Tremor Detect
Tremor Detect is an innovative Machine Learning web application that serves as an early detection and tracking tool for Parkinson's Disease (PD) using Machine Learning thereby saving patients from immobilizing consequences. PD is a neurodegenerative illness w
2025-06-28 16:36:28 - Adil Khan
Tremor Detect
Project Area of Specialization Artificial IntelligenceProject SummaryTremor Detect is an innovative Machine Learning web application that serves as an early detection and tracking tool for Parkinson's Disease (PD) using Machine Learning thereby saving patients from immobilizing consequences.
PD is a neurodegenerative illness whose treatment is only effective if caught early. Scarcity of neurologists, costly consultation and the ongoing pandemic make timely consultation difficult.
Tremor Detection Tests:
Tremor Detect provides cost-effective early detection with the convenience of taking tests at home by providing a digitized version of the following two tests:
- Spiral Test – a commonly used method for detecting tremors by asking a person to draw a spiral on the paper.
- Wave Test – a method for detecting tremors by asking a person to draw a wave on paper.
The patients will be required to trace a spiral/wave on the screen and if the person has Parkinson’s Disease, the spiral/wave will show visible signs of tremors in hand. The image will be run through a Machine Learning algorithm after processing and the user will be notified of the diagnosis.
Progress Tracking:
The system will also allow the users to track their progress and determine whether the tremors have reduced or worsened.
Accurate Quantitative analysis of the user’s condition will be calculated based on previous tremor detection tests. A simple and easily understandable figure will be displayed showing the user’s progress.
Following are the objectives of Tremor Detect:
- To be able to detect Parkinson's Disease with the maximum achievable accuracy.
- To be able to develop a responsive web application for usage over a number of different devices.
- To be able to track the progress of patients through quantitative analysis and display in user understandable format.
- To enable users to download their reports which can in turn assist medical practitioners.
- The test results and the progress report should be clearly understandable by the users.
Tremor Detect aims to modernize Parkinson's Disease detection by incorporating a telemedicine component through digitizing the Spiral and Wave Tests for tremor detection.
Machine Learning will be used to develop models for detecting Parkinson's Disease. Pre-available Kaggle Datasets will be used. Alhough these have to be preprocessed and modified according to the project needs, this data will then be used to train and test machine learning models through Microsoft Azure.
Through a responsive web application, user's traced patterns will be processed through digital image processing. The keypoints needed for the tremor detection will be extracted and passed through trained and tested Machine Learning Models for the detection of Parkinson's Disease. Detection will be based on comparison of the extracted keypoints with the pre-available data set. The downloadable test results and the calculated accuracy will be stored in the SQL Database for progress tracking. The system will quantitavely analyze the user's previous results and calculate if the condition of the user is being improved or worsened. A statistical figure will be plotted for better understanding and the reports displayed will be available for download.
Benefits of the ProjectTremor Detect is a replacement of an existing solution to the following problem:
Parkinson’s is a progressive, chronic illness. The cause is unknown and there are no cures. Treatment is effective if the disease is caught early enough and early tremors typically begin in the fingers and hands. To prevent the major negative impact on PD patients it is necessary to detect the PD at the early stage. Because of the delay in early detection caused due to waiting for a doctor’s appointment and scarcity of neurosurgeons, the symptoms can worsen considerably. The whole process of appointments and tests is costly, time-consuming and dangerous during the ensuing COVID-19 pandemic.
To overcome this problem, the Tremor Detect team has an innovative solution that will allow the users to detect Parkinson’s at an early through a digitized spiral/wave test that will allow progress tracking and detection. This solution reduces costs because of no travel and most importantly provides early detection, which is the most crucial factor in effective treatment.
Technical Details of Final DeliverableThe final deliverable of Tremor Detect will be a responsive web application which will detect tremors using a touch interface through Machine Learning algorithms.
The web application will be hosted through Microsoft azure.
Final Deliverable of the Project Software SystemCore Industry ITOther Industries Medical Core Technology Artificial Intelligence(AI)Other Technologies Cloud InfrastructureSustainable Development Goals Good Health and Well-Being for People, Industry, Innovation and Infrastructure, Sustainable Cities and CommunitiesRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 47012 | |||
| Azure Machine Learning Bs-series B2MS | Equipment | 2 | 23506 | 47012 |