curve shows an AUC value of 0.85. Conclusion: The Qi-deficiency constitution and Balanced constitution determination method based on the ResNet residual neural network model proposed in this study can improve the efficiency of constitution recognition and provide decision support for clinical practice.

Automatic tongue image quality assessment using a multi-task deep learning model.

The quality of tongue images has a significant influence on the performance of tongue diagnosis in Chinese medicine. During the acquisition process, the quality of the tongue image is easily affected by factors such as the illumination, camera parameters, and tongue extension of the subject. To ensure that the quality of the collected images meet the diagnostic criteria of traditional Chinese Medicine practitioners, we propose a deep learning model to evaluate the quality of tongue images. First, we acquired the tongue images of the patients under different lighting conditions, exposures, and tongue extension conditions using the inspection instrument, and experienced Chinese physicians manually screened them into high-quality and unqualified tongue datasets. We then designed a multi-task deep learning network to classify and evaluate the quality of tongue images by adding tongue segmentation as an auxiliary task, as the two tasks are related and can promote each other. Finally, we adaptively designed different task weight coefficients of a multi-task network to obtain better tongue image quality assessment (IQA) performance, as the two tasks have relatively different contributions in the loss weighting scheme. Experimental results show that the proposed method is superior to the traditional deep learning tongue IQA method, and as an additional task of the network, it can output the tongue segmentation area, which provides convenience for follow-up clinical tongue diagnosis. In addition, we used network visualization to verify the effectiveness of the proposed method qualitatively.

Applied Research on the Combination of Weighted Network and Supervised Learning in Acupoints Compatibility.

To enhance the depth of excavation and promote the intelligence of acupoint compatibility, a method of constructing weighted network, which combines the attributes of acupoints and supervised learning, is proposed for link prediction. Medical cases of cervical spondylosis with acupuncture treatment are standardized, and a weighted network is constructed according to acupoint attributes. Multiple similarity features are extracted from the network and input into a supervised learning model for prediction. And, the performance of the algorithm is evaluated through evaluation indicators. The experiment finally screened 67 eligible medical cases, and the network model involved 141 acupoint nodes with 1048 edge. Except for the Preferential Attachment similarity index and the Decision Tree model, all other similarity indexes performed well in the model, among which the combination of PI index and Multilayer Perception model had the best prediction effect with an AUC value of 0.9351, confirming the feasibility of weighted networks combined with supervised learning for link prediction, also as a strong support for clinical point selection.

Which of the acupuncture treatment regimen for lumbar disc herniation is more effective and safer: A protocol for systematic review and network meta-analysis.

INTRODUCTION: Lumbar disc herniation (LDH) is the most common cause of low back pain and severely affects people's quality of life and ability to work. Although many clinical trials and medical reports conducted over the years have shown that acupuncture treatments are effective for LDH, the comparative effectiveness of these different acupuncture therapies is still unclear. This protocol of a network meta-analysis was designed to compare the effects and safety of acupuncture treatment regimens on LDH using both direct and indirect evidence. METHODS AND ANALYSIS: This protocol is reported according to the 2015 PRISMA-P and PRISMA guidelines for acupuncture. Eight databases and two platforms will be searched for articles published from their establishment to 1 December 2020 with medical subject heading terms and keywords. Three reviewers will verify the eligible randomized controlled trials independently. NoteExpress (3.2.0) software will be utilized to manage the literature. The overall quality of evidence will be evaluated by Confidence In Network Meta-Analysis (CINeMA). Additionally, we will conduct a meta-analysis of the effectiveness, recurrence rate, and symptom score of acupuncture in treating LDH using Review Manager (RevManV.5.4.1) and R4.0.2 software (The R Foundation for Statistical Computing). RESULTS: The results of the study will be published in journals or relevant conferences. CONCLUSION: This proposed systematic review will evaluate the comparative efficacy and safety of various acupuncture methods and combination protocols for LDH.

[Acupoint selection rules of acupuncture for cervical spondylotic radiculopathy based on data mining].

OBJECTIVE: To analyze the rules of acupoint selection in the acupuncture treatment of cervical spondylotic radiculopathy by data mining. METHODS: The randomized controlled trials (RCTs) regarding acupuncture for cervical spondylotic radiculopathy published from July 15 of 2009 to July 15 of 2019 were retrieved from databases of CNKI, VIP, Wanfang, SinoMed, PubMed and EMbase. A database was established with Microsoft Excel 2016. The frequency and total effective rate of high-frequency acupoints, meridians and acupoint combinations were analyzed, and the association rules of acupoints and meridians were analyzed by Apriori algorithm. RESULTS: A total of 87 RCTs were included, involving 104 acupoints with a total frequency of 921. Among them, the high-frequency acupoints were cervical Jiaji (EX-B 2, 87 times), Fengchi (GB 20, 70 times), Houxi (SI 3, 54 times), etc. The frequently-used acupoints were mainly distributed in the hand yangming large intestine meridian, the foot shaoyang gallbladder meridian and hand taiyang small intestine meridian. The frequently-used acupoint combination was Fengchi (GB 20)-cervical Jiaji (EX-B 2), and most of the combinations were acupoints at the proximal end and acupoints at the far and near end. With the analysis of association rules, 15 groups of acupoint association rules and meridian association rules were obtained. CONCLUSION: It is feasible to explore the acupoint selection and compatibility rules of acupuncture for cervical spondylotic radiculopathy by data mining. This study could provide corresponding reference for clinical treatment.

Deep generative learning for automated EHR diagnosis of traditional Chinese medicine.

