INTRODUCING A DECISION SUPPORT SYSTEM FOR MULTIPLE SCLEROSIS BASED ON POSTURAL TREMOR: A HOPE FOR SEPARATION OF PEOPLE WHO MIGHT BE AFFECTED BY MULTIPLE SCLEROSIS IN THE FUTURE

Multiple Sclerosis (MS) is a progressive and disabling autoimmune disease of the Central Nervous System (CNS). Regretfully, no reliable MS diagnostic test has been proposed so far. Early diagnosis of the disease has significant benefits. Based on studies, it seems that the movement disorders have a good potential to help the researchers diagnose MS in its early stages. The main objective of this study was to propose a decision support system (DSS) for MS based on balance disorder. To this end, 14 MS patients and 20 healthy subjects were recruited in this study. An infrared marker was set on the participants' forehead and in between eyebrows. They were asked to stand silently in front of a black background for 3 min with their feet shoulder-width apart. Their balance disorder was recorded with a camcorder. Then, the video images were analyzed using MATLAB software and an image processing algorithm was used to get the marker position on images. In the next step, appropriate features that had the ability to separate healthy subjects from patients were extracted. Finally, an artificial neural network classifier was designed that could separate the patients from healthy subjects with high accuracy (92.35%). Due to the high accuracy of the classifier, it can be used as a system in order to introduce a DSS for the diagnosis of MS. Besides, an intermediate state between healthy and diseased states was imagined and people belonging to this state were simulated based on their features behaviors. Hence, another classifier was designed with accuracy of 84.8%. It seems that the second classifier is capable of separating people suspected of having MS in the future. The main contribution of this study is the classification of suspected cases of having MS in the future. Moreover, the presented DSS is simple, easy to use and has high accuracy. As the presented test method is simple and the duration of imaging time is only 3 min, it can be performed annually as a routine checkup and immediately after the observation of the slightest disturbances in the postural behavior the person can be asked to visit a specialist for a more detailed examination.