Tremor Recognition via Convolutional Neural Network Spiral Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for recognizing tremor symptoms, such as subjective rating scales and invasive electrophysiological techniques, lack accuracy and patient convenience, while accelerometer-based devices are prone to motion artifacts.
Innovation Solution
A method utilizing a pre-trained convolutional neural network regression device that evaluates tremor levels from spiral graphs drawn by patients, processed through image preprocessing and enhancement techniques, providing a non-invasive and accurate assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If subjective rating scales are used for tremor recognition, then patient convenience is improved, but measurement precision deteriorates due to influence from anxiety, prejudice, or poor memory
Solution Approach 1:
The patent replaces the subjective mechanical assessment system (patient self-rating) with an objective image-based recognition system using convolutional neural networks. The system captures images of tremor symptoms and uses deep learning algorithms to objectively quantify tremor severity, eliminating the influence of patient anxiety, prejudice, or memory issues while maintaining ease of use.
2Measurement precision
If invasive electrophysiological techniques are used for tremor recognition, then measurement precision is improved, but device complexity and patient discomfort increase due to electrode insertion requirements
Solution Approach 1:
The patent replaces invasive electrophysiological measurement systems with non-invasive image-based recognition. By using cameras to capture tremor symptoms and applying convolutional neural networks for analysis, the system achieves accurate tremor quantification without requiring electrode insertion, thereby eliminating device complexity and patient discomfort associated with invasive procedures.
3Device complexity
If accelerometer-based devices are used for tremor recognition, then device complexity is reduced, but measurement precision deteriorates due to motion artifacts affecting the measurements
Solution Approach 1:
The patent introduces image processing and deep learning algorithms as intermediaries between the simple camera input and tremor measurement output. By capturing images of tremor symptoms and using convolutional neural networks to extract meaningful features while filtering out motion artifacts, the system maintains device simplicity while achieving high measurement precision.
Data Source
AI summary
The present application relates to a method, apparatus and system for recognizing a tremor symptom, a recognition terminal and a storage medium. The method includes: receiving an image to be recognized which is uploaded by a user terminal, where the image to be recognized includes spiral graph used for recognizing whether a drawing person has a tremor state or not and evaluating a tremor level; and taking the image to be recognized as an input value of a pre-trained convolutional neural network regression device to obtain a tremor level. By using the method, the tremor can be recognized by means of images, which is non-invasive and patient-friendly, and since a model is a deep learning algorithm trained on a large spiral image dataset, results of evaluating the severity of essential tremor are accurate and consistent.