Computer-Aided Patient Stratification for Radiology Workload Balancing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional medical diagnosis systems, such as radiology information systems and picture archiving and communication systems, do not consider the difficulty level of patient cases, leading to inefficient allocation of complex cases to junior physicians and uneven workload distribution among radiologists.

Innovation Solution

A computer-aided stratification system that retrieves patient images, analyzes demographic and clinical information, and calculates a stratification score to rank cases by difficulty, enabling the system to assign cases based on predicted diagnostic difficulty, thereby optimizing workload and ensuring complex cases are handled by senior personnel.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If cases are sorted only by imaging modality and specialty in conventional systems, then the system structure remains simple, but workload distribution becomes uneven and junior physicians receive difficult cases leading to potential misdiagnosis

Engineering Contradiction:

Improvediagnostic accuracyVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system performs preliminary analysis of case difficulty using machine learning models before cases are assigned to physicians. The stratification score is calculated in advance based on image features, abnormality characteristics, and clinical information, allowing cases to be pre-categorized by difficulty level before the diagnostic workflow begins

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

A computer-aided stratification module acts as an intermediary between the case management system and physicians. This module calculates difficulty scores and generates stratification recommendations, serving as a mediator that translates complex case characteristics into actionable assignment guidance without replacing physician judgment

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If difficult cases are assigned to junior physicians to maximize utilization, then productivity increases, but the likelihood of misdiagnosis increases

Engineering Contradiction:

Improvephysician workload utilizationVSAvoiddiagnostic accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system applies different case assignment strategies to different difficulty levels. Easy cases are assigned to junior physicians for training and efficiency, while difficult cases are assigned to senior physicians or flagged for double-reading. This localized quality approach ensures that each physician receives cases appropriate to their expertise level

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system changes the assignment parameter from simple random or round-robin distribution to difficulty-based stratified assignment. By introducing the stratification score as a new parameter, the system transforms the assignment logic to balance both productivity and diagnostic accuracy based on case characteristics

Inventive Principle:

Principle #35Parameter changes

3Reliability

If all cases are reviewed by senior physicians to ensure accuracy, then diagnostic reliability improves, but the time and cost of diagnosis increases significantly

Engineering Contradiction:

Improvediagnostic accuracyVSAvoiddiagnosis time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

Instead of applying full senior physician review to all cases, the system applies partial action by selectively flagging only difficult cases for additional review or senior physician assignment. Easy cases proceed through the normal workflow with standard assignment, avoiding unnecessary time consumption on cases that do not require elevated scrutiny

Inventive Principle:

Principle #16Partial or excessive action

4Productivity

If cases are distributed evenly among radiologists without considering difficulty, then workload balance is achieved, but complex cases may be missed or misdiagnosed

Engineering Contradiction:

Improveworkload distributionVSAvoidcase assessment accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system transitions from static even distribution to dynamic stratified assignment. The stratification score is calculated based on real-time case characteristics, allowing the system to adaptively adjust case assignment based on actual case difficulty rather than following a fixed rotation schedule

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS10585940B2Method and system for computer-aided patient stratification based on case difficulty

Publication Date: 2020.03.10 KONINKLIJKE PHILIPS NV

AI summary

When evaluating patient cases to determine complexity thereof, a computer-aided stratification technique is applied to analyze historical patient case diagnoses and correctness thereof in order to calculate a stratification score (20) for each of a plurality of abnormality types and/or anatomical locations. When a new patient case is received, the computer-aided stratification technique is applied to evaluate the patient case in view of historical data and assign a stratification score thereto. A ranked list (21) of current patient cases can be generated according to stratification scores, and physician workload can be adjusted as a function thereof so that workload is balanced across physicians and/or according to physician experience level.