Physiotherapy Assistance for Patients Using Human Pose Estimation With Raspberry Pi
Year: 2024
Type: article
Abstract: In this work, we employed a device that utilizes Raspberry Pi 4, a camcorder constituent, and a set of audio apparatus to provide real-time assistance to patients during rehabilitation exercises. A person’s lifestyle and physical activity explicitly influence their cerebral health. Exercise routines are crucial for maintaining a proper hormone level and physical fitness. Therefore, the workout routine must be constantly examined and adjusted if any changes are needed. With the help of this device, patients may perform their exercises without a physiotherapist. A physiotherapist can show how to perform the exercises during the first few appointments; after that, the patient can utilize the system to track their routines. This will prevent injuries caused by performing exercises inaccurately when not under the guidance of a medical practitioner. The device monitors how frequently a certain exercise is performed and guides the patient in performing the exercises correctly, promoting quicker recovery. The voice generated also helps the patients analyze and correct the exercises if needed. When detecting a slump, an alarm is triggered to alert the individual. We focused on human pose detection using the OpenCV and MediaPipe libraries to capture and dissect in real-time accurately. OpenCV and MediaPipe libraries were used to capture and detect poses accurately in real time.
Authors: Khasim Vali Dudekula, Maruthi Venkata Chalapathi Mukkoti, Venkata Pavan Kumar Yellapragada, Purna Prakash Kasaraneni, Pradeep Reddy Challa, Dhiren Gangishetty, Manaswita Solanki, Rohith Singhu
Institutions: VIT-AP University, Koneru Lakshmaiah Education Foundation
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4400564129