Computational Simulation of Virtual Patients Reduces Dataset Bias and Improves Machine Learning-Based Detection of ARDS from Noisy Heterogeneous ICU Datasets
<italic>Goal:</italic> Machine learning (ML) technologies that leverage large-scale patient data are promising tools predicting disease evolution in individual patients. However, the limited generalizability of ML models developed on single-center datasets, and their unproven performance...
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Open Journal of Engineering in Medicine and Biology |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10040737/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832583220153024512 |
|---|---|
| author | Konstantin Sharafutdinov Sebastian Johannes Fritsch Mina Iravani Pejman Farhadi Ghalati Sina Saffaran Declan G. Bates Jonathan G. Hardman Richard Polzin Hannah Mayer Gernot Marx Johannes Bickenbach Andreas Schuppert |
| author_facet | Konstantin Sharafutdinov Sebastian Johannes Fritsch Mina Iravani Pejman Farhadi Ghalati Sina Saffaran Declan G. Bates Jonathan G. Hardman Richard Polzin Hannah Mayer Gernot Marx Johannes Bickenbach Andreas Schuppert |
| author_sort | Konstantin Sharafutdinov |
| collection | DOAJ |
| description | <italic>Goal:</italic> Machine learning (ML) technologies that leverage large-scale patient data are promising tools predicting disease evolution in individual patients. However, the limited generalizability of ML models developed on single-center datasets, and their unproven performance in real-world settings, remain significant constraints to their widespread adoption in clinical practice. One approach to tackle this issue is to base learning on large multi-center datasets. However, such heterogeneous datasets can introduce further biases driven by data origin, as data structures and patient cohorts may differ between hospitals. <italic>Methods:</italic> In this paper, we demonstrate how mechanistic virtual patient (VP) modeling can be used to capture specific features of patients’ states and dynamics, while reducing biases introduced by heterogeneous datasets. We show how VP modeling can be used for data augmentation through identification of individualized model parameters approximating disease states of patients with suspected acute respiratory distress syndrome (ARDS) from observational data of mixed origin. We compare the results of an unsupervised learning method (clustering) in two cases: where the learning is based on original patient data and on data derived in the matching procedure of the VP model to real patient data. <italic>Results:</italic> More robust cluster configurations were observed in clustering using the model-derived data. VP model-based clustering also reduced biases introduced by the inclusion of data from different hospitals and was able to discover an additional cluster with significant ARDS enrichment. <italic>Conclusions:</italic> Our results indicate that mechanistic VP modeling can be used to significantly reduce biases introduced by learning from heterogeneous datasets and to allow improved discovery of patient cohorts driven exclusively by medical conditions. |
| format | Article |
| id | doaj-art-44befcf235114feabf6f4ed4938e85a4 |
| institution | Kabale University |
| issn | 2644-1276 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of Engineering in Medicine and Biology |
| spelling | doaj-art-44befcf235114feabf6f4ed4938e85a42025-01-29T00:01:28ZengIEEEIEEE Open Journal of Engineering in Medicine and Biology2644-12762024-01-01561162010.1109/OJEMB.2023.324319010040737Computational Simulation of Virtual Patients Reduces Dataset Bias and Improves Machine Learning-Based Detection of ARDS from Noisy Heterogeneous ICU DatasetsKonstantin Sharafutdinov0https://orcid.org/0000-0002-9604-5742Sebastian Johannes Fritsch1https://orcid.org/0000-0002-8350-8584Mina Iravani2Pejman Farhadi Ghalati3Sina Saffaran4Declan G. Bates5https://orcid.org/0000-0003-1395-9846Jonathan G. Hardman6https://orcid.org/0000-0002-8959-8058Richard Polzin7https://orcid.org/0000-0001-6831-3001Hannah Mayer8Gernot Marx9Johannes Bickenbach10Andreas Schuppert11https://orcid.org/0000-0003-3783-6605Institute for Computational Biomedicine, RWTH Aachen University, Aachen, GermanySMITH Consortium of the German Medical Informatics Initiative, Leipzig, GermanyInstitute for Computational Biomedicine, RWTH Aachen University, Aachen, GermanyInstitute for Computational Biomedicine, RWTH Aachen University, Aachen, GermanySchool of Engineering, University of Warwick, Coventry, U.K.School of Engineering, University of Warwick, Coventry, U.K.School of Medicine, University of Nottingham, Nottingham, U.K.Institute for Computational Biomedicine, RWTH Aachen University, Aachen, GermanySMITH Consortium of the German Medical Informatics Initiative, Leipzig, GermanySMITH Consortium of the German Medical Informatics Initiative, Leipzig, GermanySMITH Consortium of the German Medical Informatics Initiative, Leipzig, GermanyInstitute for Computational Biomedicine, RWTH Aachen University, Aachen, Germany<italic>Goal:</italic> Machine learning (ML) technologies that leverage large-scale patient data are promising tools predicting disease evolution in individual patients. However, the limited generalizability of ML models developed on single-center datasets, and their unproven performance in real-world