Occult Sepsis Detection with Composite Clinical Scoring and ML
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing clinical risk scores have poor performance in detecting occult sepsis, leading to delayed diagnosis and increased risk of severe complications due to subtle and non-specific symptoms.
Innovation Solution
A computer-implemented method and system for determining the risk of occult sepsis by obtaining sepsis scores and analyzing subject data against occult sepsis criteria, including vital signs and laboratory results, to provide early detection and intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If simple rule-based clinical risk scores (SIRS, qSOFA, MEWS) are used for sepsis screening, then the screening process is simple and quick, but the detection accuracy for occult sepsis is poor
Solution Approach 1:
The patent segments the sepsis detection process into multiple independent analysis modules: traditional score calculation (SIRS, qSOFA, MEWS), machine learning model prediction, and feature importance analysis. Each module operates independently and contributes to the overall detection result, allowing the system to maintain operational simplicity while improving detection accuracy through multiple parallel assessment pathways
Solution Approach 2:
The patent creates a composite detection system that combines traditional clinical scoring methods with machine learning algorithms. This composite approach integrates the interpretability of rule-based scores with the predictive power of ML models, achieving both operational simplicity and high detection accuracy for occult sepsis by leveraging the strengths of both methodologies
2Productivity
If traditional sepsis scores are used, then the diagnostic process is fast, but false negatives occur due to subtle and non-specific symptoms of occult sepsis
Solution Approach 1:
The patent performs preliminary analysis by calculating traditional sepsis scores and identifying high-risk patients before applying more sophisticated machine learning models. This preliminary action filters the patient population and allows rapid identification of obvious cases while directing more complex computational resources to patients with subtle symptoms, maintaining fast diagnosis speed while reducing false negatives
Solution Approach 2:
The system incorporates feedback mechanisms where the results from traditional scores inform the machine learning model input, and the feature importance analysis provides feedback on which clinical parameters are most predictive. This multi-layered feedback loop continuously refines the diagnostic accuracy while maintaining efficient processing speed through intelligent resource allocation
3Measurement precision
If a comprehensive analysis of multiple parameters is performed to improve occult sepsis detection, then the detection accuracy improves, but the system complexity increases
Solution Approach 1:
The patent implements a universal machine learning model framework that can process multiple different input parameters (vital signs, laboratory results, traditional scores) through a single integrated system. This multi-functional approach allows the system to achieve high detection accuracy by analyzing comprehensive patient data without proportionally increasing system complexity, as the same computational infrastructure handles diverse data types
4Loss of time
If early detection of occult sepsis is achieved, then treatment time is reduced and clinical outcomes improve, but more advanced detection methods are required
Solution Approach 1:
The patent introduces machine learning models as intermediary components that bridge traditional clinical assessment and occult sepsis detection. These ML intermediaries process complex patterns in patient data that are not apparent through traditional scoring alone, enabling early detection of occult sepsis while maintaining a structured, clinically-integrated workflow that limits the perceived complexity for end users
Data Source
AI summary
A computer-implemented method of determining, for a subject, a risk of having an occult sepsis is described. The method comprises obtaining at least one sepsis score indicative of a risk of the subject developing a sepsis event; confirming that the obtained at least one sepsis score meets at least one threshold value for the at least one sepsis score; and analysing a set of subject data indicative of a health state of the subject with respect to one or more occult sepsis criteria indicative of a risk of the subject having an occult sepsis.