Body Temperature Patch Monitoring for Early CRS Fever Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current temperature monitoring methods for patients undergoing CAR-T therapy are insufficient to detect the early onset of cytokine release syndrome (CRS) due to their low frequency, often missing the initial fever signs that can escalate to life-threatening conditions.
Innovation Solution
Implementing continuous temperature monitoring with a body temperature logging patch and utilizing machine learning systems, such as Multi-step Multi-layer Perceptron (MLP) networks, to analyze temperature data for distinguishing characteristics of CRS fevers, providing real-time alerts and predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If continuous temperature monitoring is implemented, then early detection of CRS fever is improved, but device complexity and data processing requirements increase
Solution Approach 1:
A machine learning system acts as an intermediary between the continuous temperature monitoring device and clinicians. The system receives continuous temperature data, processes it through trained algorithms, and outputs predictions about CRS fever onset. This intermediary handles the complexity of continuous data analysis, enabling early detection without requiring clinicians to directly manage the complex monitoring system.
Solution Approach 2:
The patent replaces manual temperature checking with automated continuous electronic monitoring. Instead of mechanical thermometers used at discrete intervals, an electronic patch continuously measures temperature and transmits data to the machine learning system, enabling automated early detection of CRS fever.
2Measurement precision
If temperature data is collected at high frequency, then measurement precision for detecting CRS onset is improved, but data processing load and energy consumption increase
Solution Approach 1:
The machine learning system processes temperature data at optimized intervals rather than continuously analyzing every data point. The system collects high-frequency temperature measurements but processes them periodically, balancing the need for precise detection with energy conservation. This periodic processing approach maintains measurement precision while reducing overall energy consumption.
3Measurement precision
If machine learning analysis is applied to predict CRS fever, then prediction accuracy is improved, but computational requirements and system complexity increase
Solution Approach 1:
The machine learning system is trained in advance on historical temperature data from patients who developed CRS fever. This preliminary training phase allows the system to learn patterns and characteristics of CRS fever onset before actual patient monitoring begins. When deployed, the pre-trained model can make accurate predictions without requiring complex real-time computational resources, as the heavy lifting was done during the preliminary training phase.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Continuous temperature monitors can be used to collect frequent temperature data over large periods of time. A machine learning system trained with collected temperature data can be used to predict future temperature data for patients. Such predictions can be clinically beneficial for disease states in which fever can be fatal, particularly for those fevers having fast onset. For example, patients undergoing chimeric antigen receptor T-cell (CAR-T) therapy may suffer fevers caused by cytokine release syndrome (CRS). Accordingly, a continuous temperature monitor is used to collect temperature data from CAR-T patients with high risk of CRS (or like patients with high fever risk), and the collected temperature data processed by a machine learning system to predict the patient's future temperature profiles. Results of analysis of the predicted temperature profiles may then be provided to the patient and/or clinician.