Determining the potential benefits of endovascular thrombectomy (EVT) in acute ischemic stroke (AIS) patients requires an accurate early prognosis. This study employs and compares multi-modal machine learning algorithms for predicting the National Institutes of Health Stroke Scale (NIHSS) score at discharge from the hospital, utilizing initial computed tomography (CT) scans and clinical data. Prior to formulating the advanced predictive models, a comprehensive literature review on AIS outcome prediction is performed alongside the implementation of several simpler benchmark models to set a comparative baseline. Subsequently, preliminary models are selected and screened until finally three machine learning models, including Convolutional Neural Networks (CNN), are developed, tested, and compared to assess their effectiveness in guiding the decision to proceed with EVT. The imaging-incorporating models, particularly those utilizing CNNs, have demonstrated a significant ability to successfully leverage the imaging data when compared to the baseline models. The study concludes with a comparative analysis, illustrating the performance of these models in predicting the clinical outcome in AIS patients.