Filter and export

Publications

What is a Publication?
4 Publications visible to you, out of a total of 4
Deep Learning and Explainable Artificial Intelligence for Improving Specificity and Detecting Metabolic Patterns in Newborn Screening

Abstract

Not specified

Authors: Elaine Zaunseder, Ulrike Mütze, Sven F. Garbade, Saskia Haupt, Stefan Kölker, Vincent Heuveline

Date Published: 5th Dec 2023

Publication Type: Journal

Personalised metabolic whole-body models for newborns and infants predict growth and biomarkers of inherited metabolic diseases

Abstract (Expand)

Abstract Extensive whole-body models (WBMs) accounting for organ-specific dynamics have been developed to simulate adult metabolism. However, there is currently a lack of models representing infant …ls representing infant metabolism taking into consideration its special requirements in energy balance, nutrition, and growth. Here, we present a resource of organ-resolved, sex-specific, anatomically accurate models of newborn and infant metabolism, referred to as infant-whole-body models (infant-WBMs), spanning the first 180 days of life. These infant-WBMs were parameterised to represent the distinct metabolic characteristics of newborns and infants accurately. In particular, we adjusted the changes in organ weights, the energy requirements of brain development, heart function, and thermoregulation, as well as dietary requirements and energy requirements for physical activity. Subsequently, we validated the accuracy of the infant-WBMs by showing that the predicted neonatal and infant growth was consistent with the recommended growth by the World Health Organisation. We assessed the infant-WBMs’ reliability and capabilities for personalisation by simulating 10,000 newborn models, personalised with blood concentration measurements from newborn screening and birth weight. Moreover, we demonstrate that the models can accurately predict changes over time in known blood biomarkers in inherited metabolic diseases. By this, the infant-WBM resource can provide valuable insights into infant metabolism on an organ-resolved level and enable a holistic view of the metabolic processes occurring in infants, considering the unique energy and dietary requirements as well as growth patterns specific to this population. As such, the infant-WBM resource holds promise for personalised medicine, as the infant-WBMs could be a first step to digital metabolic twins for newborn and infant metabolism for personalised systematic simulations and treatment planning.

Authors: Elaine Zaunseder, Ulrike Mütze, Jürgen G. Okun, Georg F. Hoffmann, Stefan Kölker, Vincent Heuveline, Ines Thiele

Date Published: 23rd Oct 2023

Publication Type: Journal

Machine Learning Methods Improve Specificity in Newborn Screening for Isovaleric Aciduria

Abstract (Expand)

Isovaleric aciduria (IVA) is a rare disorder of leucine metabolism and part of newborn screening (NBS) programs worldwide. However, NBS for IVA is hampered by, first, the increased birth prevalence due …prevalence due to the identification of individuals with an attenuated disease variant (so-called “mild” IVA) and, second, an increasing number of false positive screening results due to the use of pivmecillinam contained in the medication. Recently, machine learning (ML) methods have been analyzed, analogous to new biomarkers or second-tier methods, in the context of NBS. In this study, we investigated the application of machine learning classification methods to improve IVA classification using an NBS data set containing 2,106,090 newborns screened in Heidelberg, Germany. Therefore, we propose to combine two methods, linear discriminant analysis, and ridge logistic regression as an additional step, a digital-tier, to traditional NBS. Our results show that this reduces the false positive rate by 69.9% from 103 to 31 while maintaining 100% sensitivity in cross-validation. The ML methods were able to classify mild and classic IVA from normal newborns solely based on the NBS data and revealed that besides isovalerylcarnitine (C5), the metabolite concentration of tryptophan (Trp) is important for improved classification. Overall, applying ML methods to improve the specificity of IVA could have a major impact on newborns, as it could reduce the newborns’ and families’ burden of false positives or over-treatment.

Authors: Elaine Zaunseder, Ulrike Mütze, Sven F. Garbade, Saskia Haupt, Patrik Feyh, Georg F. Hoffmann, Vincent Heuveline, Stefan Kölker

Date Published: 1st Feb 2023

Publication Type: Journal

Opportunities and challenges in machine learning‐based newborn screening—A systematic literature review

Abstract

Not specified

Authors: Elaine Zaunseder, Saskia Haupt, Ulrike Mütze, Sven F. Garbade, Stefan Kölker, Vincent Heuveline

Date Published: 1st May 2022

Publication Type: Journal

Powered by
 (v.1.14.2)
About HITS Publication SEEK | About HITS-Publication-SEEK | Funding and Programmes | Credits
Copyright © 2008 - 2023 The University of Manchester and HITS gGmbH