Early Detection for Elderly Wellness

AI that reads the
early signs of
Parkinson's

NeuroSketch combines three AI-powered diagnostic tests — spiral drawing, voice analysis, and motor tracking — to detect Parkinson's disease before it's too late.

Launch App ↗ ▶ Watch Demo
7L+
Indians affected by Parkinson's
10%
of global PD cases are in India
1,200
neurologists for 1.4B people
3
AI diagnostic tests combined
Scroll

A silent crisis
hiding in plain sight

Parkinson's affects over 7 lakh Indians, yet the country has fewer than 1,200 neurologists. Early-onset cases, misdiagnosis, and rural healthcare gaps are leaving millions without answers.

🇮🇳
10%
of all global Parkinson's cases originate in India
Source: Amrita Hospitals
👤
1 in 100
people over 60 are affected — with 68% male prevalence
Source: medRxiv
🩺
1,200
neurologists for a nation of 1.4 billion — critically understaffed
Source: medRxiv
⏱️

Early Onset Cases

A significant number of Indian patients develop PD symptoms between ages 22 and 49 — far earlier than commonly assumed.

🌫️

Lack of Awareness

Symptoms like tremor and rigidity are often dismissed as normal aging, causing critical delays in diagnosis and treatment.

📋

No Definitive Test

Diagnosis relies entirely on clinical evaluation — subjective, inconsistent, and prone to misdiagnosis without specialist access.

🏥

Infrastructure Gaps

Inadequate healthcare facilities and financial constraints prevent timely access to specialist care, especially in rural India.

Three tests.
One diagnosis.
Accessible anywhere.

NeuroSketch is an AI-powered web application that runs three parallel diagnostic tests using nothing more than a browser and a camera — no specialist required.

"NeuroSketch combines multiple AI-powered diagnostic tests to enhance early detection of Parkinson's disease — accessible to anyone, anywhere."
01
🌀

Spiral Test

Patients draw a spiral on screen. The AI analyzes kinematic features — stroke velocity, acceleration, and radial velocity — to detect the micro-tremors invisible to the naked eye.

100% Complete
CNN Random Forest SVC
02
🎙️

Voice Assessment

A 5-second voice recording is analyzed for jitter, shimmer, and frequency variations. Parkinson's subtly alters vocal quality long before other symptoms appear.

60% Complete
SVM 22+ Features Streamlit
03
🤲

Motor Detection

Computer vision tracks finger-tapping and hand movements via webcam. Speed, amplitude, frequency, and period are extracted to detect motor impairments in real time.

38% Complete
LightGBM MediaPipe OpenCV

How each test works

1

Input: Hand-drawn Spiral

Patient draws a spiral on a digital canvas — a standard clinical Archimedes spiral test, now digitized.

2

Image Processing

The drawing is converted to binary (black/white) and normalized to remove scale/position bias before feature extraction.

3

Feature Extraction

Kinematic features are extracted: stroke velocity, acceleration, radial velocity, and geometric regularity measures.

4

ML Classification

Logistic Regression, SVC, KNN, and Random Forest classifiers vote on the binary Healthy/Parkinson's classification.

Sample Output — Spiral Test
Classification Result
Healthy
15%
Parkinson's
85%
1

Input: 5-Second Voice Recording

Patient records a sustained vowel sound ("ahhh"). The audio is captured directly in the browser.

2

Signal Processing

Audio waveform is converted to spectrogram. MDVP (Multi-Dimensional Voice Program) features are extracted.

3

22+ Feature Extraction

Jitter (%), Shimmer, HNR, RPDE, DFA, PPE and more — capturing subtle vocal cord irregularities.

4

SVM Classification

Support Vector Machine trained on the Oxford Parkinson's Dataset classifies with ~87% accuracy on test data.

Sample Output — Voice Assessment
⚠ Parkinson's Detected
Key indicators: elevated jitter (0.00997), shimmer (0.05492), and reduced HNR (0.02924) detected in vocal analysis.
Accuracy score on test data: 0.8717948718
1

Input: Webcam Video

Patient performs finger-tapping or hand movement tasks in front of the webcam for ~60 seconds.

2

Keypoint Extraction

MediaPipe Hands tracks 21 3D landmarks per hand frame-by-frame, computing inter-finger distances.

3

Motion Feature Analysis

Speed, amplitude, frequency, and period of tapping are calculated from the distance-over-time signal.

4

LightGBM Regression

A LightGBM regressor predicts PD symptom severity score based on extracted video motion features.