BACKGROUND: Computer-aided medical decision-making (CAMDM) is the method to utilize massive EMR data as both empirical and evidence support for the decision procedure of healthcare activities. Well-developed information infrastructure, such as hospital information systems and disease surveillance systems, provides abundant data for CAMDM. However, the complexity of EMR data with abstract medical knowledge makes the conventional model incompetent for the analysis. Thus a deep belief networks (DBN) based model is proposed to simulate the information analysis and decision-making procedure in medical practice. The purpose of this paper is to evaluate a deep learning architecture as an effective solution for CAMDM. METHODS: A two-step model is applied in our study. At the first step, an optimized seven-layer deep belief network (DBN) is applied as an unsupervised learning algorithm to perform model training to acquire feature representation. Then a support vector machine model is adopted to DBN at the second step of the supervised learning. There are two data sets used in the experiments. One is a plain text data set indexed by medical experts. The other is a structured dataset on primary hypertension. The data are randomly divided to generate the training set for the unsupervised learning and the testing set for the supervised learning. The model performance is evaluated by the statistics of mean and variance, the average precision and coverage on the data sets. Two conventional shallow models (support vector machine / SVM and decision tree / DT) are applied as the comparisons to show the superiority of our proposed approach. RESULTS: The deep learning (DBN + SVM) model outperforms simple SVM and DT on two data sets in terms of all the evaluation measures, which confirms our motivation that the deep model is good at capturing the key features with less dependence when the index is built up by manpower. CONCLUSIONS: Our study shows the two-step deep learning model achieves high performance for medical information retrieval over the conventional shallow models. It is able to capture the features of both plain text and the highly-structured database of EMR data. The performance of the deep model is superior to the conventional shallow learning models such as SVM and DT. It is an appropriate knowledge-learning model for information retrieval of EMR system. Therefore, deep learning provides a good solution to improve the performance of CAMDM systems.

Depressive neurosis treated by acupuncture for regulating the liver--a report of 176 cases.

OBJECTIVE: To observe therapeutic effect of acupuncture for regulating the liver on depressive neurosis. METHODS: In a multi-center randomized controlled trial, 440 patients were divided into 3 groups: Acupuncture group for regulating the liver (Acup., 176 cases) was treated by acupuncture at Siguan Points, i.e., bilateral Hegu (LI 4) and Taichong (LR 3), Baihui (GV 20) and Yintang (EX-HN3) plus ear-acupuncture, Prozac group (P., 176 cases) by oral administration of Prozac, and Non-acupoint needling group (NAN, 88 cases) by acupuncture at non-acupoints as acupuncture placebo. Self-rating Depression Scale (SDS) was examined before treatment, and one month, two and three months after treatment respectively to evaluate therapeutic effect, and Rating Scale for Side Effects (SERS) was used to evaluate the safety. RESULTS: After one month of treatment, SDS scores in Acup. Group were significantly lower than that in P. Group (P < 0.05) and than that in NAN Group (P < 0.01), and SDS scores in P. Group were lower than that in NAN Group (P < 0.05), showing the SDS scores in Acup. Group < P. Group < NAN Group. After 2 months of treatment, SDS scores in Acup. Group were also significantly lower than that in P. Group (P < 0.01) and than that in NAN Group (P < 0.01), and SDS scores in P. Group were also lower than that in NAN Group (P < 0.05), showing the SDS scores in Acup. Group 0.05), showing the SERS scores in Acup. Group < NAN Group < P. Group. No side effect was found in Acup. and NAN groups. CONCLUSION: The therapeutic effect of acupuncture on depressive neurosis is better than or similar to that of Prozac but with less side effect.

[Acupuncture for treatment of depressive neurosis: a multi-center randomized controlled study].

OBJECTIVE: To observe the clinical therapeutic effect of acupuncture on depressive neurosis. METHODS: With a multi-center randomized controlled study, 440 cases were randomly divided into an acupuncture group, a prozac group, a non-acupoint needling group. In the acupuncture group, Hegu (LI 4) and Taichong (LR 3) were selected, and the Prozac group were treated with administration of 20 mg/d and the non-acupoint needling group were treated with needling the points deviating from the acupoints. The therapeutic effect was evaluated by HAMD score reduction rate, and Asberg's anti-depressant side-effect rating scale (SERS) and severe adverse reaction were used for safety evaluation, and the data were analyzed with ITT. RESULTS: The total effective rate was 86. 4% in the acupuncture group, which was better than 59.1% in the non-acupoint needling group and 72.7% in the prozac group; HAMD score in the acupuncture group was similar to that in the Prozac group, which was better than that in the non-acupoint needling group; the SERS scores in the acupuncture group and the non-acupoint needling group were significantly lower than that in the Prozac group, with no severe side-effects found for acupuncture. CONCLUSION: Acupuncture is an effective and safe therapy for depressive neurosis; therapeutic effect of acupuncture on depressive neurosis possibly is better than or similar to that of Prozac, but with less side-effects.

[Randomized controlled study on picking therapy for treatment of cervical spondylosis].

OBJECTIVE: To use randomized controlled clinical research method to assess therapeutic effect of picking therapy on cervical spondylosis. METHODS: One hundred and fifty-eight cases were randomly divided into a picking therapy group (n=56), a routine acupuncture group (n=55) and a local anesthesia group (n=47). They were treated respectively with picking therapy, routine acupuncture and local anesthesia at Jing bailao (EX HN 15), Dazhui (GV 14), Jianjing (GB 21), etc. Brief McGill Pain Questionaire was used for score, which was combined with clinical symptoms and signs to analyze the therapeutic effect. RESULTS: The cured rate was 57.1% in the picking therapy group, better than 23.6% in the acupuncture group and 14.9% in the local anesthesia group (P < 0.01), and adverse reaction was basically not found in the picking therapy group. CONCLUSION: Picking therapy is a highly effective and safe therapy for cervical spondylosis.