Pulse Oximetry Predictive Scores for Early Preterm Risk Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current clinical practices in neonatal intensive care units (NICUs) utilize only a fraction of the abundant physiologic information from pulse oximetry, failing to effectively predict adverse events in preterm infants such as death, intraventricular hemorrhage, sepsis, necrotizing enterocolitis, bronchopulmonary dysplasia, retinopathy of prematurity, and prolonged NICU stay.

Innovation Solution

Development of pulse oximetry predictive scores (POPS) using algorithms that analyze heart rate (HR) and oxygen saturation (SpO2) data, incorporating demographic and laboratory variables, to predict adverse outcomes in preterm infants by measuring mean and standard deviation, cross-correlation, HR decelerations, entropy, hypoxia, and hyperoxia over specific time periods, and alerting clinicians before overt signs of illness emerge.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If pulse oximetry data is collected and analyzed using advanced algorithms (POPS), then prediction accuracy for adverse events improves, but system complexity and computational requirements increase

Engineering Contradiction:
Improveprediction accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The POPS system segments the analysis by dividing pulse oximetry data into multiple time periods (e.g., first 24 hours, first week, first month) and applies different algorithmic approaches to each segment. This allows the system to achieve high prediction accuracy for specific adverse events while managing computational complexity through modular processing of different time segments.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transforms one-dimensional pulse oximetry measurements into multi-dimensional analysis by incorporating multiple parameters (mean SpO2, standard deviation, entropy, cross-correlation with heart rate) and multiple time dimensions. This dimensional expansion enables more accurate predictions while the structured approach manages the resulting complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Reliability

If comprehensive pulse oximetry analysis is performed over multiple time periods, then early warning capability improves, but computational time and processing resources increase

Engineering Contradiction:
Improveearly warning capabilityVSAvoidcomputational time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary analysis by pre-processing pulse oximetry data as it is collected, calculating baseline statistics and identifying trends in real-time. This preliminary action prepares the data for more intensive analysis only when adverse events are likely, reducing overall computational time while maintaining early warning capability.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The POPS system implements periodic analysis at predetermined time intervals (24 hours, 1 week, 1 month) rather than continuous intensive processing. This periodic approach maintains reliable early warning detection while significantly reducing computational time and resource requirements compared to continuous analysis.

Inventive Principle:
Principle #19Periodic action

3Measurement precision

If POPS is applied to all preterm infants, then identification of highest-risk infants improves, but resource allocation and clinical workload increase

Engineering Contradiction:
Improverisk identification accuracyVSAvoidclinical efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The POPS system applies different levels of analysis intensity to different infants based on their individual risk profiles. High-risk infants receive comprehensive multi-parameter analysis, while lower-risk infants receive streamlined monitoring. This local quality approach improves risk identification accuracy for those who need it most while maintaining clinical efficiency across the entire preterm infant population.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system replaces manual clinical assessment with automated algorithmic analysis of pulse oximetry data. This substitution enables precise risk identification across all infants without proportionally increasing clinical workload, as the computational analysis occurs automatically without requiring additional manual intervention from healthcare providers.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP3451927B1Method, system, and computer readable medium for generating pulse oximetry predictive scores (POPS) for adverse preterm infants
Publication Date: 2025.10.22 UNIV OF VIRGINIA PATENT FOUND
  • EP3451927B1 patent drawingFigure 1~2
  • EP3451927B1 patent drawingFigure 3
  • EP3451927B1 patent drawingFigure 4

AI summary

A method for generating pulse oximetry predictive scores for adverse preterm infants.