A DenseNet-based diagnosis algorithm for automated diagnosis using clinical ECG data / 南方医科大学学报

OBJECTIVE@#To train convolutional networks using multi-lead ECG data and classify new data accurately to provide reliable information for clinical diagnosis.@*METHODS@#The data were pre-processed with a bandpass filter, and signal framing was adopted to adjust the data of different lengths to the same size to facilitate network training and prediction. The dataset was expanded by increasing the sample size to improve the detection rate of abnormal samples. A depth-wise separable convolution structure was used for more specific feature extraction for different channels of twelve-lead ECG data. We trained the two classifiers for each label using the improved DenseNet to classify different labels.@*RESULTS@#The propose model showed an accuracy of 80.13% for distinguishing between normal and abnormal ECG with a sensitivity of 80.38%, a specificity of 79.91% and a F1 score of 79.35%.@*CONCLUSIONS@#The model proposed herein can rapidly and effectively classify the ECG data. The running time of a single dataset on GPU is 33.59 ms, which allows real-time prediction to meet the clinical requirements.

Study on Pre-processing Methods of Clinical Data from TCM Individual Treatment of Insomnia Based on Syndrome Differentiation / 中国中医药信息杂志

Objective To explore suitable pre-processing methods for the TCM clinical data based on prospective study on insomnia treated by syndrome differentiation. Methods Based on the TCM shared clinical and research information platform and by using man-machine combination method, data cleaning rules, physician review, rule revision, procedural import and batch processing were used to conduct pre-processing for data in prospective study on insomniac treated by syndrome differentiation of 8 TCM doctors. Results Totally 27534 rules for symptoms data of individual treatment of insomnia were made and 1036 rules for diagnostic data, 842 rules for therapeutic ways, 540 rules for formula data, 3785 rules for data of Chinese materia medica. Conclusion Different kinds of terminology concepts were normalized at different levels, at the same time, characteristics of individualized treatment based on syndrome differentiation were reserved. Appropriate pre-processing methods can be used in the reaserch of individualization and standardization of TCM syndrome differentiation clinical data and can provide support for data mining.

Analysis and application of TCM data preprocessing method based on Apriori algorithm / 医疗卫生装备

To present a data processing method of traditional Chinese medicine based on Apriori algorithm to improve the efficiency of data mining and ensure the accuracy of the knowledge or conclusion in the data mining. The importance of data preprocessing in data mining was analyzed, along with the characteristics of TCM data and the requirements of Apriori algorithm for mining data. Some new functions were formed with considerations on the exam-ples. The data preprocessing was explored from the aspects of terminology standardization, eliminating unqualified data, structured prescription data, data sorting and etc. The new functions were simple and easy to operate, and the preprocessed data made the efficiency of TCM data mining enhanced greatly. The preprocessing method based on Apriori algorithm for TCM data facilities the TCM data mining.