Machine Learning Biomarker Models for Early Preeclampsia Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for diagnosing preeclampsia are limited by the need for both hypertension and proteinuria, which can lead to late detection and increased maternal and fetal morbidity and mortality, especially in developing countries, and there is a lack of effective early prediction and prevention strategies.

Innovation Solution

A method involving the analysis of microparticle-associated protein biomarkers in a blood sample using mass spectrometry to assess the risk of preeclampsia, followed by a classification rule to predict and potentially intervene in the condition before clinical symptoms appear, utilizing panels of biomarkers identified through machine learning and curation against protein databases.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If diagnosis requires both hypertension and proteinuria findings, then diagnostic reliability is improved, but detection timing is delayed leading to increased morbidity and mortality

Engineering Contradiction:

Improvediagnostic reliabilityVSAvoiddetection timing

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by measuring microparticle-associated protein biomarkers during routine prenatal visits before the actual onset of preeclampsia symptoms. This allows prediction and early intervention before the condition develops, rather than waiting for diagnostic criteria to be met. The machine learning model analyzes biomarker patterns in advance to identify women at risk, enabling preventive treatment with aspirin before hypertension and proteinuria develop.

Inventive Principle:

Principle #10Preliminary action

2Loss of time

If early prediction methods are developed, then detection timing is improved, but diagnostic precision may be compromised without validated biomarkers

Engineering Contradiction:

Improvedetection timingVSAvoiddiagnostic precision

Core Design Contradiction:

Loss of timeVSMeasurement precision

Solution Approach 1:

The patent applies parameter changes by transitioning from traditional diagnostic parameters (hypertension and proteinuria) to alternative parameters (microparticle-associated protein biomarkers). The machine learning model analyzes multiple biomarker parameters simultaneously to create a predictive score, maintaining diagnostic precision while enabling earlier detection. The model was validated against clinical outcomes to ensure accuracy.

Inventive Principle:

Principle #35Parameter changes

3Loss of time

If machine learning models are used for prediction, then early detection capability is improved, but system complexity increases

Engineering Contradiction:

Improveearly detection capabilityVSAvoidsystem complexity

Core Design Contradiction:

Loss of timeVSDevice complexity

Solution Approach 1:

The patent applies universality by designing a machine learning model that can be integrated into existing electronic health record systems and used across diverse clinical settings. The same model architecture processes biomarker data from different sources and populations, providing universal early prediction capability without requiring separate complex systems for each clinical environment.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enables early detection and risk assessment of preeclampsia with high sensitivity and specificity, allowing for timely interventions to reduce maternal and fetal complications, and provides a basis for personalized therapeutic interventions.

Implementation Method 1

A method involving the analysis of microparticle-associated protein biomarkers in a blood sample using mass spectrometry

Methodology Applied

Scientific EffectMass spectrometry:

Data Source

PatentUS20250329468A1Systems, devices, and methods for generating machine learning models and using the machine learning models for early prediction and prevention of preeclampsia

Publication Date: 2025.10.23 NX PRENATAL INC

AI summary

Disclosed herein are methods and systems for determining risk of preeclampsia. The system can include (a) a computer comprising: (i) a processor; and (II) a memory, coupled to the processor, the memory storing a module comprising: (1) test data for a sample from a subject including values indicating a quantitative measure of one or more markers; (2) a classification rule which, based on values including the measurements, classifies the subject as being at risk of preeclampsia, wherein the classification rule is configured to have a sensitivity of at least 75%, at least 85% or at least 95%; and (3) computer executable instructions for implementing the classification rule on the test data.