Accurate hospital stay length prediction is critical for effective patient care and resource allocation. To improve the precision of hospital stay length prediction, this research presents using the Internet of Things (IoT)-assisted Linear Regression (LR) model. Using IoT devices and sensors installed across healthcare facilities, it collects real-time data on various variables impacting patients' stays, including vital signs, room occupancy, prescription delivery, and staff actions. It builds an LR model that can accurately predict the length of hospital stays by integrating this rich information with past patient records. Due to factors such as the ever-changing nature of patients' ailments, the intricacy of medical treatments, and the unpredictability of healthcare procedures, the proposed method overcomes some of the challenges inherent in predicting the length of a hospital stay. The algorithm improves prediction accuracy by responding to changing conditions in real time by monitoring patient data and ambient variables using IoT devices. The most important factors for stay length prediction to improve the model's performance are chosen using feature selection approaches. The effectiveness of the IoT-assisted LR model is assessed using real-world hospital data and conventional regression methodologies. This method outperforms the competition in estimating how long patients will remain in the hospital, improving patient care, streamlining hospital operations, and making better use of available resources. Challenges include making sure models are accurate and understandable, managing real-time processing needs, and managing patient conditions that might vary widely. © 2025 Elsevier B.V., All rights reserved.