Researchers at IBM and the Michael J. Fox Foundation have developed a model that can predict the evolution of Parkinson’s disease in people. That is, when and in what form the symptoms will appear as the disease progresses.
This information would be essential to provide patients with a treatment that helps control symptoms as the disease progresses.
This is not the first time we have seen the potential of AI being used to help Parkinson’s patients, but each of the initiatives has a different perspective that contributes to making great strides in improving the quality of life of patients.
For example, we saw how a group of engineers at Imperial College is using AI and wearable to monitor and treat the symptoms of Parkinson’s disease. And now IBM’s AI is proposing the same goal, but anticipating the symptoms, predicting how the disease will evolve.
Our goal is to use artificial intelligence to help with patient management and clinical trial design. These goals are important because, despite the prevalence of Parkinson’s, patients experience a unique variety of motor and non-motor symptoms.
The Michael J. Fox Foundation was a partner in this effort, which made it possible to draw on an anonymized dataset with information from more than 1,400 individuals.
The dataset served as input for the machine learning approach, which enabled the discovery of patterns of progression and complex symptoms. While many previous studies have focused on characterizing Parkinson’s disease using only baseline information, our method is based on up to seven years of patient data.
As a result, they have developed an Artificial Intelligence that groups disease symptoms and establish patterns to predict disease progression, taking into account that it does not progress in the same way in all patients.