DTCWT Based EEG Biomarker Detection and Classification for Amyotrophic Lateral Sclerosis

Amyotrophic Lateral Sclerosis (ALS) is the most common progressive neurodegenerative disorder reflecting the degeneration of upper and lower motor neurons. Motor neurons controls the communication between nervous system and muscles of the body. ALS results in the loss of voluntary control over muscular activities along with the inability to breathe and the maximum life expectancy of affected individual will be 3-5 years from the onset of symptoms. But the lifetime of affected people can be extended by early detection of disease. The usual methods for diagnosis are Electromyography (EMG), Nerve Conduction Study (NCS), Magnetic Resonance Imaging (MRI) and Magneto-encephalography (MEG). But some of these methods may erroneously result in neuropathy or myopathy instead of ALS and some do not provide any biomarker. EEG is comparatively least expensive method and it provides biomarker for ALS detection. ALS is always associated with fronto-temporal dementia (FTD). The spectral analysis of EEG will reveal the structural and functional connectivity alterations of the underlying neural network that occurs due to FTD and it can generate potential biomarkers for the early detection of ALS. A novel algorithm has been developed by exploiting the Dual Tree Complex Wavelet Transform (DTCWT) technique and it can overcome the short comes of existing methods for the analysis and feature extraction of EEG. Deterministic biomarkers were obtained from spectral analysis of EEG and the proposed algorithm provided 100% accuracy for all the test datasets.