Oxygen Tank Duration Prediction Using ML and Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users of oxygen therapy are often unaware of the remaining duration of oxygen in their tanks, leading to anxiety and potential life risks due to unexpected depletion while away from home.
Innovation Solution
A method and system using machine learning models to determine the remaining duration of oxygen in tanks by collecting data from sensors and user inputs, incorporating features like usage patterns, tank capacity, and environmental factors, and providing notifications through a connected device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If no monitoring system is installed in the oxygen tank, then the device complexity is low, but the user cannot know the remaining oxygen duration leading to safety risks
Solution Approach 1:
The patent introduces a separate computing device that acts as an intermediary between the oxygen tank and the user. The computing device receives data from the tank's sensor, calculates the remaining duration using machine learning models, and provides notifications to the user through a mobile device. This intermediary approach enables safety monitoring without requiring complex integrated systems within the tank itself.
Solution Approach 2:
The oxygen tank is equipped with sensors that automatically monitor its own state (oxygen level, usage rate) without external intervention. The system performs self-diagnosis and self-reporting by continuously collecting data and transmitting it to the computing device, enabling autonomous safety monitoring that reduces the need for user manual checking.
2Measurement precision
If simple sensors are used in the oxygen tank, then the device complexity is low, but the measurement precision of remaining duration is insufficient
Solution Approach 1:
The system collects and stores historical usage data, environmental conditions, and tank characteristics in advance before they are needed for prediction. Machine learning models are pre-trained with extensive datasets to learn patterns of oxygen consumption under various conditions. This preliminary preparation enables accurate real-time predictions without requiring complex hardware during actual operation.
Solution Approach 2:
The system continuously monitors actual oxygen consumption and compares it with predicted values, using this feedback to refine and update the machine learning models. The computing device adjusts predictions based on real-time sensor data and historical patterns, improving measurement precision over time while maintaining relatively simple tank sensors.
Data Source
AI summary
The exemplary embodiments disclose a method, a computer program product, and a computer system for determining a duration of use left in an oxygen tank. The exemplary embodiments may include collecting data of a user and corresponding oxygen tank, extracting one or more features from the collected data, and determining a duration of use left in the oxygen tank based on the extracted one or more features and one or more models.