Stroke is one of the leading causes of long-term disabilities in motor and cognition functionality. An early and accurate prediction of rehabilitation outcomes can lead to a tailor-made treatment that can significantly improve the post-stroke quality of life of a person. This scoping review aimed to summarize studies that use Artificial Intelligence (AI) for the prediction of language and cognition rehabilitation outcomes and the need to use AI in this domain. This study followed the PRISMA-ScR guidelines for two databases, Scopus and PubMed. The results, which are measured with several metrics depending on the task, regression, or classification, present encouraging outcomes as they can predict the cognitive functionality of post-stroke patients with relative precision. Among the results of the paper are the identification of the most effective Machine Learning (ML) algorithms, and the identification of the key factors that influence rehabilitation outcomes. The majority of studies focus on aphasia and present high performance achieving up to 97% recall and 91.4% precision. The main limitations of the studies were the small subject population and the lack of an external dataset. However, effective ML algorithms along with explainability are expected to become among the most prominent solutions for precision medicine due to their ability to overcome non-linearities on data and provide insights and transparent predictions that can help healthcare professionals make more informed and accurate decisions.

Stroke remains a predominant cause of mortality and disability worldwide. The endeavor to diagnose stroke through biomechanical time-series data coupled with Artificial Intelligence (AI) poses a formidable challenge, especially amidst constrained participant numbers. The challenge escalates when dealing with small datasets, a common scenario in preliminary medical research. While recent advances have ushered in few-shot learning algorithms adept at handling sparse data, this paper pioneers a distinctive methodology involving a visualization-centric approach to navigating the small-data challenge in diagnosing stroke survivors based on gait-analysis-derived biomechanical data. Employing Siamese neural networks (SNNs), our method transforms a biomechanical time series into visually intuitive images, facilitating a unique analytical lens. The kinematic data encapsulated comprise a spectrum of gait metrics, including movements of the ankle, knee, hip, and center of mass in three dimensions for both paretic and non-paretic legs. Following the visual transformation, the SNN serves as a potent feature extractor, mapping the data into a high-dimensional feature space conducive to classification. The extracted features are subsequently fed into various machine learning (ML) models like support vector machines (SVMs), Random Forest (RF), or neural networks (NN) for classification. In pursuit of heightened interpretability, a cornerstone in medical AI applications, we employ the Grad-CAM (Class Activation Map) tool to visually highlight the critical regions influencing the model’s decision. Our methodology, though exploratory, showcases a promising avenue for leveraging visualized biomechanical data in stroke diagnosis, achieving a perfect classification rate in our preliminary dataset. The visual inspection of generated images elucidates a clear separation of classes (100%), underscoring the potential of this visualization-driven approach in the realm of small data. This proof-of-concept study accentuates the novelty of visual data transformation in enhancing both interpretability and performance in stroke diagnosis using limited data, laying a robust foundation for future research in larger-scale evaluations.