VTE Risk Prevention System Based on Big Data Analysis and Multimodule System

Venous thromboembolism (VTE) is one of the most common complications in perioperative period and the most common cause of postoperative death. VTE includes deep vein thrombosis (DVT) and acute pulmonary thromboembolism (PTE). Since the embolus of PTE comes from the deep vein thrombosis, and not all PE patients can present obvious clinical symptoms, VTE is currently considered as a disease for research, prevention, diagnosis and treatment at home and abroad. Therefore, we urgently need to develop a more comprehensive and reliable perioperative VTE risk prevention system based on medical big data and multi-module computer in current clinical practice, so as to effectively guide the prevention of DVT/PE, and thus reduce the perioperative mortality.

As early as 1991, Caprini et al. collated and published the clinical scale of VTE risk assessment based on the study data of 538 patients. In 2010, Caprini et al. conducted retrospective verification of this scale again among 8216 patients, thus making this scale an important reference for many medical institutions to assess the risk of VTE occurrence in perioperative patients, and it has been used ever since. However, over the past 30 years, the disease spectrum, surgical methods, anesthesia methods and the cognition of thrombus formation susceptibility of surgical patients have all undergone great changes. Perioperative patients who are evaluated by Caprini scale and take preventive measures according to its suggested measures will still have postoperative VTE. Therefore, in recent years, scholars have published new VTE risk prediction methods, such as Kucher model, Padua model and so on. However, the study cases established by the model are almost only about 1000 cases, and the included research indicators are limited.

The new technology realizes for the first time the identification, reading and integration of VTE-related risk indicators under the guidance of informatization and digitalization, providing technical support for big data analysis; To realize forward-looking, big data, multi-variable, multi-module VTE-related risk assessment and prediction; The existing scoring system should be improved to provide a more reliable and operable scoring model for perioperative VTE prevention and treatment.