Development and validation of explainable machine learning models for risk of mortality in transcatheter aortic valve implantation: TAVI risk machine scores
Abstract: Aims
Identification of high-risk patients and individualized decision support based on objective criteria for rapid discharge after transcatheter aortic valve implantation (TAVI) are key requirements in the context of contemporary TAVI treatment. This study aimed to predict 30-day mortality following TAVI based on machine learning (ML) using data from the German Aortic Valve Registry.
Methods and results
Mortality risk was determined using a random forest ML model that was condensed in the newly developed TAVI Risk Machine (TRIM) scores, designed to represent clinically meaningful risk modelling before (TRIMpre) and in particular after (TRIMpost) TAVI. Algorithm was trained and cross-validated on data of 22 283 patients (729 died within 30 days post-TAVI) and generalisation was examined on data of 5864 patients (146 died). TRIMpost demonstrated significantly better performance than traditional scores [C-statistics value, 0.79; 95% confidence interval (CI)] [0.74; 0.83] compared to Society of Thoracic Surgeons (STS) with C-statistics value 0.69; 95%-CI [0.65; 0.74]). An abridged (aTRIMpost) score comprising 25 features (calculated using a web interface) exhibited significantly higher performance than traditional scores (C-statistics value, 0.74; 95%-CI [0.70; 0.78]). Validation on external data of 6693 patients (205 died within 30 days post-TAVI) of the Swiss TAVI Registry confirmed significantly better performance for the TRIMpost (C-statistics value 0.75, 95%-CI [0.72; 0.79]) compared to STS (C-statistics value 0.67, CI [0.63; 0.70]).
Conclusion
TRIM scores demonstrate good performance for risk estimation before and after TAVI. Together with clinical judgement, they may support standardised and objective decision-making before and after TAVI
- Standort
-
Deutsche Nationalbibliothek Frankfurt am Main
- Umfang
-
Online-Ressource
- Sprache
-
Englisch
- Anmerkungen
-
European heart journal - digital health. - 4, 3 (2023) , 225-235, ISSN: 2634-3916
- Ereignis
-
Veröffentlichung
- (wo)
-
Freiburg
- (wer)
-
Universität
- (wann)
-
2023
- Urheber
-
Leha, Andreas
Huber, Cynthia
Friede, Tim
Bauer, Timm
Beckmann, Andreas
Bekeredjian, Raffi
Bleiziffer, Sabine
Herrmann, Eva
Möllmann, Helge
Walther, Thomas
Beyersdorf, Friedhelm
Hamm, Christian
Künzi, Arnaud
Windecker, Stephan
Stortecky, Stefan
Kutschka, Ingo
Hasenfuß, Gerd
Ensminger, Stephan
Frerker, Christian Hermann Lukas
Seidler, Tim
- DOI
-
10.1093/ehjdh/ztad021
- URN
-
urn:nbn:de:bsz:25-freidok-2367480
- Rechteinformation
-
Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
- Letzte Aktualisierung
-
25.03.2025, 13:48 MEZ
Datenpartner
Deutsche Nationalbibliothek. Bei Fragen zum Objekt wenden Sie sich bitte an den Datenpartner.
Beteiligte
- Leha, Andreas
- Huber, Cynthia
- Friede, Tim
- Bauer, Timm
- Beckmann, Andreas
- Bekeredjian, Raffi
- Bleiziffer, Sabine
- Herrmann, Eva
- Möllmann, Helge
- Walther, Thomas
- Beyersdorf, Friedhelm
- Hamm, Christian
- Künzi, Arnaud
- Windecker, Stephan
- Stortecky, Stefan
- Kutschka, Ingo
- Hasenfuß, Gerd
- Ensminger, Stephan
- Frerker, Christian Hermann Lukas
- Seidler, Tim
- Universität
Entstanden
- 2023
Ähnliche Objekte (12)
Towards explainable automated machine learning
Challenges in developing and validating machine learning models for TAVI mortality risk prediction: reply
Explainable Machine Learning and its Limitations
eXplainable cooperative machine learning with NOVA
Machine Learning for Credit Risk Analytics
Explainable machine learning for default privacy setting prediction
Explainable machine learning for precise fatigue crack tip detection
Machine learning with knowledge graphs for explainable Artificial Intelligence
NOVA - a tool for eXplainable Cooperative Machine Learning
Machine Learning im Supply Chain Risk Management : Studie
Tail-risk protection: Machine Learning meets modern Econometrics
Privacy risk assessment of emerging machine learning paradigms
Towards explainable automated machine learning
Challenges in developing and validating machine learning models for TAVI mortality risk prediction: reply
Explainable Machine Learning and its Limitations
eXplainable cooperative machine learning with NOVA
Machine Learning for Credit Risk Analytics
Explainable machine learning for default privacy setting prediction
Explainable machine learning for precise fatigue crack tip detection
Machine learning with knowledge graphs for explainable Artificial Intelligence
NOVA - a tool for eXplainable Cooperative Machine Learning
Machine Learning im Supply Chain Risk Management : Studie
Tail-risk protection: Machine Learning meets modern Econometrics
Privacy risk assessment of emerging machine learning paradigms
Towards explainable automated machine learning
Challenges in developing and validating machine learning models for TAVI mortality risk prediction: reply
Explainable Machine Learning and its Limitations
eXplainable cooperative machine learning with NOVA
Machine Learning for Credit Risk Analytics
Explainable machine learning for default privacy setting prediction
Explainable machine learning for precise fatigue crack tip detection
Machine learning with knowledge graphs for explainable Artificial Intelligence
NOVA - a tool for eXplainable Cooperative Machine Learning
Machine Learning im Supply Chain Risk Management : Studie
Tail-risk protection: Machine Learning meets modern Econometrics