Application of artificial neural network in daily prediction of bleeding in ICU patients treated with anti-thrombotic therapy

Abstract

Objectives Anti-thrombotic therapy is the basis of thrombosis prevention and treatment. Bleeding is the main adverse event of anti-thrombosis. Existing laboratory indicators cannot accurately reflect the real-time coagulation function. It is necessary to develop tools to dynamically evaluate the risk and benefits of anti-thrombosis to prescribe accurate anti-thrombotic therapy. Methods The prediction model,daily prediction of bleeding risk in ICU patients treated with anti-thrombotic therapy, was built using deep learning algorithm recurrent neural networks, and the model results and performance were compared with clinicians. Results There was no significant statistical discrepancy in the baseline. ROC curves of the four models in the validation and test set were drawn, respectively. One-layer GRU of the validation set had a larger AUC (0.9462; 95%CI, 0.9147–0.9778). Analysis was conducted in the test set, and the ROC curve showed the superiority of two layers LSTM over one-layer GRU, while the former AUC was 0.8391(95%CI, 0.7786–0.8997). One-layer GRU in the test set possessed a better specificity (sensitivity 0.5942; specificity 0.9300). The Fleiss’ k of junior clinicians, senior clinicians, and machine learning classifiers is 0.0984, 0.4562, and 0.8012, respectively. Conclusions Recurrent neural networks were first applied for daily prediction of bleeding risk in ICU patients treated with anti-thrombotic therapy. Deep learning classifiers are more reliable and consistent than human classifiers. The machine learning classifier suggested strong reliability. The deep learning algorithm significantly outperformed human classifiers in prediction time.

Machine Learning-Based Predictive Models for Patients with Venous Thromboembolism: A Systematic Review

“…Overall, only 10 studies were identified, and are summarized in Supplementary Table S3 (available in the online version). 69 70 71 72 73 74 75 76 77 78 Two studies predicted PTS with good to excellent performance (AUC: 0.83–0.89) 69 70 The diagnosis of PTS was based on the Villalta score, but immediate comparison with the score was not performed. Regarding prediction of mortality, one shortcoming is the variability of its definition between studies, in terms of time to event.…”

Rapid-Test Kit for Cardiac Troponin I: A Reliable Enzyme-Linked-Immuno-Substrate-Assay-Based Biosensor for Daily-Use Naked-Eye Detection and Pharmacokinetic Studies for Myocardial Infarction in Cardiovascular Disease

“…In cardiovascular diseases (CVDs), plaque accumulates in the coronary artery and affects blood flow, − resulting in a lack of oxygen or nutrition in the heart that can lead to death. The American Heart Association annually releases statistical reports on heart disease, stroke, and cardiovascular risk factors, and considerable research efforts are devoted to CVD. , This focus includes artificial polymer bypass graft transplants, ,− stent material innovations, antithrombosis medicine studies, ,− tissue engineering for artificial vascular grafts, ,, tissue repair medicine for ischemia, ,− and creating biomarkers for detection. ,,, …”