کنفرانس مهندسی برق ایران

صفحه اصلی / سی و سومین کنفرانس بین المللی مهندسی برق

A New Protocol to Improve Effect of repetitive Transcranial Magnetic Stimulation in Treatment of Alzheimer's Disease

نویسندگان :

Ali Abedi¹ Gholamreza Moradi² Reza Sarraf Shirazi³ Mehran Jahed⁴

1- دانشگاه صنعتی امیرکبیر 2- دانشگاه صنعتی امیرکبیر 3- دانشگاه صنعتی امیرکبیر 4- دانشگاه صنعتی امیرکبیر

کلمات کلیدی :

repetitive Transcranial Magnetic Stimulation،Machine Learning،Alzheimer Disease،Support Vector Machine،Electroencephalography

چکیده :

Despite significant breakthroughs in the clinical and instrumental evaluation of Alzheimer's Disease (AD) diagnosis, as well as therapeutic efficacy achieved to date, we still face challenges for early public classification. Recent studies show that the use of electroencephalographic (EEG) network analysis allows dynamic brain connectivity to be frozen, and that this is successful in increasing classification accuracy when EEG signals are used together with neuropsychological tests. In conclusion, this study sought to evaluate the therapeutic potential of rTMS using an innovative protocol on cognitive performances in AD treatment. Using an SVM-based classifier on EEG data, we obtained excellent sensitivity = (97%±3%), specificity = (97%±2%), and accuracy = (98%±2%), with AUC= (0.98±0.05) for the classification of healthy controls and AD patients. We have established, to our knowledge, a unique modulation of pulse train, interpulse intervals, and pulse width in rTMS protocol, which will potentiate its therapeutic response. Further, we adopted the Common Mode Features (CMF) approach to delineate common biomarkers between Alzheimer's disease and Parkinson's disease as well as between Huntington's Disease with Amyotrophic Lateral Sclerosis. This method can improve SVM classifier performance by securing diagnostic as well as pan-condition biomarkers and therefore could enhance classification power in a clinical setting. This study was conducted with a total sample of 59 subjects (34 healthy, and 25 AD), proving that rTMS combined with EEG and machine learning can serve as an inexpensive and non-invasive individualized approach to diagnosis improvement or treatment augmentation in cases of Alzheimer's disease.