"Seeing" the internal structures of the body during physical rehabilitation

More and more people are living in conditions that can benefit from physical rehabilitation, but there are not enough physical therapists (PTs).

As the population grows, the demand for PTs continues to grow, and aging and the rising incidence of serious diseases exacerbate the problem
.

The rise of sensor-based technologies, such as body motion sensors, has provided patients with a degree of autonomy and precision that they can benefit from robotic systems to complement human therapists
.
However, minimalist watches and rings currently on the market rely mainly on exercise data, and beyond exercise, there is a lack of more comprehensive data cobbled together by physical therapists, including muscle contact and tension
.

This musculomotor language disorder recently prompted researchers at MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) and Massachusetts General Hospital to create an unsupervised physical rehabilitation system, MuscleRehab
.
It has three components: motion tracking that captures motor activity, electrical impedance tomography (EIT), an imaging technology that measures muscle activity, and a virtual reality (VR) helmet and tracking suit
that allows patients to watch themselves behave with the help of a physical therapist.

Patients wear a stylish ninja-style all-black tracking suit and perform various exercises such as arrow steps, knee bends, leg raises, leg raises, knee straightening, squats, hydrants, and bridges to measure the movement of the
quadriceps, sartorius, hamstring, and abductors.
VR captures 3D motion data
.

In a virtual environment, the patient has two situations
.
In both cases, their avatars perform
with the physiotherapist.
In the first case, only the motion-tracking data is overlaid on the patient's avatar
.
In the second case, the patient wears an EIT induction band, and then they have all the information about
movement and muscle bonding.

In both cases, the research team compared the accuracy of the exercises and gave the results to a professional therapist who explained which muscle groups
should be used in each exercise.
By visualizing muscle contact and movement data in these unsupervised exercises, not just movement, the overall accuracy of the workout improved by 15%.

The team then cross-compared
the timing of the correct muscle group being triggered during exercise in both cases.
In cases where they display muscle contact data in real time, that's feedback
.
By monitoring and recording most engagement data, PTs report a better understanding of the quality of patients' movements and help better assess their current status and exercise
based on this data.

Junyi Zhu, a doctoral student in electrical engineering and computer science at MIT, said, "We hope that our sensing scenarios go beyond clinical settings and better provide data-driven, unsupervised rehabilitation for athletes recovering from injuries, patients currently undergoing physical therapy, or people with physically limited medical conditions, and ultimately see if we can not only help with recovery, but may also help with prevention
.
" Junyi Zhu is an affiliate of CSAIL and lead author
of a new paper on MuscleRehab.
"By actively measuring deep muscle contact, we can observe whether the data is abnormal compared to the patient's baseline, providing insight into potential muscle trajectories
.
"

Current sensing technology is primarily focused on tracking behavior and heart rate, but Zhu is interested in finding a better way than electromyography (EMG) to sense the contact of different layers of muscle (blood flow, stretching, contraction).

EMG can only capture muscle activity under the skin, unless it is invasive
.

Professor Zhu has been
researching personal health sensing devices for some time.
He used EIT to measure muscle conductivity in a 2021 project that used noninvasive imaging techniques to create a toolkit for designing and manufacturing health and motion sensing devices
.
As far as he knew, EIT was commonly used to monitor lung function, detect chest tumors and diagnose pulmonary embolism, and had never been used before
.

For MuscleRehab, the EIT sensor board acts as the "brain"
behind the system.
It also comes with two electrode-filled straps that can be put on the upper part of the user's thigh to capture 3D volumetric data
.
The motion capture process uses 39 markers and many cameras
that perceive very high frame rates per second.
EIT sensing data shows that actively triggered muscles are highlighted on the display, and the color of specific muscles darkens
with increasing participation.

Currently, MuscleRehab focuses on the major muscle groups in and inside the upper thighs, but next they want to expand to the hips
.
The team is also working with Piotr Zygmanski, a medical physicist at Brigham and Women's Hospital and Dana-Farber Cancer Institute and associate professor of radiology at Harvard Medical School, to explore potential avenues
for using EIT in radiation therapy.

Zygmanski said: "We are exploring the use of electric fields and currents to detect radiation, as well as to image
the dielectric properties of patient anatomy during radiation therapy treatment or in treatment outcomes.
In addition to causing direct damage at the molecular level (DNA damage), radiation induces electrical currents
inside tissues, cells, and other media, such as detectors.
We found that the EIT instrument developed by the MIT team is particularly well suited to explore this new application
of EIT in radiation therapy.
We hope that by customizing the electronic parameters of the EIT system, we can achieve these goals
.
”

Yang Zhang, assistant professor of electrical and computer engineering at UCLA's Samueli School of Engineering, said: "This work advances EIT, a sensing approach commonly used in the clinic, that combines ingeniously and uniquely with
virtual reality.
" Yang Zhang was not involved in the paper
.
"Enabling apps that promote recovery could have wide-ranging implications for society as a whole, helping patients to rehabilitate
physically safely and effectively at home.
Due to the lack of labor in the healthcare industry, such tools have long been needed to eliminate the need
for clinical resources and personnel.
”