A computational cardiopulmonary physiology simulator accurately predicts individual patient responses to changes in mechanical ventilator settings

Mistry, Sonal; Brook, Bindi S.; Saffaran, Sina; Chikhani, Marc; Hannon, D. M.; Laffey, John G.; Scott, Tim E.; Camporota, Luigi; Hardman, Jonathan G.; Bates, Declan G.

doi:10.1109/embc48229.2022.9871182

Cited by 2 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The simulator used in this study includes a comprehensive model of the pulmonary system based on mechanistic models of ventilation and gas exchange [38]. It was later extended to include cardiovascular components [39]. The simulator has already been validated using real patient data [40,41].…”

Section: A Computational Modelmentioning

confidence: 99%

“…For instance, if patients of Hosp A are encountered in a particular cluster more frequently than in the overall patient population formed of 5 hospitals, then that cluster is enriched with patients of Hosp A. Observed statistical significance values for each of conditions under consideration were corrected for multiple Modules used in the study In this study, the RBFOpt package [39] was used for fitting the VP model to real patient data in the optimization procedure. The following Python programming language [47] implementations were used in the study: scikit-learn [48] implementation of k-means clustering was used in the consensus clustering algorithm (sklearn.cluster.KMeans); scipy [49] implementations of hierarchical clustering were used in the consensus clustering algorithm (scipy.cluster.hierarchy, scipy.spatial.distance); statistical analysis was performed with scipy library (scipy.stats.hypergeom, scipy.stats.ttest_ind).…”

Section: Consensus Clustering and Enrichment Analysismentioning

confidence: 99%

See 1 more Smart Citation

Computational Simulation of Virtual Patients Reduces Dataset Bias and Improves Machine Learning-Based Detection of ARDS from Noisy Heterogeneous ICU Datasets

Sharafutdinov

Fritsch

Iravani

et al. 2024

IEEE Open J. Eng. Med. Biol.

View full text Add to dashboard Cite

Computational simulation of virtual patients reduces dataset bias and improves machine learning-based detection of ARDS from noisy heterogeneous ICU datasets. Mechanistic virtual patient modeling is used to infer model-derived parameters on individual patients, significantly reducing biases introduced by learning from heterogeneous datasets and allowing improved discovery of patient cohorts driven exclusively by medical conditions.

show abstract

Section: A Computational Modelmentioning

confidence: 99%

Section: Consensus Clustering and Enrichment Analysismentioning

confidence: 99%

Computational Simulation of Virtual Patients Reduces Dataset Bias and Improves Machine Learning-Based Detection of ARDS from Noisy Heterogeneous ICU Datasets

Sharafutdinov

Fritsch

Iravani

et al. 2024

IEEE Open J. Eng. Med. Biol.

View full text Add to dashboard Cite

show abstract

“…The simulator used in this study includes a comprehensive simulation model of the pulmonary system based on mechanistic models of ventilation and gas exchange [32]. It was later extended to include cardiovascular components [33]. The simulator has already been validated using real patient data [34,35].…”

Section: Computational Modelmentioning

confidence: 99%

Computational simulation of virtual patients reduces dataset bias and improves machine learning-based detection of ARDS from noisy heterogeneous ICU datasets

Sharafutdinov

Fritsch

Iravani

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Goal: 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. Methods: 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 to extract relevant medical information on individual 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 "filtered" through a VP model. Results: More robust cluster configurations were observed in clustering using the VP model-based filtered 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. Conclusions: Our results indicate that mechanistic VP modeling can be used as a filter to significantly reduce biases introduced by learning from heterogeneous datasets and to allow improved discovery of patient cohorts driven exclusively by medical conditions.

show abstract

A computational cardiopulmonary physiology simulator accurately predicts individual patient responses to changes in mechanical ventilator settings

Cited by 2 publications

References 10 publications

Computational Simulation of Virtual Patients Reduces Dataset Bias and Improves Machine Learning-Based Detection of ARDS from Noisy Heterogeneous ICU Datasets

Computational Simulation of Virtual Patients Reduces Dataset Bias and Improves Machine Learning-Based Detection of ARDS from Noisy Heterogeneous ICU Datasets

Computational simulation of virtual patients reduces dataset bias and improves machine learning-based detection of ARDS from noisy heterogeneous ICU datasets

Contact Info

Product

Resources

About