Neurological Assessment via Mobile Biometric ML
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing and assessing neurological disorders are subjective, cumbersome, and lack objective, remote assessment tools for quantifying motor symptoms like tremor, muscle tone, and rigidity, which are critical for monitoring disease progression and treatment effects.
Innovation Solution
A computer-implemented method using machine learning models that analyze biometric activity data from mobile devices to generate scores for neurological disorder classification, incorporating muscle activity data from sensors to assess tremor, muscle tone, and rigidity, enabling remote and objective monitoring of neurological disorders.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional subjective assessment methods are used by expert neurologists, then diagnostic accuracy can be maintained through expert interpretation, but the process becomes cumbersome and lacks objective quantification capabilities
Solution Approach 1:
The patent replaces the mechanical/manual system of expert neurological examination with an automated computational system using machine learning models. The system processes biometric data from mobile device sensors (accelerometers, gyroscopes, microphones) to objectively quantify motor symptoms like tremor, rigidity, and bradykinesia, eliminating the need for subjective expert interpretation while maintaining diagnostic accuracy
Solution Approach 2:
The system enables patients to perform self-assessment at home using their own mobile devices. The automated machine learning models process the biometric data without requiring expert clinician involvement for data collection or initial analysis, allowing patients to independently monitor their neurological symptoms and facilitating remote patient monitoring
2Reliability
If detailed neurological examination by expert neurologist is performed, then proper diagnosis and severity assessment can be achieved, but the process requires significant time and expert resources
Solution Approach 1:
The system performs preliminary neurological assessment automatically through machine learning analysis of biometric data collected during routine patient activities. This preliminary assessment provides objective quantification of motor symptoms and generates diagnostic recommendations before expert clinician review, reducing the time required for detailed examination while maintaining diagnostic reliability
Solution Approach 2:
The patent introduces an intermediary automated analysis system that processes biometric data and generates diagnostic recommendations. This intermediary system acts as a bridge between raw sensor data and expert clinician interpretation, providing pre-processed, objectively quantified symptoms that accelerate the diagnostic process while maintaining accuracy
3Ease of operation
If remote assessment tools are implemented, then patient monitoring accessibility is improved, but the ability to capture comprehensive biometric data for accurate neurological assessment is limited
Solution Approach 1:
The patent leverages the universal multi-functionality of mobile devices, which already contain multiple sensors (accelerometers, gyroscopes, microphones, cameras) for various purposes. By repurposing these existing sensors for neurological assessment, the system achieves comprehensive biometric data collection for remote monitoring without requiring specialized medical equipment, thus maintaining both accessibility and data completeness
Solution Approach 2:
The system merges multiple data sources from the mobile device (motion sensors, audio sensors, visual sensors) into a unified neurological assessment framework. By combining biometric data from different sensor types, the system captures comprehensive information about motor symptoms including tremor frequency, rigidity, bradykinesia, and postural instability, achieving complete neurological assessment through remote monitoring
Data Source
AI summary
Embodiments herein disclose computer-implemented methods, computer program products and computer systems for performing neurological diagnostic assessments. The computer-implemented method may include processors configured for receiving biometric activity data corresponding to user extremity movement from a mobile device associated with a user. Further, the computer-implemented method may include processors configured for transmitting the biometric activity data to a machine learning model. Furthermore, the computer-implemented may be configured for processing, using the machine learning model, the biometric activity data to generate first model output data corresponding to a first score. Even further, the computer-implemented method may include processors configured for determining that the first model output data corresponds to a neurological disorder classification based at least on the first score exceeding a predetermined threshold.