Researchers from the Polytechnic University of Madrid (UPM), in collaboration with University Carlos III of Madrid, Oxford University and Coventry University, developed a system of deep brain stimulation on demand that can be implemented in the neurostimulation systems that treat Parkinson’s. The proposed system is capable of operating with 100% accuracy; which means that it is able to detect without fail the state in which the patient is at each moment.
Parkinson’s disease is currently the second most important neurodegenerative condition in terms of its incidence, but it is estimated that it will become the first by 2040, surpassing Alzheimer’s disease. As noted by the Spanish Parkinson Federation, it is expected that this disease will become a pandemic within 20 years. Therefore, there is a need to investigate and implement new measures to combat it.
To improve the associated symptoms, the first option is based on a pharmacological treatment. However, not all patients respond well to medication. In addition, over time it begins to generate adverse effects such as dyskinesias (abnormal and involuntary movement disorders).
In these cases, a treatment pathway is deep brain stimulation, which involves the implantation of a neurostimulator that provides electrical current through a series of electrodes in the area of the brain that controls movement (the target nucleus, usually the subthalamic core). This makes the neuronal population, altered by the lack of dopamine, recover its functioning. However, the current neurostimulators once implanted stimulate continuously, which induces adverse effects in the patient (such as paraesthesia or cognitive impairment, among others).
“Brain stimulation on demand represents a better functioning strategy, in which the device should stimulate only when the patient needs it, on demand and in real time,” says Carmen Cámara, a researcher at the Center for Biomedical Technology (CTB) of the UPM.
“Developing a system like this requires unraveling the functioning of the brain networks involved, understanding what neuronal behavior is generated in different clinical states —such as when the patient has symptoms, for example tremor, and how it behaves when the patient does not have them ,” Carmen Camera continues.
A system with total precision
To carry out this work, researchers have studied this behavior by observing the synchronization of neurons in different clinical states through mathematical methods of functional connectivity. They have found that when the patient has tremor, the synchronization of the neurons changes.
Researchers hope that systems like the one they propose will be incorporated into clinical practice in the coming years.
This change in neuronal communication can be used as a decision element so that the device knows when to start stimulation. The system that researchers have developed has been designed under the paradigm of artificial intelligence or data stream mining.
“This is a type of novel algorithm capable of working in demanding scenarios, having to process and offer a quick response. Such is the case of neurostimulators, which record brain signal continuously throughout the patient’s life, with permanent monitoring and decision making being necessary,” says the researcher.
The proposed system is capable of working with total precision, being able to detect, without failure, the state in which the patient is. Otherwise, there may be situations in which the patient is left without stimulation, a situation to avoid.
Given the prediction in the growth of Parkinson’s disease, the researchers hope that systems like the one they propose will be incorporated into clinical practice in the coming years. For them, the results are promising and represent an important step in the development of intelligent systems that provide a better treatment.