A new machine learning algorithm with high interpretability for improving the safety and efficiency of thrombolysis for stroke patients: A hospital-based pilot study

Abstract

Background Thrombolysis is the first-line treatment for patients with acute ischemic stroke. Previous studies leveraged machine learning to assist neurologists in selecting patients who could benefit the most from thrombolysis. However, when designing the algorithm, most of the previous algorithms traded interpretability for predictive power, making the algorithms hard to be trusted by neurologists and be used in real clinical practice. Methods Our proposed algorithm is an advanced version of classical k-nearest neighbors classification algorithm (KNN). We achieved high interpretability by changing the isotropy in feature space of classical KNN. We leveraged a cohort of [Formula: see text] patients to prove that our algorithm maintains the interpretability of previous models while in the meantime improving the predictive power when compared with the existing algorithms. The predictive powers of models were assessed by area under the receiver operating characteristic curve (AUC). Results In terms of interpretability, only onset time, diabetes, and baseline National Institutes of Health Stroke Scale (NIHSS) were statistically significant and their contributions to the final prediction were forced to be proportional to their feature importance values by the rescaling formula we defined. In terms of predictive power, our advanced KNN (AUC 0.88) outperformed the classical KNN (AUC 0.75, [Formula: see text]). Conclusions Our preliminary results show that the advanced KNN achieved high AUC and identified consistent significant clinical features as previous clinical trials/observational studies did. This model shows the potential to assist in thrombolysis patient selection for improving the successful rate of thrombolysis.

Artificial Intelligence and Machine Learning Applications in Critically Ill Brain Injured Patients

“…97 However, various supervised ML algorithms, using demographic, clinical, and CT imaging data, have shown a significant boost in predictive performance for sICH postthrombolysis, with AUCs ranging from 0.77 to 0.87, demonstrating adaptability across diverse ethnic groups. [98][99][100] Final infarct volume post-AIS is a critical radiographic measure linked to functional outcomes and complications like malignant cerebral edema and herniation. 101 Although imaging in the hyperacute/acute setting may not accurately reflect final infract volume, ML applications on acute MRI sequences have predicted it with high accuracy (AUC: 0.88).…”

Knowledge structure and global trends of machine learning in stroke over the past decade: A scientometric analysis

“…The advances in ML have presented opportunities to harness these massive medical datasets to inform medical practice across various domains. In recent years, ML models have been widely used to solve various complex challenges in stroke, such as early stroke detection and thrombolysis decision-making [ 6 , 7 ] neuroimaging analysis [ 8 , 9 ], stroke diagnosis and severity assessment [ 10 , 11 ], candidate selection for therapeutic intervention [ 12 , 13 ], prediction of short- and long-term functional outcomes and prognosis [ [ [14] , [15] , [16] , [17] ]]. Early detection of stroke is a crucial step in ensuring effective treatment and ML has demonstrated significant value in facilitating this process [ 18 ].…”

Uncertainty-aware deep learning for trustworthy prediction of long-term outcome after endovascular thrombectomy

“…Similar results have been achieved using extreme gradient boosting (XGBoost) and gradient boosting machines 17 . Conversely, while instance-based methods like k-Nearest-Neighbors have been employed in outcome prediction tasks 18 , 19 , they have not been applied to this specific task.…”