Ischemic Stroke is one of the leading causes of death and disability in low- and middle-income countries. Given the time-dependent nature of ischemic stroke, understanding patient care pathways across emergency, therapeutic decision-making, hospitalization, and discharge is essential. This study analyzed the care pathway of acute ischemic stroke patients in a Brazilian hospital, aiming to identify factors associated with hospital length of stay. Secondary data from 232 patients were evaluated using a hybrid analytical approach combining machine learning (Gradient Boosting Machine) with interpretable statistical modeling through knowledge distillation. Clinical and demographic variables were considered, such as age, sex, history of stroke, NIHSS score at admission, and TOAST classification. Age and baseline NIHSS were the factors most strongly associated with length of stay, highlighting the relevance of initial neurological severity. Additionally, TOAST classification was associated with length of stay, particularly for the Large-Artery Atherosclerosis and Other Etiologies categories.