Sample Output — Motor Detection
Expected Classification
Healthy
15%
Parkinson's
85%
⚡ Currently 38% complete — in active development

Built on established
clinical methods

Our diagnostic approach is grounded in the same 15 clinical assessment methods used by neurologists — just made accessible via AI.

🔬 Technology Stack

Spiral Analysis CNN + Random Forest COMPLETE
Voice SVM Oxford PD Dataset 87% ACC
MediaPipe Hands 21-point keypoints CV
LightGBM Motor severity scoring ACTIVE
Streamlit + Keras Web interface DEPLOYED

📋 Clinical Basis — 15 PD Indicators

Speech patterns
Facial movement
Rigidity
Finger tapping
Hand movement
Pronation/Supination
Toe tapping
Gait
Postural stability
Tremor amplitude
87.2%
Voice model accuracy on test dataset
Oxford Parkinson's Disease Detection Dataset

📡 Sources & References

Amrita Hospitals 10% global PD prevalence in India
medRxiv Research Gender distribution & neurologist deficit data
UCI ML Repository Oxford PD voice measurements dataset

12-Week Roadmap

A structured plan to bring all three diagnostic modules together into a polished, unified MVP.

🎙️
Week 1–3

Voice UI Build

Streamlit interface + model integration

🔬
Week 4

Motor Research

CV tremor detection & data collection

🤲
Week 5–10

Motor Dev

Feature extraction, model training, frontend

🔗
Week 10–12

Integration

Combine all tools, internal testing & evaluation

🚀
Week 12

Launch

Documentation, demo video, submission

🖥️

Voice Analysis Tool

Display 22+ vocal features. Integrate UI + model output in unified dashboard.

🎯

Motor Detection Test

CV-based tremor tracking with keypoint extraction and model training.

Diagnostics Integration

Combine spiral, voice, and motor into one simple, unified interface.

User Testing

Feedback-driven design refinement, accessibility improvements, and final polish.

Where we need help

We're a team of young builders with strong foundations — but we're actively seeking mentorship to push NeuroSketch to the next level.

🚩 Current Challenges

Need guidance on advanced model optimization and hyperparameter tuning for our classifiers

Limited experience building video-based motion detection models at scale

Early-stage user interface is functional but not yet user-friendly for elderly patients

Need help scaling our prototype tools into a polished, deployable MVP

🙌 What We're Seeking

Technical Mentorship: Support on CV model architecture, training pipelines, and feature engineering for medical data

Dataset Guidance: Help sourcing and handling appropriate Parkinson's motor/video datasets

UI/UX Mentorship: Creating a clean, intuitive diagnostic flow with accessibility best practices for elderly users

Clinical Validation: Connecting with neurologists to validate our approach and refine diagnostic accuracy

What Our Users Think

Real feedback from evaluators, educators, and healthcare professionals who've seen NeuroSketch in action.

So futuristic, if taken forward will be very helpful!!
SA
Shobhit Agarwal
Evaluator
The idea presented and research done were both extraordinary. This is a really advanced and well done project. Kudos to bytebusters.
KB
Kartikey Bansal
Reviewer
Commendable job. Parkinson's affects day to day activities and if diagnosed early by the ideas mentioned will help patients a lot!! Way to go kids.
SS
Sukhvinder Sodhi
Healthcare Professional
Excellent information… well done Students… keep it up.
ST
Santosh Tiwari
Educator
Appreciate your thoughts on different topics… really impressive work from such young minds.
HS
Harpreet Singh
Reviewer
A brilliant blend of AI and healthcare. The spiral test alone is a clever, non-invasive diagnostic approach.
AK
Ananya Kapoor
Tech Mentor
So futuristic, if taken forward will be very helpful!!
SA
Shobhit Agarwal
Evaluator
The idea presented and research done were both extraordinary. This is a really advanced and well done project. Kudos to bytebusters.
KB
Kartikey Bansal
Reviewer
Commendable job. Parkinson's affects day to day activities and if diagnosed early by the ideas mentioned will help patients a lot!! Way to go kids.
SS
Sukhvinder Sodhi
Healthcare Professional
Excellent information… well done Students… keep it up.
ST
Santosh Tiwari
Educator
Appreciate your thoughts on different topics… really impressive work from such young minds.
HS
Harpreet Singh
Reviewer
A brilliant blend of AI and healthcare. The spiral test alone is a clever, non-invasive diagnostic approach.
AK
Ananya Kapoor
Tech Mentor
Give your feedback here!

Ready to see it in action?

Explore the live app, watch our demo, or follow our journey as we build toward a world where Parkinson's is caught early — for everyone.