Search

Polycystic kidney disease (Autosomal Dominant Polycystic Kidney Disease, ADPKD) is the most common genetic disease of the kidney, and the measurement of Total Kidney Volume (TKV) in clinical research of this disease is essential to study the progression of ADPKD. At present, the volume segmentation of polycystic kidneys mainly relies on doctors to manually outline the kidney boundary on the radiological image. This process is time-consuming, labor-intensive, inefficient, subjective, and difficult to guarantee consistency. In the research of this paper, A multi-module hybrid U-shape segmentation method is proposed (HUNet), which introduces wavelet pooling, cascade residual, and efficient multi-head self-attention into the U-shape structure. We use wavelet pooling instead of traditional down-sampling to reduce the loss of detailed features, the use of cascaded residual modules can improve the ability of model feature reuse, and the use of efficient multi-head self-attention modules can effectively capture global multi-scale information. In the decoding process of the U-shape, the corresponding loss value of each decoder will be calculated, and finally, the total loss value of the model will be obtained by weighted average. The method was trained and tested on the polycystic kidney dataset provided by Shanghai Changzheng Hospital. We automatically segmented the ADPKD in MRI images using the proposed method with a remarkably high Dice similarity coefficient relative to the manual segmentation (mean=0.915). The percentage difference between the total kidney volume values using manual and HUNet methods was only 0.4%. The proposed approach enables fast and accurate TKV measurement.

Traditional Chinese medicine (TCM) is based on a unique disease diagnosis and treatment system that has been developed over the last 2,300 years. In the TCM, “syndrome differentiation and treatment”(SDAT) is a core method for doctors to deal with diseases. This diagnostic and therapeutic technique that infer the occurrence and the development of diseases by observing symptoms as a whole, not only has its own uniqueness but also has been recognized by the public in oriented medical fields for its clinical efficacy. With recent developments in computer science, the Internet, big data, and artificial intelligence, a study based on the SDAT algorithm has aroused much attention. This paper encompasses three stages spanning 30 years to accomplish the following: 1) the TCM data and the modern SDAT system were collated and summarized based on 35,706 reference data on the TCM, starting from the syndrome differentiation of four aspects, such as the cause, location, characteristics, and conditions of the disease (CLCC), we constructed a quantitative model of the TCM SDAT regarding the CLCC of the disease, collected the symptom information on the diagnosed subject, and transferred them to the SDAT assistant algorithm for calculation and analysis, to determine the CLCC, Based on the therapy recommended by the differentiation results in the knowledge base and the prescription and traditional Chinese medicines recommended by the therapy, any stage of all diseases could determine a syndrome type by differentiating the CLCC, we constructed the basic SDAT algorithm integrating theory, method, prescription, and medicine and realized the calculability in the TCM diagnosis and treatment process; 2) based on the SDAT algorithm, we developed the TCM doctor's workstation software and introduced it to more than 80 TCM institutions in Sichuan province, China, we collated a large-scale trove of samples of the TCM data platform that was established with more than 2.9 million TCM electronic medical records (EMRs) and reference data, and had the compliance tested and algorithm verified on the 9,300 EMRs of the common diseases in the TCM; and 3) based on the dimension reduction and degree elevation optimization of the technology with a directed graph to the basic algorithm, the algorithm complexity was reduced and the accuracy of the algorithm was improved. It was demonstrated that the coincidence rate of the basic model was 80.47% and the basic coincidence rate was 96.19%. After optimizing the basic algorithm (for example, for gastric abscess), the coincidence rate increased by 7.04%. The test results demonstrated the efficacy of the model study. This model realized a computable SDAT to specify and assist in the differentiation diagnosis and in the treatment processes of the TCM and improve the service quality of the TCM diagnosis and treatment.