Rapid triage for ischemic stroke: a machine learning-driven approach in the context of predictive, preventive and personalised medicine

Abstract

Background Recognising the early signs of ischemic stroke (IS) in emergency settings has been challenging. Machine learning (ML), a robust tool for predictive, preventive and personalised medicine (PPPM/3PM), presents a possible solution for this issue and produces accurate predictions for real-time data processing. Methods This investigation evaluated 4999 IS patients among a total of 10,476 adults included in the initial dataset, and 1076 IS subjects among 3935 participants in the external validation dataset. Six ML-based models for the prediction of IS were trained on the initial dataset of 10,476 participants (split participants into a training set [80%] and an internal validation set [20%]). Selected clinical laboratory features routinely assessed at admission were used to inform the models. Model performance was mainly evaluated by the area under the receiver operating characteristic (AUC) curve. Additional techniques—permutation feature importance (PFI), local interpretable model-agnostic explanations (LIME), and SHapley Additive exPlanations (SHAP)—were applied for explaining the black-box ML models. Results Fifteen routine haematological and biochemical features were selected to establish ML-based models for the prediction of IS. The XGBoost-based model achieved the highest predictive performance, reaching AUCs of 0.91 (0.90–0.92) and 0.92 (0.91–0.93) in the internal and external datasets respectively. PFI globally revealed that demographic feature age, routine haematological parameters, haemoglobin and neutrophil count, and biochemical analytes total protein and high-density lipoprotein cholesterol were more influential on the model’s prediction. LIME and SHAP showed similar local feature attribution explanations. Conclusion In the context of PPPM/3PM, we used the selected predictors obtained from the results of common blood tests to develop and validate ML-based models for the diagnosis of IS. The XGBoost-based model offers the most accurate prediction. By incorporating the individualised patient profile, this prediction tool is simple and quick to administer. This is promising to support subjective decision making in resource-limited settings or primary care, thereby shortening the time window for the treatment, and improving outcomes after IS.

Serum Olink Proteomics-Based Identification of Protein Biomarkers Associated with the Immune Response in Ischemic Stroke

“…In addition, we constructed and verified an IS diagnostic model consisting of six biomarkers, i.e., the abovementioned four proteins and two other immune response-related proteins: MASP1, CLEC4D, PIK3AP1, STC1, PTH1R, and HCLS1. The model demonstrated comparable or superior accuracy to IS models utilizing routine clinical hematology and biochemical characteristics, 43 inflammation-associated proteins, 44 or noncoding RNAs 45 in the bloodstream. However, it is important to note that our study focused solely on elucidating protein alterations at the immune level.…”

“…Finally, the dynamic changes of proteins may provide more information for the prognostic prediction of IS. In the further research, integrate machine learning algorithms should be applied to efficiently identify diagnostic biomarker panels with high validation and significant classification effects, 47 then multiindicator joint inspection including proteomics is realized via microfluidics technology, 48 and an online intelligent screening and evaluation auxiliary system is established 43 to assist clinical diagnosis, gain time window, and guide clinical decision-making.…”

Mutual effect of homocysteine and uric acid on arterial stiffness and cardiovascular risk in the context of predictive, preventive, and personalized medicine

“…In our subsequent analysis of the association of UA and Hcy with 10-year CVD risk, people with both high UA and Hcy showed the highest risk, supporting our hypothesis that UA and Hcy can mediate CVD by causing arterial stiffness. The study by Zheng et al suggests that ischemic stroke can be predicted based on routine hematological and biochemical features, and early identification of disease risk may facilitate the formulation of primary care strategies and improve the prognosis of the disease [ 71 ]. In this study, the nomogram of CVD risk assessment based on Hcy and UA indicators showed that high Hcy and UA levels predicted the increase of 10-year CVD risk, and the CVD risk was highest when both of them were at high levels.…”

Diagnostic accuracy of autoverification and guidance system for COVID-19 RT-PCR results

“…PPPM combines the advantages and minimizes the disadvantages of the existing approaches and has been considered as the "medicine of the future" [5]. It has been successfully applied to non-communicable diseases including cancer [32,33], cardiovascular diseases [34], ischemic stroke [35] and neurodegenerative diseases [36] and infectious diseases such as COVID-19 [37].…”