Universität zu Köln

Kramer, Tilmann ORCID: 0000-0003-0265-7607, Weis, Henning ORCID: 0000-0001-8638-3281, Kramer, Mira ORCID: 0009-0001-3464-0802, Baldus, Stephan ORCID: 0000-0001-8259-1737, Rosenkranz, Stephan ORCID: 0000-0001-6237-1470 and Spinler, Stefan (2026). Machine learning for prediction of key haemodynamic parameters in pulmonary arterial hypertension. European Heart Journal - Digital Health, 7 (5). pp. 1-6. Oxford University Press. ISSN 2634-3916

PDF ztaf074.pdf Bereitstellung unter der CC-Lizenz: Creative Commons Attribution. Download (2MB)

Abstract

Aims: Machine learning (ML) is increasingly recognized for its ability to identify and structure variables for predictive tasks. Pulmonary arterial hypertension (PAH) is a progressive disease characterized by elevated mean pulmonary arterial pressure (mPAP) and pulmonary vascular resistance (PVR) with normal pulmonary arterial wedge pressure (PAWP), as assessed by right heart catheterization (RHC). Despite increased awareness, delays between onset of non-specific symptoms and diagnosis continue to hinder early initiation of targeted therapies, leading to poorer outcomes. To develop and evaluate ML models for predicting key haemodynamic parameters in PAH, based on routinely available non-invasive data collected within 8 weeks prior to RHC, as a proof of concept. Methods and results: We analysed data from 181 patients with invasively confirmed PAH, incorporating 56 variables, including demographics, echocardiography, blood gas analyses, 6-min walk distances, laboratory tests, and WHO functional class. An 80/20 train-test split and fivefold cross-validation were applied across multiple ML models, including least absolute shrinkage and selection operator (lasso) regression, ridge regression, k-nearest neighbours, decision trees, random forest, and gradient boosting machine. Lasso achieved best performance for predicting mPAP (r = 0.80, R² = 0.64, RMSE = 8.49). For PVR, ridge performed best (r = 0.71, R² = 0.51, RMSE = 3.60). Random forest and gradient boosting machines achieved modest but consistent performance for cardiac index (r = 0.38 and 0.37), while PAWP prediction remained limited across all models. Conclusion: Machine learning models can estimate mPAP and PVR from routine clinical data obtained prior to RHC in patients with confirmed PAH. External validation is required to confirm generalizability and clinical applicability.

Item Type:	Article
Creators:	Creators Email ORCID ORCID Put Code Kramer, Tilmann UNSPECIFIED https://orcid.org/0000-0003-0265-7607 UNSPECIFIED Weis, Henning UNSPECIFIED https://orcid.org/0000-0001-8638-3281 UNSPECIFIED Kramer, Mira UNSPECIFIED https://orcid.org/0009-0001-3464-0802 UNSPECIFIED Baldus, Stephan UNSPECIFIED https://orcid.org/0000-0001-8259-1737 UNSPECIFIED Rosenkranz, Stephan UNSPECIFIED https://orcid.org/0000-0001-6237-1470 UNSPECIFIED Spinler, Stefan UNSPECIFIED UNSPECIFIED UNSPECIFIED
URN:	urn:nbn:de:hbz:38-811272
Identification Number:	10.1093/ehjdh/ztaf074
Journal or Publication Title:	European Heart Journal - Digital Health
Volume:	7
Number:	5
Page Range:	pp. 1-6
Number of Pages:	6
Date:	18 June 2026
Publisher:	Oxford University Press
ISSN:	2634-3916
Language:	English
Faculty:	Faculty of Medicine
Divisions:	Faculty of Medicine > Anästhesiologie und Operative Intensivmedizin > Klinik für Anästhesiologie und Operative Intensivmedizin Faculty of Medicine > Chirurgie > Klinik und Poliklinik für Herzchirurgie Faculty of Medicine > Innere Medizin > Klinik III für Innere Medizin - Kardiologie, Pneumologie, Angiologie und internistische Intensivmedizin Faculty of Medicine > Nuklearmedizin > Klinik und Poliklinik für Nuklearmedizin
Subjects:	Medical sciences Medicine
Uncontrolled Keywords:	Keywords Language Pulmonary arterial hypertension ; Machine learning ; Haemodynamic assessment ; Non-invasive prediction ; Right heart catheterization ; Cardiopulmonary haemodynamics UNSPECIFIED
['eprint_fieldname_oa_funders' not defined]:	Publikationsfonds UzK
Refereed:	Yes
URI:	http://kups.ub.uni-koeln.de/id/eprint/81127