N1332727 - Hannah Brown

Digital Twin-Enabled Framework for Enhanced Fetal Movement Monitoring and Prediction

Abstract

  • Reduction in fetal movements indicates fetal distress; intervention may be crucial.

  • Traditional monitoring relies on maternal perception, which is prone to errors; no accurate home counting methods exist.

  • Review discusses various technologies for fetal movement monitoring in clinical and research settings.

  • Clinical devices: accurate but too complex for home use.

  • Wearable technology (accelerometers) offers continuous monitoring but is susceptible to noise and maternal movements.

  • Need for widely available, continuous, precise technology for Fetal Movement Counting (FMC).

  • Project aims to establish a digital twin model capable of replicating fetal movements for accurate long-term FMC.

    • Model development includes:

      • Generating typical fetal movements.

      • Modeling a finite element uterine environment.

      • Calculating strain on the maternal abdomen.

      • Determining Bragg's wavelength concerning deformation.

    • Comparison with optical fibre sensor experimental data.

  • Addresses a gap in fetal health monitoring with accurate, continuous measurements outside clinical settings.

  • Proposed method results in a scaled fetal model generating a kick in the range of 16.97 to 53.15 N; Bragg's wavelength at 1543.4 nm.

  • Application of medical imaging for patient-specific geometries, incorporating bidirectional data transmission and machine learning for high-risk pregnancies, potentially leading to a digital twin for FMC.

Acknowledgements

  • Special thanks to academic supervisor Dr. Qimei Zhang for guidance, support, and mentorship throughout the project timeline.

  • Gratitude for feedback, encouragement, and prompt answers to project-related queries.

Table of Contents

  • 1. Introduction

    • 1.1. Background

    • 1.2. Aims

    • 1.3. Objectives

    • 1.4. Relevance

  • 2. Literature Review

    • 2.1. Fetal Movement Counting

    • 2.2. Use of Technology for Fetal Monitoring

    • 2.3. Digital Twins

    • 2.4. Finite Element Analysis (FEA) Fetal Modeling

    • 2.5. Musculoskeletal Modeling

    • 2.6. Knowledge Gap

    • 2.7. Simulation Software

  • 3. Methodology

    • 3.1. Ethics

    • 3.2. Methodology

    • 3.3. Fetal Kick

    • 3.4. Force Generation

    • 3.5. Leg Geometry

    • 3.6. Uterine Modeling

    • 3.7. Finite Element Analysis

    • 3.8. Optical Fibre Sensors

  • 4. Results

    • 4.1. Fetal Kick

    • 4.2. Kick Force

    • 4.3. Load Testing

    • 4.4. Wavelength

  • 5. Discussion

    • 5.1. Literature Review

    • 5.2. Methodology

    • 5.3. Results

    • 5.4. Limitations and Future Work

  • 6. Conclusions

  • 7. Reference List

  • 8. Appendix

    • 8.1. MSc Supervision Forms

    • 8.2. Scaling

    • 8.3. FEA

1. Introduction

1.1. Background

  • Pregnancy complications can lead to miscarriage, stillbirth, or low birth weight.

  • Worldwide stillbirth rate: 14.3 per 1000 births (UNICEF, 2023).

  • Many fetal deaths preventable with timely intervention.

  • Fetal Movement Counting (FMC) is a method for assessing fetal health; a decrease suggests potential fetal distress and requires intervention (Mangesi et al., 2015).

  • Study (2000): 54.7 ext{%} of intrauterine fetal deaths had recorded reduced movements prior (Efkarpidis et al., 2004).

  • NHS (2020) recommends reporting reduced fetal movements (RFM) immediately for electronic fetal monitoring, potential induced labor, or further observation (Weller et al., 2019).

  • Financial impact of RFM attendances: estimated £35,627 per 1000 patients (Camacho et al., 2022).

  • Current manual FMC by mothers: subjective and inaccurate; need for innovative technologies to improve accuracy of FMC outside clinical settings.

  • Finite Element Analysis (FEA) models the uterus and fetus, providing new biomarkers for fetal health.

1.2. Aims

  • Address gaps in maternal care regarding FMC due to inconsistent advice, insufficient home-available monitoring technologies, and lack of continuous monitoring.

  • Prior works explored sensors for detecting fetal movements; few utilized finite element analysis to model fetal kicks and their effects on maternal abdomen.

  • Aim of the project: Use computational techniques to analyze fetal kick effects on maternal abdomen deformation and strain, comparing results with novel optical fiber sensor data.

