Telehealth and remote doctor appointments really took off during the COVID-19 lockdown, but not everything can be done virtually. Children with cerebral palsy usually need to go in for in-person, often hour-long appointments for evaluations on a regular basis.
With new research from the Massachusetts Institute of Technology (MIT), this could change. New technology could allow doctors to assign clinical scores and evaluate progress remotely. Machine learning is also proving accurate in evaluating these young patients.
The New Technology
The research team at MIT developing this new technology works on projects using robotics to help patients with a variety of movement disorders improve strength and mobility. These technologies require the patient to be there in person to work with the robot.
The research group’s latest work involves replacing the in-person visit for some patients. Children with cerebral palsy must be evaluated regularly to assign clinical scores and determine how they are progressing.
The work is based on what is called pose estimation algorithms. These are algorithms that can use a video of someone moving and translate it into a simple series of lines and dots. The algorithm under study currently makes simple, real-time visualizations of a person’s movements that a doctor can evaluate remotely.
The current research project is testing how effectively machine learning can use the same information to assign clinical scores. They used sets of data for people with and without cerebral palsy to help the machine learn. With thousands of data sets on both typical and atypical movements, the algorithm has a wealth of information to learn from.
So far, the results show that machine learning can evaluate a child’s movement almost as effectively as a physician can. As the technology advances, it may even become better at assessing a child’s progress.
The Benefits of Remote Analysis
There are several benefits of using remote analysis for a child with cerebral palsy. Currently, parents must take their children to a doctor’s office or clinic for a session that can last up to an hour. This is time-consuming and often stressful for the child.
The researchers plan to implement this service by developing an app that anyone can download and use. Parents would record their child’s movements on a smartphone and send it to their doctor for analysis.
This would reduce the number of required in-person visits while still providing important clinical information that directs care.
As the technology improves, it could predict how a child will respond to specific types of treatment. Also, the costs of care for a child should decrease with machine learning and with fewer in-office visits.
The current model is designed for children with cerebral palsy, but researchers are hopeful they can apply it to other conditions. They hope it could improve care for children and adults with any type of movement disorder, such as Parkinson’s.
The technology being tested isn’t available yet, but it won’t be long before parents and children will benefit from more remote medical technologies like this.
