Severity Prediction over Parkinson's Disease Prediction by Using the Deep Brooke Inception Net Classifier

Area of Science:

• Neurology
• Biomedical Engineering
• Data Science

Background:

• Parkinson's disease (PD) is a progressive neurodegenerative disorder impacting motor functions due to dopamine neuron loss.
• Symptom severity in PD gradually increases over time, necessitating accurate and timely assessment.
• Analyzing voice data presents challenges due to unintelligible information in raw recordings.

Purpose of the Study:

• To develop and evaluate a novel strategy for predicting the severity of Parkinson's disease using voice data.
• To improve the accuracy of PD severity prediction compared to existing methods.
• To establish voice abnormality classification as a reliable metric for PD severity.

Main Methods:

• Utilized the Parkinson's Telemonitoring Voice Data Set from UCI.
• Preprocessed raw speech signals using signal error drop standardization.
• Employed wavelet cleft fuzzy algorithm for feature grouping and firming bacteria foraging algorithm for feature selection.
• Classified disease severity using a deep brooke inception net (DBIN) classifier.

Main Results:

• The proposed DBIN model demonstrated superior accuracy in detecting PD severity compared to conventional methods.
• Classification based on extracted voice abnormality data achieved a high accuracy of 99.8% for PD prediction.
• The developed strategy effectively addresses challenges posed by unintelligible data in raw speech recordings.

Conclusions:

• The proposed deep neural network strategy offers a highly accurate method for predicting Parkinson's disease severity.
• Voice abnormality analysis using the DBIN model is a promising and effective metric for PD severity assessment.
• This approach has the potential to enhance remote patient monitoring and management of Parkinson's disease.