Predictive Diagnosis Transfer Learning for Telehealth Visit Scoring

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing predictive data analysis solutions are inefficient and unreliable in generating telehealth visit recommendation scores, requiring computationally expensive operations and lacking in computational efficiency.

Innovation Solution

An end-to-end machine learning framework utilizing a diagnosis prediction model, a hybrid diagnosis-provider classification model, and a telehealth visit recommendation scoring model to generate telehealth visit recommendation scores through transfer learning, reducing the need for redundant computations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If existing predictive data analysis solutions are used to generate telehealth visit recommendation scores, then the scores can be produced, but the computational cost is expensive and efficiency is low

Engineering Contradiction:

Improvecomputational efficiencyVSAvoidcomputational overhead

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The system segments the telehealth recommendation task into three distinct machine learning models: a diagnosis prediction model that processes service request data, a hybrid classification model that categorizes diagnoses and providers, and a recommendation scoring model that generates final scores. This segmentation allows each model to specialize in specific computations, improving overall efficiency and reducing redundant calculations.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The diagnosis prediction model performs preliminary analysis of service request data to generate predicted diagnoses before the recommendation scoring model processes the data. This preliminary action filters and structures the input data, reducing the computational burden on subsequent models and improving overall system efficiency.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If comprehensive diagnostic analysis is performed to ensure reliable telehealth recommendations, then prediction accuracy improves, but computational complexity increases

Engineering Contradiction:

Improveprediction reliabilityVSAvoidmodel complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system merges multiple functions into a unified machine learning framework where the diagnosis prediction model, hybrid classification model, and recommendation scoring model work together as an integrated system. This merging ensures that comprehensive diagnostic analysis is performed while managing complexity through coordinated model interactions rather than isolated complex processes.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The hybrid classification model acts as an intermediary between the diagnosis prediction model and the recommendation scoring model. It processes predicted diagnoses and provider information, transforming them into structured classifications that the scoring model can efficiently process, thereby maintaining reliability while managing computational complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12424338B2Transfer learning techniques for using predictive diagnosis machine learning models to generate telehealth visit recommendation scores

Publication Date: 2025.09.23 UNITEDHEALTH GROUP INC

AI summary

Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing predictive data analysis operations. For example, certain embodiments of the present invention utilize systems, methods, and computer program products that perform predictive data analysis operations by an end-to-end machine learning framework that performs at least the following steps/operations: (i) a service request data object is processed by a diagnosis prediction machine learning model to generate a probabilistic diagnosis data object, (ii) the probabilistic diagnosis data object is processed by the hybrid diagnosis-provider classification machine learning model to generate a variable-length classification for the service request data object, and (iii) the variable-length classification is processed by a telehealth visit recommendation scoring machine learning model to generate a telehealth visit recommendation score for the service request data object.