EEG-Based MDD Diagnosis Using Classification Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current diagnostic methods for major depressive disorders, such as fMRI, face challenges in accuracy due to high costs, spatial and temporal restrictions, and failure to consider altered cognitive processes, making it difficult to distinguish between similar symptoms and provide reliable information for diagnosis.
Innovation Solution
A method and device utilizing brain wave data and a classification model to determine the presence of major depressive disorders by generating brain activity data, extracting features, and applying a statistical scoring method to predict the condition, overcoming the limitations of fMRI by considering cognitive characteristics and reducing overfitting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fMRI is used for diagnosis of major depressive disorders, then neural activity information can be obtained, but the diagnosis accuracy is limited due to spatial and temporal restrictions and high costs
Solution Approach 1:
The patent replaces the complex mechanical fMRI system with an EEG-based diagnostic system that uses electrical signal detection and computational processing. The EEG device records brain wave signals electrically, which are then processed through feature extraction and classification algorithms to achieve accurate diagnosis without the spatial and temporal restrictions of fMRI
Solution Approach 2:
The patent creates a computational model that copies and simulates the diagnostic capabilities of fMRI using EEG data. By training a classification model on EEG features that represent neural activity patterns, the system replicates the diagnostic information obtained from fMRI while avoiding its limitations
2Measurement precision
If fMRI focuses only on neural activity during emotional information processing, then neural responses can be measured, but important pathologies such as altered cognitive processes are not considered
Solution Approach 1:
The patent makes the EEG diagnostic system universal by designing it to capture multiple aspects of brain function simultaneously. The EEG records various brain wave frequencies and patterns that reflect both emotional processing and cognitive processes, allowing a single measurement system to provide comprehensive diagnostic information across multiple functional domains
3Productivity
If brain wave data is used for diagnosis, then temporal and spatial restrictions are eliminated allowing continuous monitoring, but accurate classification requires sophisticated feature extraction and modeling
Solution Approach 1:
The patent applies preliminary action by pre-processing the EEG data through filtering, artifact removal, and feature extraction before classification. The system prepares the raw brain wave signals in advance by extracting relevant features such as power spectral density and temporal patterns, which are then fed into the classification model for efficient and accurate diagnosis
Data Source
AI summary
The present disclosure provides a method for providing information on a major depressive disorder, which is implemented by a processor, and a device using the same, the method comprising receiving brain wave data of an individual; generating brain activity data based on the brain wave data; and determining whether the individual's major depressive disorder is present by using a classification model configured to classify the major depressive disorder based on the brain activity data.