Intra-individual Variations in Voice Variables Among Individuals With and Without Parkinson's Disease

Introduction Voice changes affect 70 to 90% of individuals with Parkinson's disease (PD). Voice and speech changes in individuals with PD include increased shimmer, increased jitter, reduced harmonics-to-noise ratio (HNR), and changes in fundamental frequency. Studies have identified inter-individual variations in voice variables as PD progresses. However, intra-individual variations in voice variables have not been studied extensively. When individuals without PD experience stress or nervousness, their jitter and shimmer may also increase. Therefore, the purpose of our study was to compare the mean intra-individual variations in voice variables for individuals with PD and those without PD to determine whether the mean intra-individual variations differ between the two groups. Methodology We utilized the "Oxford Parkinson's Disease Detection Dataset," which consists of various biomedical voice measurements, and each individual's voice was measured six or seven times. The changes between each voice measurement for each individual were calculated. Independent samples t-tests were performed to determine significant differences in intra-individual variations in voice variables between individuals with PD and those without PD for all voice variables. Results The independent samples t-tests revealed no statistically significant differences in intra-individual variations between individuals with (n = 24) and without PD (n = 8) for any of the voice variables. For vocal fundamental frequency variables, the mean differences ranged from -1.98e5 (Flo) to 2.58e5 (Fhi). For jitter variables, the mean differences ranged from 1.05e-6 (Jitter(Abs)) to 5.32e-4 (DDP). For shimmer variables, the mean differences ranged from 2.11e-4 (APQ5) to 7.07e2 (Shimmer(dB)). For other variables, the mean differences ranged from -2.76e6 (HNR) to 6.02e4 (spread1). Conclusion Our findings indicate that intra-individual variations in voice variables do not significantly differ between individuals with and without PD. This suggests that intra-individual voice variability may not serve as a distinguishing factor for PD diagnosis. Future research should explore alternative methods to assess intra-individual voice variability and its potential role in PD diagnostics.