Closed-loop therapy control with sensor confidence evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Medical devices face challenges in delivering accurate therapy due to loss of sensor data, which can lead to the selection of less effective therapy parameters, as they rely on inaccurate or incomplete sensor data for parameter determination.
Innovation Solution
The implementation of a hierarchical arrangement of control policies and confidence level evaluation to determine therapy parameter values, where confidence levels are assessed based on the accuracy of sensor data, allowing for interpolation of missing or out-of-band samples and default therapy parameter selection when data accuracy is low.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If sensor data is used to determine therapy parameters in autonomous adaptive therapy systems, then therapy delivery is automated and responsive to patient condition, but accuracy of therapy parameter selection deteriorates when sensor data is lost or inaccurate
Solution Approach 1:
The system implements feedback by evaluating the confidence level of sensor data before using it to adjust therapy parameters. When sensor data confidence is high, the system automatically adapts therapy; when confidence is low or data is lost, the system falls back to previously determined parameters, ensuring reliable therapy delivery throughout the automation process.
Solution Approach 2:
The system prepares for potential sensor data loss by establishing a fallback mechanism in advance. Default therapy parameters are determined before sensor data may be lost, and the system is pre-configured to switch to these default parameters when sensor data confidence drops, cushioning against the reliability deterioration that would otherwise occur during data loss events.
2Adaptability or versatility
If the system relies on sensor data for therapy parameter determination, then therapy can be adapted in real-time, but the system may select inappropriate therapy parameters when sensor data accuracy is reduced
Solution Approach 1:
The system dynamically adjusts its behavior based on sensor data confidence levels. When confidence is high, the system fully utilizes real-time sensor data for therapy adaptation. When confidence decreases, the system transitions to a more conservative mode using default parameters, maintaining measurement precision awareness throughout the adaptation process.
Solution Approach 2:
The confidence level evaluation acts as an intermediary between sensor data and therapy parameter selection. This intermediary assesses the accuracy of sensor data and mediates whether the sensor data should directly influence therapy parameters or whether default parameters should be used instead, preventing inappropriate parameter selection when sensor accuracy is compromised.
3Reliability
If multiple sensing sources are used to generate sensor data, then data accuracy and reliability improve, but system complexity increases
Solution Approach 1:
The system merges data from multiple sensing sources by evaluating their combined confidence level to determine therapy parameters. Rather than processing each sensor independently, the system combines their inputs through a unified confidence evaluation process, achieving improved reliability while managing complexity through integration.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A medical device may receive sensor data from sensing sources, and determine confidence levels for sensor data received from each of the plurality of sensing sources. Each of the confidence levels of the sensor data from each of the sensing sources is a measure of accuracy of the sensor data received from respective sensing sources. The medical device may also determine one or more therapy parameter values based on the determined confidence levels, and cause delivery of therapy based on the determined one or more therapy parameter values.