Investigational Clinical Decision Support

Early voice screening for
Parkinson's vs Essential Tremor

VISTA-PD captures 14 acoustic biomarkers from a 3-minute voice protocol and aids primary care clinicians in differentiating early Parkinson's disease from Essential Tremor.

See how it works Join the study

Investigational · Not FDA-cleared · For research use only · HIPAA-compliant data handling · Zero PHI retained

Protocol

3 minutes. 2 tasks. 14 features.

A clinician-administered voice protocol captures the acoustic signatures associated with hypokinetic dysphonia and vocal tremor — two hallmarks that separate PD from ET years before motor symptoms become unambiguous.

1

Sustained phonation

Patient holds the /a/ vowel for 5 seconds. Eight Praat-derived features are extracted: jitter, shimmer, HNR, CPPS, F0 mean/std, and 4–8 Hz vocal tremor.

2

Reading passage

Patient reads a standardised passage (EN/ES/ZH/HI/PA). Six features are extracted: F0 prosodic std, pause count, pause-to-speech ratio, and pause duration statistics.

3

Acoustic analysis

Audio is processed on HIPAA-covered Cloud infrastructure. Features are extracted; audio is cryptographically wiped before any data leaves the server.

4

Clinical output

Phase 1 builds the labelled dataset. Phase 2 returns a PD vs ET probability with top-3 SHAP biomarker drivers for the clinician to review.


Signal Science

What the algorithm listens for

Each feature maps to a known pathophysiological mechanism. The classifier weights them jointly — no single feature is diagnostic.

Feature Task Clinical significance
Jitter %PhonationMicro-period perturbation from basal ganglia hypokinesia
Shimmer %PhonationAmplitude flutter from reduced vocal fold closure force
HNR (dB)PhonationHarmonic-to-noise ratio — reduced in breathy PD voice
CPPS (dB)PhonationCepstral peak prominence — key hypokinetic dysphonia marker
F0 mean / stdPhonationMonotone pitch (reduced std) characteristic of PD
Vocal tremor 4–8 HzPhonationResting tremor frequency band in the voice signal
F0 prosodic stdReadingReduced prosodic variability in connected speech
Pause / speech ratioReadingMotor planning gaps — festination and speech freezing
Pause count / durationReadingBradykinesia signature in speech timing

Privacy Architecture

Zero-retention by design

PHI never leaves HIPAA-covered infrastructure. Audio never leaves the analysis server. Nothing identifiable is stored anywhere.

🌐

Stateless frontend

The web UI (vista-pd.com) stores nothing. No cookies, no local storage, no database. Study IDs only — no names, DOB, or MRN.

🔒

Ephemeral audio handling

Audio streams to GCP Cloud Run, is written to an ephemeral /tmp path, features are extracted, then the file is overwritten with cryptographic random bytes before deletion — before any response is returned.

✓ GCP HIPAA BAA covered
📊

Features only, never audio

The dataset bucket stores a JSON object of 14 numeric features per participant — never audio. A study ID links the row to the clinical outcome label provided by the enrolling clinician.

📱

iOS + Android app

The VISTA-PD mobile app records uncompressed 44.1 kHz WAV, transmits once over TLS, then discards the recording from device memory. Mic permission is granted only during recording.


Interested in the study?

We are recruiting primary care clinicians and patients with suspected early PD or ET. Enrollment is open at participating sites.

Contact the research team