Patients learn to self-rehabilitate through new AI system

A new system will help patients rehabilitate themselves in a more efficient and cost-effective manner. Developed by students and researchers at the United Arab Emirates University, in collaboration with the intelligent behaviour control unit at the RIKEN National Science Institute in Nagoya, Japan, the project uses technology and artificial intelligence (AI) to have easier and quicker access to neuro-therapists.

“With the technology, we can actually overcome this limitation of not having enough high-quality neuro-therapists,” said Dr Fady Najjar, Assistant Professor at the College of Information Technology. “The idea is to use technology with cloud computing to extend the power of these people – we can decrease the work of each doctor and let him do more service work.”

The AI system will perform routine work and learn from the doctor’s feedback to be able to adapt it on its own. “The system already learnt how to solve this problem from the therapist’s long-term experience, which will allow it to take decisions without consulting the doctor,” Dr Najjar said. “The idea first started with a few therapists in Nagoya, Japan, where residents living around have no access to good quality neuro-therapists. They need to travel long distances to meet them, so we found a way for them to do the training in their house.”

Their data is recorded and uploaded on the internet in the cloud, where doctors can access it online. Based on the data, the system can monitor and give feedback to the patient, as well as change his training, based on his evolvement. Doctors will also be able to access the data through their dashboard to monitor the number of registered patients.

To motivate the patients to perform the training, students at the university flew to the Japanese institute early March for two months to build interactive training games. “If we just ask them to wear sensors, they may do it once or twice but not continually,” Dr Najjar said. “But if they do it as part of a game, they will most likely continue the games every day to see where they reach.”

Students also plan to involve social media, whereby competitions will take place between patients for them to track the number of hours they train every week. According to Dr Najjar, this will further challenge them and give them the idea that they are not alone in their physical evolvement. “Everything exists already,” he added. “We’re just putting it together.”

The system will also learn, through the patient’s input and the doctor’s output, the most probable feedback for any potential issue raised. Once it reaches an expert level, it will attempt to decrease the load on the doctor by automatically sending out the type of training needed by the patient. “We put them all together in a good content instead of these patients having to travel long distances to visit doctors, who are limited due to the newness of the field,” said Hamda Al Shamsi, one of the students working on the project. “At first, there will be many evaluations and feedback but gradually, the system will learn and try to suggest an output without going back to the doctor. This is the main merit of this study – to have more patients with less doctors.”

Students are currently continuing their research in the university’s laboratory, where they are looking to enhance the games, and find ways to build the expert system which will learn from monitoring input and output. This will increase the patients’ chances of meeting an expert instead of going to local hospitals to recover, as they will have more facilities to train better and recover.

According to Jamila Al Nuaimi, another student, the goal is to implement the project to rehabilitate patients who spend years in hospital for treatment. “The benefits include reduced costs, reduced waiting time, improved outcomes of treatment and a generally enhanced patient experience,” she said. “Although there are some challenges, such as security issues by sharing data through the cloud, internet availability or users not knowing how to deal with technology, we will develop a secure encoding system by encrypting the patient’s data and the doctor’s feedback. Our method could also work on a number of different muscles.”

For 23-year-old Maythaa Aleghfeli, a student in information security, there is a real need for the project. “One patient showed us her video and we couldn’t find any existing system using this technology,” she said. “Since many countries, including the UAE and Japan, lack neuro-therapists, this motivated us. It should take us another year before we complete the project.”

Share this news
 
Last updated:  Site best viewed at 1024 x 768 resolution in I.E 9+, Mozilla 3.5+, Google Chrome 3.0+, Safari 5.0+

عفوا

لايوجد محتوى عربي لهذه الصفحة

عفوا

يوجد مشكلة في الصفحة التي تحاول الوصول إليها

Jun 6, 2018