Table 1 Demographic, clinical, and recording characteristics of participants across Dataset-1 to Dataset-5
From: FedOcw: optimized federated learning for cross-lingual speech-based Parkinson’s disease detection
Dataset (Language) | Gender Male/Female | Age Range (Mean ± SD) | Disease Severity (Mean ± SD) | Recording Conditions | Speech Tasks |
|---|---|---|---|---|---|
Dataset-1 (Spanish)40 | PD: 25/25 HC: 25/25 | PD: 33–81 (61 ± 9.4) HC: 31–86 (61 ± 9.5) | UPDRS speech score: 6–93 (37.7 ± 18.3) | Soundproof booth; 44.1 kHz, 16-bit; professional recording setup | Sustained vowels, isolated words, sentence reading, spontaneous speech |
Dataset-2 (Italian)41 | PD: 19/9 HC: 23/14 | PD: 40–80 (67.2 ± 8.7) HC: 19–77 (48.3 ± 23.4) | UPDRS II speech score: 0–4 (1.1 ± 1.2) | Echo-free environment; 15–25 cm mic distance | Text and phrase reading, syllable repetition (/pa/, /ta/), sustained vowels |
Dataset-3 (Chinese)37 | PD: 16/14 HC: 7/8 | PD: 36–86 (60 ± 13.6) HC: 23–72 (51.9 ± 14.1) | Hoehn and Yahr: 1–5 (2.5 ± 0.8) | Smartphone; 10 cm from mouth | Sustained vowels (/a/, /e/), short sentence reading |
Dataset-4 (Czech)43 | PD: 10/ 12 HC: 11/ 11 | PD: 48–82 (64.4 ± 9.6) HC: 41–79 (63.6 ± 10.0) | UPDRS III: 6–34 (15.9 ± 7.6) | Headset mic; 5 cm distance; 48 kHz, 16-bit | Sustained vowels (/A/, /I/) |
Dataset-5 (English)44 | PD: 9/7 HC: 19/2 | – | UPDRS II Part 5: 0–3 (0.8 ± 0.9) | Smartphone (Moto G4); 44.1 kHz, 16-bit | Not specified in full; smartphone-based voice tasks |