  • Lay groundwork for a digital twin (DT) model replicating fetal movement as a health indicator.

1.3. Objectives

  1. Generate fetal movement patterns.

  2. Obtain resulting forces from these movements.

  3. Analyze strain and deformation on the maternal abdomen from fetal kicks.

  4. Convert measured forces into wavelengths for comparison with optical fiber sensor data.

1.4. Relevance

  • NHS England report (2023): Increased awareness for RFM crucial for lowering death rates; need for improved fetal monitoring emphasized.

  • Recognized disparities in death rates between deprived and less deprived regions; continuous monitoring could bridge this gap (Sands, 2024).

2. Literature Review

2.1. Fetal Movement Counting

  • FMC through maternal perception is the oldest method for assessing fetal well-being (AlAmri and Smith, 2022).

  • First introduced in the 1960s, this method requires mothers to track fetal movements within specific intervals; decreases may indicate fetal issues (Mangesi et al., 2015).

  • Studies indicate perceptions of reduced movements lead to electronic fetal monitoring, increasing hospital visits (McCarthy et al., 2016).

  • Awareness campaigns in 2018 aimed to lower stillbirth rates demonstrated no effectiveness in reducing incidents (Camacho et al., 2022).

  • Despite lack of evidence supporting FMC efficacy, it’s globally recommended as a qualitative method to understand fetal well-being (NHS, 2020).

  • Known inaccuracies arise from concentration lapses of mothers, generating unnecessary anxiety (Mangesi et al., 2015; AlAmri and Smith, 2022).

  • Need for technology arises for effective and continuous monitoring across populations.

2.2. Use of Technology for Fetal Monitoring

  • FMC via maternal perception provides no equipment needs; however, subject to inaccuracy.

  • Factors influencing maternal perception include individual physiology, fetal position, and weight (Liu et al., 2022).

  • Mobile applications developed help mothers with counting (Huang et al., 2024), though limited impact on users' motivation to adhere to recommendations.

  • Advancements in technology (wearable devices, machine learning) improve fetal monitoring outcomes.

  • Gold standard clinical techniques include ultrasound and MRI (Murugesan et al., 2024).

  • Current applications such as CTG measure fetal heart rate in a clinical setting but not for continuous home usage.

  • CTG accuracy can reach 90 ext{%} only in clinical supervision (Murugesan et al., 2024).

  • UK guidelines (SBLCB v3) recommend computerized cardiotocography (cCTG) for RFM, still relying on maternal perceptions (Patel et al., 2024).

  • Devices like actographs monitor fetal activity non-invasively but can be obtrusive or expensive (Murugesan et al., 2024).

  • Fetal Electrocardiogram (fECG) and Fetal Magnetocardiography (fMCG) are advanced techniques but not suitable for home monitoring due to high costs and complexity.

  • Need for accessible wearable devices; accelerometers prevalent in fetal monitoring research (Morita et al., 2019).

2.3. Digital Twins

  • With advancements in machine learning, digital twins (DTs) are being explored to model physiological changes in medicine (Scott and Oyen, 2024).

  • Current literature shows promise with DT applications in fetal heart rate variability and circulation but not specifically for FMC.

  • Challenges with imaging techniques for patient-specific geometries hinder DT development.

  • Combining ultrasound, imaging processing, and sophisticated models can yield detailed representations of fetal health (Calcaterra et al., 2023).

2.4. Finite Element Analysis Fetal Modeling

  • FEA has potential for developing new biomarkers for fetal health, assessing fetal environment, cervix, and placenta (Plitman Mayo et al., 2016; Nowlan, 2015).

  • FEA used for birthing simulations, potentially predicting outcomes using medical imaging data (Gerikhanov, 2017).

  • Models predicting uterine ruptures created by Scott et al. (2024) demonstrate patient-specific applicability (Fernandez et al., 2016).

2.5. Musculoskeletal Modeling

  • Used to simulate human movement biomechanics; combines with FEA for analyzing physiological operations loads (Hua and Shu, 2024).

2.6. Knowledge Gap

  • Issues in FMC due to maternal perception highlight inaccuracies; need for tech to monitor fetal movements efficiently highlighted.

  • Existing research focuses on fetal responses to external factors more than movements themselves, marking an area for further study.

2.7. Simulation Software

  • Identifying appropriate software through literature review: COMSOL suitable for electromagnetic modeling, Ansys preferred for biomechanics (stress