settings, remain significant constraints to their widespread adoption in clinical practice. One approach to tackle this issue is to base learning on large multi-center datasets. However, such heterogeneous datasets can introduce further biases driven by data origin, as data structures and patient cohorts may differ between hospitals. <italic>Methods:</italic> In this paper, we demonstrate how mechanistic virtual patient (VP) modeling can be used to capture specific features of patients’ states and dynamics, while reducing biases introduced by heterogeneous datasets. We show how VP modeling can be used for data augmentation through identification of individualized model parameters approximating disease states of patients with suspected acute respiratory distress syndrome (ARDS) from observational data of mixed origin. We compare the results of an unsupervised learning method (clustering) in two cases: where the learning is based on original patient data and on data derived in the matching procedure of the VP model to real patient data. <italic>Results:</italic> More robust cluster configurations were observed in clustering using the model-derived data. VP model-based clustering also reduced biases introduced by the inclusion of data from different hospitals and was able to discover an additional cluster with significant ARDS enrichment. <italic>Conclusions:</italic> Our results indicate that mechanistic VP modeling can be used to significantly reduce biases introduced by learning from heterogeneous datasets and to allow improved discovery of patient cohorts driven exclusively by medical conditions.https://ieeexplore.ieee.org/document/10040737/ARDScomputational simulationdataset biasmachine learningvirtual patients |
| spellingShingle | Konstantin Sharafutdinov Sebastian Johannes Fritsch Mina Iravani Pejman Farhadi Ghalati Sina Saffaran Declan G. Bates Jonathan G. Hardman Richard Polzin Hannah Mayer Gernot Marx Johannes Bickenbach Andreas Schuppert Computational Simulation of Virtual Patients Reduces Dataset Bias and Improves Machine Learning-Based Detection of ARDS from Noisy Heterogeneous ICU Datasets IEEE Open Journal of Engineering in Medicine and Biology ARDS computational simulation dataset bias machine learning virtual patients |
| title | Computational Simulation of Virtual Patients Reduces Dataset Bias and Improves Machine Learning-Based Detection of ARDS from Noisy Heterogeneous ICU Datasets |
| title_full | Computational Simulation of Virtual Patients Reduces Dataset Bias and Improves Machine Learning-Based Detection of ARDS from Noisy Heterogeneous ICU Datasets |
| title_fullStr | Computational Simulation of Virtual Patients Reduces Dataset Bias and Improves Machine Learning-Based Detection of ARDS from Noisy Heterogeneous ICU Datasets |
| title_full_unstemmed | Computational Simulation of Virtual Patients Reduces Dataset Bias and Improves Machine Learning-Based Detection of ARDS from Noisy Heterogeneous ICU Datasets |
| title_short | Computational Simulation of Virtual Patients Reduces Dataset Bias and Improves Machine Learning-Based Detection of ARDS from Noisy Heterogeneous ICU Datasets |
| title_sort | computational simulation of virtual patients reduces dataset bias and improves machine learning based detection of ards from noisy heterogeneous icu datasets |
| topic | ARDS computational simulation dataset bias machine learning virtual patients |
| url | https://ieeexplore.ieee.org/document/10040737/ |
| work_keys_str_mv | AT konstantinsharafutdinov computationalsimulationofvirtualpatientsreducesdatasetbiasandimprovesmachinelearningbaseddetectionofardsfromnoisyheterogeneousicudatasets AT sebastianjohannesfritsch computationalsimulationofvirtualpatientsreducesdatasetbiasandimprovesmachinelearningbaseddetectionofardsfromnoisyheterogeneousicudatasets AT minairavani computationalsimulationofvirtualpatientsreducesdatasetbiasandimprovesmachinelearningbaseddetectionofardsfromnoisyheterogeneousicudatasets AT pejmanfarhadighalati computationalsimulationofvirtualpatientsreducesdatasetbiasandimprovesmachinelearningbaseddetectionofardsfromnoisyheterogeneousicudatasets AT sinasaffaran computationalsimulationofvirtualpatientsreducesdatasetbiasandimprovesmachinelearningbaseddetectionofardsfromnoisyheterogeneousicudatasets AT declangbates computationalsimulationofvirtualpatientsreducesdatasetbiasandimprovesmachinelearningbaseddetectionofardsfromnoisyheterogeneousicudatasets AT jonathanghardman computationalsimulationofvirtualpatientsreducesdatasetbiasandimprovesmachinelearningbaseddetectionofardsfromnoisyheterogeneousicudatasets AT richardpolzin computationalsimulationofvirtualpatientsreducesdatasetbiasandimprovesmachinelearningbaseddetectionofardsfromnoisyheterogeneousicudatasets AT hannahmayer computationalsimulationofvirtualpatientsreducesdatasetbiasandimprovesmachinelearningbaseddetectionofardsfromnoisyheterogeneousicudatasets AT gernotmarx computationalsimulationofvirtualpatientsreducesdatasetbiasandimprovesmachinelearningbaseddetectionofardsfromnoisyheterogeneousicudatasets AT johannesbickenbach computationalsimulationofvirtualpatientsreducesdatasetbiasandimprovesmachinelearningbaseddetectionofardsfromnoisyheterogeneousicudatasets AT andreasschuppert computationalsimulationofvirtualpatientsreducesdatasetbiasandimprovesmachinelearningbaseddetectionofardsfromnoisyheterogeneousicudatasets |