De Filippo, Ovidio and Cammann, Victoria L. and Pancotti, Corrado and Di Vece, Davide and Silverio, Angelo and Schweiger, Victor and Niederseer, David and Szawan, Konrad A. and Wuerdinger, Michael and Koleva, Iva and Dusi, Veronica and Bellino, Michele and Vecchione, Carmine and Parodi, Guido and Bossone, Eduardo and Gili, Sebastiano and Neuhaus, Michael and Franke, Jennifer and Meder, Benjamin and Jaguszewski, Milosz and Noutsias, Michel and Knorr, Maike and Jansen, Thomas and Dichtl, Wolfgang and von Lewinski, Dirk and Burgdorf, Christof and Kherad, Behrouz and Tschoepe, Carsten and Sarcon, Annahita and Shinbane, Jerold and Rajan, Lawrence and Michels, Guido and Pfister, Roman and Cuneo, Alessandro and Jacobshagen, Claudius and Karakas, Mahir and Koenig, Wolfgang and Pott, Alexander and Meyer, Philippe and Roffi, Marco and Banning, Adrian and Wolfrum, Mathias and Cuculi, Florim and Kobza, Richard and Fischer, Thomas A. and Vasankari, Tuija and Airaksinen, K. E. Juhani and Napp, L. Christian and Dworakowski, Rafal and Maccarthy, Philip and Kaiser, Christoph and Osswald, Stefan and Galiuto, Leonarda and Chan, Christina and Bridgman, Paul and Beug, Daniel and Delmas, Clement and Lairez, Olivier and Gilyarova, Ekaterina and Shilova, Alexandra and Gilyarov, Mikhail and El-Battrawy, Ibrahim and Akin, Ibrahim and Polednikova, Karolina and Tousek, Petr and Winchester, David E. and Massoomi, Michael and Galuszka, Jan and Ukena, Christian and Poglajen, Gregor and Carrilho-Ferreira, Pedro and Hauck, Christian and Paolini, Carla and Bilato, Claudio and Kobayashi, Yoshio and Kato, Ken and Ishibashi, Iwao and Himi, Toshiharu and Din, Jehangir and Al-Shammari, Ali and Prasad, Abhiram and Rihal, Charanjit S. and Liu, Kan and Schulze, P. Christian and Bianco, Matteo and Joerg, Lucas and Rickli, Hans and Pestana, Goncalo and Nguyen, Thanh H. and Boehm, Michael and Maier, Lars S. and Pinto, Fausto J. and Widimsky, Petr and Felix, Stephan B. and Braun-Dullaeus, Ruediger C. and Rottbauer, Wolfgang and Hasenfuss, Gerd and Pieske, Burkert M. and Schunkert, Heribert and Budnik, Monika and Opolski, Grzegorz and Thiele, Holger and Bauersachs, Johann and Horowitz, John D. and Di Mario, Carlo and Bruno, Francesco and Kong, William and Dalakoti, Mayank and Imori, Yoichi and Muenzel, Thomas and Crea, Filippo and Luescher, Thomas F. and Bax, Jeroen J. and Ruschitzka, Frank and De Ferrari, Gaetano Maria and Fariselli, Piero and Ghadri, Jelena R. and Citro, Rodolfo and D'Ascenzo, Fabrizio and Templin, Christian (2023) Machine learning-based prediction of in-hospital death for patients with takotsubo syndrome: The InterTAK-ML model. WILEY, HOBOKEN.
Full text not available from this repository. (Request a copy)Abstract
Aims Takotsubo syndrome (TTS) is associated with a substantial rate of adverse events. We sought to design a machine learning (ML)-based model to predict the risk of in-hospital death and to perform a clustering of TTS patients to identify different risk profiles.Methods and results A ridge logistic regression-based ML model for predicting in-hospital death was developed on 3482 TTS patients from the International Takotsubo (InterTAK) Registry, randomly split in a train and an internal validation cohort (75% and 25% of the sample size, respectively) and evaluated in an external validation cohort (1037 patients). Thirty-one clinically relevant variables were included in the prediction model. Model performance represented the primary endpoint and was assessed according to area under the curve (AUC), sensitivity and specificity. As secondary endpoint, a K-medoids clustering algorithm was designed to stratify patients into phenotypic groups based on the 10 most relevant features emerging from the main model. The overall incidence of in-hospital death was 5.2%. The InterTAK-ML model showed an AUC of 0.89 (0.85-0.92), a sensitivity of 0.85 (0.78-0.95) and a specificity of 0.76 (0.74-0.79) in the internal validation cohort and an AUC of 0.82 (0.73-0.91), a sensitivity of 0.74 (0.61-0.87) and a specificity of 0.79 (0.77-0.81) in the external cohort for in-hospital death prediction. By exploiting the 10 variables showing the highest feature importance, TTS patients were clustered into six groups associated with different risks of in-hospital death (28.8% vs. 15.5% vs. 5.4% vs. 1.0.8% vs. 0.5%) which were consistent also in the external cohort.Conclusion A ML-based approach for the identification of TTS patients at risk of adverse short-term prognosis is feasible and effective. The InterTAK-ML model showed unprecedented discriminative capability for the prediction of in-hospital death.
| Item Type: | Other |
|---|---|
| Uncontrolled Keywords: | OUTCOMES; CARDIOMYOPATHY; MORTALITY; DISEASE; Takotsubo syndrome; Outcome; Mortality prediction; Machine learning; Artificial intelligence |
| Subjects: | 600 Technology > 610 Medical sciences Medicine |
| Divisions: | Medicine > Lehrstuhl für Innere Medizin II |
| Depositing User: | Dr. Gernot Deinzer |
| Date Deposited: | 07 May 2024 09:25 |
| Last Modified: | 07 May 2024 09:25 |
| URI: | https://pred.uni-regensburg.de/id/eprint/61001 |
Actions (login required)
 |
View Item |
