Effective Noise Reduction in Biomedical Speech Signals: A Case Study on Parkinson’s Disease

Speech signals play a vital role in various biomedical applications, aiding in disease diagnosis, monitoring, therapy, and enhancing healthcare services. They can detect multiple medical disorders, including neurological, speech and language, respiratory, pulmonary, neuromuscular, and mental conditions. Advancements in speech analysis and processing techniques significantly contribute to patient well-being and the efficiency of healthcare delivery. This paper focuses on Parkinson’s disease as a case study to propose a comprehensive noise reduction framework combining the Hamming window technique, a Butterworth bandpass filter, and an Least Mean Squares adaptive filter. This approach achieves a substantial signal-to-noise ratio improvement of 22.6 dB, demonstrating its effectiveness in enhancing speech signal quality for biomedical applications, particularly in the context of Parkinson’s disease.