OBJECTIVETo find an appropriate feature representation in the biclustering of symptom-herb relationship in Chinese medicine (CM).

METHODSFour different representation schemes were tested in identifying the complex relationship between symptoms and herbs using a biclustering algorithm on an insomnia data set. These representation schemes were effective count, binary value, relative success ratio, or modified relative success ratio. The comparison of the schemes was made on the number and size of biclusters with respect to different threshold values.

RESULTS AND CONCLUSIONSThe modified relative success ratio scheme was the most appropriate feature representation among the four tested. Some of the biclusters selected from this representation scheme were known to follow the therapeutic principles of CM, while others may offer clues for further clinical investigations.

Algorithms ; Cluster Analysis ; Drugs, Chinese Herbal ; therapeutic use ; Humans ; Medicine, Chinese Traditional ; Sleep Initiation and Maintenance Disorders ; drug therapy

Coronavirus disease 2019 (COVID-19) is now pandemic worldwide and has heavily overloaded hospitals in Wuhan City, China during the time between late January and February. We reported the clinical features and therapeutic characteristics of moderate COVID-19 cases in Wuhan that were treated via the integration of traditional Chinese medicine (TCM) and Western medicine. We collected electronic medical record (EMR) data, which included the full clinical profiles of patients, from a designated TCM hospital in Wuhan. The structured data of symptoms and drugs from admission notes were obtained through an information extraction process. Other key clinical entities were also confirmed and normalized to obtain information on the diagnosis, clinical treatments, laboratory tests, and outcomes of the patients. A total of 293 COVID-19 inpatient cases, including 207 moderate and 86 (29.3%) severe cases, were included in our research. Among these cases, 238 were discharged, 31 were transferred, and 24 (all severe cases) died in the hospital. Our COVID-19 cases involved elderly patients with advanced ages (57 years on average) and high comorbidity rates (61%). Our results reconfirmed several well-recognized risk factors, such as age, gender (male), and comorbidities, as well as provided novel laboratory indications (e.g., cholesterol) and TCM-specific phenotype markers (e.g., dull tongue) that were relevant to COVID-19 infections and prognosis. In addition to antiviral/antibiotics and standard supportive therapies, TCM herbal prescriptions incorporating 290 distinct herbs were used in 273 (93%) cases. The cases that received TCM treatment had lower death rates than those that did not receive TCM treatment (17/273 = 6.2% vs. 7/20= 35%, P = 0.0004 for all cases; 17/77= 22% vs. 7/9= 77.7%, P = 0.002 for severe cases). The TCM herbal prescriptions used for the treatment of COVID-19 infections mainly consisted of Pericarpium Citri Reticulatae, Radix Scutellariae, Rhizoma Pinellia, and their combinations, which reflected the practical TCM principles (e.g., clearing heat and dampening phlegm). Lastly, 59% of the patients received treatment, including antiviral, antibiotics, and Chinese patent medicine, before admission. This situation might have some effects on symptoms, such as fever and dry cough. By using EMR data, we described the clinical features and therapeutic characteristics of 293 COVID-19 cases treated via the integration of TCM herbal prescriptions and Western medicine. Clinical manifestations and treatments before admission and in the hospital were investigated. Our results preliminarily showed the potential effectiveness of TCM herbal prescriptions and their regularities in COVID-19 treatment.
Adult ; Aged ; Aged, 80 and over ; COVID-19/therapy* ; China ; Combined Modality Therapy ; Drugs, Chinese Herbal/therapeutic use* ; Female ; Hospitalization ; Humans ; Male ; Medicine, Chinese Traditional ; Middle Aged ; Retrospective Studies ; Survival Rate ; Treatment Outcome