Parkinson’s disease (PD) is an irreparable neurological disorder that results in motor disabilities. The early onset yields mild non-movement related symptoms associated with Parkinson’s. PD in later stages results in motor signs like tremors, seizures, balance disabilities, rigidity etc. which worsens with time. Hence, early diagnosis is a necessity to ensure minimum damage. PD is normally detected manually by neurologists via inspection of its symptoms which may lead to an incorrect diagnosis. Researchers have proposed various automated PD inspection techniques based on MRI scans. However, the artefacts and noise may be induced during MRI acquisition which may deteriorate its quality and may change its anatomy that may lead to misdiagnosis. Hence, pre-processing is a vital phase that may reduce the effect of artefacts and noise thereby enhancing the diagnostic capability of a model. This paper proposes a preprocessing technique named SRSSN relevant for Parkinson’s detection. The performance analysis of the proposed technique has been evaluated using entropy, signal-to-noise ratio and contrast-to-noise ratio. The proposed technique performs better than the state-of-art techniques.

Parkinson’s disease, Pre-processing, Magnetic Imaging Resonance, Quality enhancement