Agitation Detection Model Using Patient and Non-Patient Biometric Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining a patient's agitated state lack accuracy, increasing the risk of adverse events such as tube removal or falling, which can lead to patient injury.
Innovation Solution
A learning device and method that acquire both patient and non-patient biometric information to generate an agitation determination model, using machine learning to differentiate between agitated and non-agitated states based on biometric data, thereby improving the accuracy of agitation detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only patient biometric information is used for training the agitation determination model, then the model training process is simpler and faster, but the determination accuracy is insufficient
Solution Approach 1:
The training data is segmented into two distinct categories: patient biometric information (indicating potential agitation risk) and non-patient biometric information (representing normal baseline states). This segmentation allows the model to learn the contrast between agitated and non-agitated states more effectively, improving determination accuracy while maintaining organized training complexity
Solution Approach 2:
Non-patient biometric information serves as an intermediary reference that mediates the learning process. By introducing this external reference group, the model can better distinguish normal variations from true agitation signals, enhancing determination accuracy without directly complicating the patient monitoring system
2Reliability
If the model training uses only patient data, then the training process is more straightforward, but the model cannot adequately distinguish normal fluctuations from agitated states
Solution Approach 1:
The training approach changes the data parameter composition by incorporating non-patient biometric information alongside patient data. This parameter change enables the model to learn more robust features for distinguishing agitated states, improving reliability while the systematic data processing framework manages the increased complexity
3Measurement precision
If a simple determination model is used, then the processing speed is faster and the system is simpler, but the accuracy in predicting problem behaviors is insufficient
Solution Approach 1:
The system performs preliminary action by training the determination model in advance using comprehensive training data including both patient and non-patient biometric information. This pre-training with diverse data establishes a more accurate model structure that can better predict problem behaviors, with the complexity managed through efficient model architecture and training protocols
Data Source
AI summary
A learning device of the present invention is provided with: an acquiring means for acquiring biometric information of a patient who may possibly become in agitation, and biometric information of a non-patient; and a model generating means for using the biometric information of the patient and the biometric information of the non-patient to generate an agitation determination model for determining whether, on the basis of the biometric information of a subject patient, the subject patient has become in agitation or has not become in agitation.