Free-Text Event Detection for Digital Therapeutics Reporting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to efficiently detect and report adverse events associated with digital therapeutics due to the decentralized nature of user access and the manual, time-consuming process of reviewing user feedback, which affects user experience and compliance with regulatory requirements.
Innovation Solution
A machine learning architecture utilizing natural language processing and generative models to analyze free text from various channels, classify events, and generate analytics reports in near-real time, enabling automated event detection and reporting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual review of user feedback submissions is used to detect events, then accuracy in determining whether users experienced events can be maintained, but the process becomes time-consuming and cannot scale to large volumes of users and multiple feedback sources
Solution Approach 1:
The patent replaces the manual mechanical review process with an automated machine learning system that uses natural language processing to analyze free text feedback. The ML architecture processes submissions from multiple channels (email, chat, social media) automatically, eliminating the need for human reviewers to manually examine each submission while maintaining accurate event detection through trained classification models.
2Adaptability or versatility
If users are required to submit feedback through multiple channels (email, chat, voicemail), then comprehensive event reporting can be achieved, but the complexity of collecting and processing feedback increases
Solution Approach 1:
The patent implements a universal feedback processing system that handles multiple feedback channels (email, chat, voicemail, social media) through a single integrated ML architecture. The system uses natural language processing to uniformly process free text from all channels, converting diverse input formats into standardized event classifications without requiring separate processing pipelines for each channel.
Solution Approach 2:
The patent introduces an intermediary layer consisting of natural language processing and machine learning models that mediate between diverse feedback channels and the event detection system. This intermediary converts various feedback formats into a unified structure that the classification system can process, simplifying the overall architecture while maintaining comprehensive channel support.
3Ease of operation
If digital therapeutics allow remote access outside clinical environments, then user convenience and accessibility are improved, but the ability to monitor and detect adverse events deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where users remotely submit feedback through multiple channels (chat, email, voicemail) that is automatically processed by the ML system to detect adverse events. This continuous feedback loop enables monitoring of users in remote settings by capturing their experiences and systematically analyzing them for event detection, maintaining surveillance capabilities despite decentralized access.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Aspects of the present disclosure are directed to systems, methods, and computer readable media for executing actions for events associated with use of applications. A computing system can identify free text associated with an application to be evaluated for at least one of a plurality of events associated with a use of the application. The computing system can apply the free text to a machine learning (ML) architecture. The computing system can determine, based on applying the free text to the ML architecture, a value indicating a likelihood of occurrence of an event associated with the use of the application. The computing system can provide to a generative ML model, a model input based on the free text and the value, to obtain data for an electronic document characterizing the event. The computing system can execute an action using the data for the electronic document.