Brazilian researchers have developed a mobile application designed to assist in the rehabilitation of individuals who have suffered a stroke. Using an accelerometer—a sensor that detects movement and inclination—attached to the user’s clothing, the app monitors posture and provides real-time feedback through voice commands, vibrations, or images to help improve body alignment.
Image Credit: Studio Romantic/Shutterstock.com
The device targets hemiparesis, a condition that causes partial muscle weakness or paralysis on one side of the body. This condition often results from strokes but can also stem from neurological disorders such as multiple sclerosis, cerebral palsy, or certain cancers. In addition to muscle weakness, hemiparesis impairs body awareness, making it difficult for individuals to perceive spatial alignment.
The person with hemiparesis loses sensitivity and perception of spatial organization. As a result, they may tip to one side and not realize it, and even experience muscle pain from the misalignment. They may also be unable to perform everyday tasks such as walking, cooking, driving, or climbing stairs without correct posture. In this way, both body awareness and trunk control, which are lost with the stroke, have to be relearned to guarantee the functionality of the upper limbs.
Amanda Polin Pereira, Professor and Study First Author, Department of Occupational Therapy, Federal University of São Carlos
The app’s development began as part of the doctoral research of Olibário José Machado Neto, a FAPESP scholarship recipient at the Institute of Mathematical and Computer Sciences of the University of São Paulo (ICMC-USP). The research was supported by two FAPESP-funded projects (16/50489-4 and 16/00351-6) and involved collaboration with the Ribeirão Preto School of Medicine (FMRP-USP).
From a software development point of view, it was a co-development job that really brought together two different areas of knowledge to meet the needs of patients. This gave us a lot of agility and support in developing the application, which is completely unique. There’s nothing like it in the clinic to help treat and rehabilitate these patients.
Maria da Graça Campos Pimentel, Professor, Institute of Mathematical and Computer Sciences of the University of São Paulo
The app was tested in a rehabilitation center, where researchers worked closely with physiotherapists and occupational therapists to tailor the software to real treatment needs. Patient feedback also played a crucial role in refining the application.
“In addition to generating a wealth of data to further our understanding of hemiparesis in the future, the app helps patients improve their posture during these sessions and allows therapists to focus on other rehabilitation issues, making the process more efficient and accurate. We’re also starting a study on the longer-term use of the app at home,” says Pereira.
Less is More
Over the course of development, three versions of the app were created.
“We started with the idea of a wearable app, but we realized that the simpler it was, the better it would be accepted by patients. That’s why we focused on developing the software, and instead of creating special clothing, we decided to sew pockets into tops or tank tops that could attach the cell phone to the patient’s torso,” said Pimentel.
“Ultimately, we created an accessible wearable technology that uses the resources of low-cost smartphones and built-in accelerometers to continuously monitor changes [from right to left and forward and backward], using the device’s visual, tactile, and auditory feedback to guide patients while standing,” explained the researcher.
The ultimate objective of the project, according to Pimentel, is to release the app for free and benefit from fresh research based on the hemiparesis data.
“What we need now is collaboration to keep the app up to date, something that takes time and financial investment,” said Pimentel.
Journal Reference:
Pereira, P. A., et al. (2025) Wearable Smartphone-Based Multisensory Feedback System for Torso Posture Correction: Iterative Design and Within-Subjects Study. JMIR Aging. doi.org/10.2196/55455