<p>Remote rehabilitation of stroke survivors is an efficient continuity to in-person rehabilitation for the recovery of motor functions. However, simultaneous monitoring of multiple stroke patients suffering from different impairments and located in different places requires real-time processing, scalability, versatility and insightful visual data representations. This study presents a 4-tiers fog-based framework for real-time visual analytics for monitoring remote rehabilitation. The objective of this paper is to evaluate the performance of the proposed framework in terms of latency and scalability and the study of its usability by the target end users. OpenTelemetry was used for the evaluation of the performance of the proposed framework. The usability study was conducted involving 32 medical doctors to evaluate the effectiveness, efficiency, and satisfaction of end-users of the proposed framework. In comparison to a Cloud-only implementation, the results of the performance evaluation of the proposed system revealed a significantly lower latency. Additionally, the system showed stable scalability, effectively managing workloads with consistent performance in terms of latency, throughput, processing, and resource utilization across four configurations. The results of the system usability study revealed an effectiveness of 84.48%, an overall relative efficiency of 80.93% and a System Usability Scale (SUS) score of 73.88 (SD = 12.69), which is classified as “Good”. The self-reported feedback provided by the participants showed their interest in adopting the proposed framework for monitoring stroke patients and they made important suggestions for its improvements.</p>