<p>The rapid growth of the Internet of Things (IoT) has enabled diverse healthcare applications to provide ease of access to medical services. Medical data analysis, service composition, and user augmentation are effectively handled using IoT elements. Electronic health records (EHR) are stored and retrieved without errors, and swift data amendments ensure reliable services. In this article, a Fine-Grained Data Analytical Model (FGDAM) is introduced to alleviate computational errors in medical data collection. This model focuses on replicated and unclassified medical input data that consumes multiple computational occasions. Such data is identified and exempted for precise error-free analysis. The unclassified-to-classified recurrences are used for augmenting consecutive data aggregations. This recurrence is identified using federated learning to prevent data unavailability and time lags. The learning relies on multi-edge data aggregators for preventing losses and amendments. Therefore, the application support is continuous, with fewer possible errors, thereby leveraging improved healthcare performance.</p>