Concentration level multi-modal prediction method and system, storage medium, and electronic device
By combining multimodal signals for attention prediction, the problems of insufficient objectivity, discontinuity, and low robustness in existing technologies are solved, enabling real-time, accurate monitoring and personalized feedback of attention, thereby improving work efficiency and learning outcomes.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-06-10
- Publication Date
- 2026-03-26
AI Technical Summary
Existing attention prediction methods lack objectivity, are inconsistent, exhibit significant individual differences, and have low robustness, making it difficult to effectively monitor and assess individual attention in complex environments.
By combining multimodal information such as EEG signals, ECG signals, EEG signals, and physiological signals, attention prediction is performed using a deep learning model through preprocessing and feature extraction. This includes bandpass filtering, baseline drift removal, short-time Fourier transform, and multimodal feature extraction. Models such as SVM, decision tree, random forest, CNN, LSTM, or EEGNet are used for training.
It enables real-time and continuous monitoring of focus, improves the accuracy and robustness of predictions, provides personalized feedback to individuals, and enhances work efficiency and learning outcomes.
Smart Images

Figure CN2025100295_26032026_PF_FP_ABST
Abstract
Citation Information
Patent Citations
A driver fatigue detection system based on electroencephalogram (EEG) signals
CN102274032A
In-class student learning status assessment system based on wearable physiological signal monitoring
CN110916631A
Concentration degree evaluation method and device
CN111160239A
Fatigue detection method and system integrating electroencephalogram, electro-oculogram and heart rate
CN114246593A
Attention assessment method, system and equipment based on multi-modal fusion and medium
CN116881853A