Training Data Collection Device for Medical Inference Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies face challenges in efficiently collecting and creating training data for generating inference models, particularly when new events occur, as they require manual effort and burden service providers with classification tasks, lacking efficient methods for collecting information on processes leading to these events.
Innovation Solution
A training data collection device and method that receives patient symptom information, specifies devices capable of acquiring previous time series data, and requests collection of similar data from other users to create training data for generating inference models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual classification tasks are performed by service providers to create training data for new events, then the quality and accuracy of training data can be ensured, but the workload and time consumption for service providers increase significantly
Solution Approach 1:
The system automatically copies and collects time series data from multiple data sources (wearable devices, medical devices, electronic health records) to create training datasets for new events, eliminating the need for manual data collection and classification while maintaining data quality through automated processes
Solution Approach 2:
The system performs self-service by automatically identifying relevant data sources, collecting time series data, and creating training data structures without requiring service provider intervention, thereby reducing both time consumption and manual workload while preserving data quality through automated quality control mechanisms
2Measurement precision
If comprehensive time series data is collected from multiple sources to create training data for new events, then the accuracy of inference models improves, but the complexity of data collection and processing increases
Solution Approach 1:
The system implements a universal data collection framework that can handle multiple data sources (wearable devices, medical devices, EHR systems) through a single standardized interface and processing pipeline, reducing system complexity while enabling comprehensive data collection for accurate inference models
Solution Approach 2:
The system introduces intermediary components including a data collection unit that acts as a mediator between diverse data sources and the training data creation process, standardizing data formats and simplifying the integration of multiple sources while maintaining comprehensive data collection capabilities
3Extent of automation
If existing classification support devices are used to classify target data into pre-prepared classification models, then the classification process can be automated, but the system cannot adapt to new events requiring creation of new inference models
Solution Approach 1:
The system transitions from static pre-prepared classification models to a dynamic framework that can automatically create new inference models for new events by collecting time series data and generating training datasets on-demand, maintaining automation while enabling adaptability to emerging classification needs
Solution Approach 2:
The system performs preliminary actions by automatically collecting and preparing time series data from multiple sources before new inference models are needed, creating ready-to-use training datasets that can be quickly utilized when new events require new classification models, thus maintaining both automation and adaptability
Data Source
AI summary
A training data collection request device, having a communication circuit receives information relating to symptoms of a specified patient, and a processor specifies a device that is capable of acquiring previous time series data of the patient, wherein the processor acquires, for another person who is using the similar type of device to the device that was specified, data and consultation information that has been collected using the similar type of device to create the training data.