Surgical complication evaluation system based on deep learning

Abstract

The invention discloses a surgical complication evaluation system based on deep learning, belonging to the technical field of medical aid decision making. According to a specific scheme, the system comprises a cloud database, a cloud server, a medical detection module, a medical case module and a doctor terminal, wherein the cloud database comprises historical clinical data of a medical unit. Artificial intelligence and medical treatment are closely combined in the invention, and the system covers more than 670 common symptoms, more than 700 symptom synonyms, more than 600 common physical examination items and more than 1200 common examination items and the method has three advantages of surgical complication prediction standardization, prevention intelligentialization and individuation ofmanagement and control. The types, time, severity and the like of complications are rapidly detected by utilizing deep learning, medical records and assay data of a patient are read within 10 seconds, prevention measures (high-grade evidences) are recommended, system preliminary testing is carried out, and the accuracy of predicting deep venous thrombosis by the system reaches 94.5%.

Description

technical field [0001] The present invention relates to the technical field of medical assistant decision-making, and more specifically, it relates to a surgical complication evaluation system based on deep learning. Background technique [0002] Statistics from the National Health Commission show that the number of surgeries in my country has grown rapidly in recent years, with the number of surgeries reaching 61.71 million in 2018 alone. However, a sample from the emergency department showed that among the 1.5 million emergency operations between January 2006 and December 2015, the medical expenses caused by surgical complications totaled 18 billion yuan, accounting for 48% of the overall medical economic burden. It shows that the medical quality system is still not perfect, and the scientific rigor still has periodic defects. Before, during and after the diagnosis, patients often aggravate their condition due to complications, which increases the difficulty of treatment, ...

AI Technical Summary

Problems solved by technology

[0004] (1) Subjective limitations: medical institutions have been using comprehensive judgments based on the subjective experience of doctors and patients' expressions, signs and appearances, and instrumental examination results. Since the prediction of surgical complications cannot have a unified standard, subjectivity is unavoidable, even if it is the same experience Rich clinicians, facing the same patient, may also come up with different diagnostic opinions, resulting in the risk of potential postoperative complications and even the death of the patient. There is a lack of an accurate, scientific and rigorous prediction system for surgical complications

[0005] (2) Objective limitations: clinicians need to evaluate the safety of the patient's surgery and the possibility of more than one hundred suspected surgical complications within about 15 minutes. Each patient has at least five preoperative routine examinations and 321 laboratory results

[0006] (3) Existing score prediction system: In modern medical care, my country currently uses more scoring systems, mainly APACHE-Ⅱ score, POSSUM score, etc., but these two systems still have disadvantages such as high limitations and unclear audiences. , has not been widely used in hospitals, and the APACHE II system has been criticized for its overestimation of mortality in practical applications, and because the POSSUM score only predicts the mortality and complication rate at 30 days after surgery, the POSSUM score Unable to predict mortality and complication rates beyond 30 days

In summary, it is not difficult to find that a sound postoperative complication prediction system has not yet been established in China's medical industry at this stage. Therefore, it is necessary to establish a surgical complication prediction and avoidance system that is applicable to a wide range of departments, has a wide audience, and is easy to operate. The decision-making AI intelligent system is imminent.

[0007] (4) In terms of auxiliary decision-making for avoiding surgical complications, at least five preoperative routine examinations and 321 laboratory results are involved before surgery, and there are strong individual differences in information, which requires the cooperation of multiple disciplines such as surgery, nursing, and rehabilitation medicine. The decision-making process Complex, currently there is no relevant practical tool to help physicians and other medical staff choose the most suitable complication prevention plan

Images