Research ExplorerPublicationsMachine-learning based prediction of in-hospital death for p...

Machine-learning based prediction of in-hospital death for patients with takotsubo syndrome: the InterTAK-ML model

Standard

Machine-learning based prediction of in-hospital death for patients with takotsubo syndrome: the InterTAK-ML model. / De Filippo, Ovidio; Cammann, Victoria L; Pancotti, Corrado; Di Vece, Davide; Silverio, Angelo; Schweiger, Victor; Niederseer, David; Szawan, Konrad A; Würdinger, Michael; Koleva, Iva; Dusi, Veronica; Bellino, Michele; Vecchione, Carmine; Parodi, Guido; Bossone, Eduardo; Gili, Sebastiano; Neuhaus, Michael; Franke, Jennifer; Meder, Benjamin; Jaguszewski, Miłosz; Noutsias, Michel; Knorr, Maike; Jansen, Thomas; Dichtl, Wolfgang; von Lewinski, Dirk; Burgdorf, Christof; Kherad, Behrouz; Tschöpe, Carsten; Sarcon, Annahita; Shinbane, Jerold; Rajan, Lawrence; Michels, Guido; Pfister, Roman; Cuneo, Alessandro; Jacobshagen, Claudius; Karakas, Mahir; Koenig, Wolfgang; Pott, Alexander; Meyer, Philippe; Roffi, Marco; Banning, Adrian; Wolfrum, Mathias; Cuculi, Florim; Kobza, Richard; Fischer, Thomas A; Vasankari, Tuija; Airaksinen, K E Juhani; Napp, L Christian; Dworakowski, Rafal; MacCarthy, Philip; Kaiser, Christoph; Osswald, Stefan; Galiuto, Leonarda; Chan, Christina; Bridgman, Paul; Beug, Daniel; Delmas, Clément; Lairez, Olivier; Gilyarova, Ekaterina; Shilova, Alexandra; Gilyarov, Mikhail; El-Battrawy, Ibrahim; Akin, Ibrahim; Poledniková, Karolina; Toušek, Petr; Winchester, David E; Massoomi, Michael; Galuszka, Jan; Ukena, Christian; Poglajen, Gregor; Carrilho-Ferreira, Pedro; Hauck, Christian; Paolini, Carla; Bilato, Claudio; Kobayashi, Yoshio; Kato, Ken; Ishibashi, Iwao; Himi, Toshiharu; Din, Jehangir; Al-Shammari, Ali; Prasad, Abhiram; Rihal, Charanjit S; Liu, Kan; Schulze, P Christian; Bianco, Matteo; Jörg, Lucas; Rickli, Hans; Pestana, Gonçalo; Nguyen, Thanh H; Böhm, Michael; Maier, Lars S; Pinto, Fausto J; Widimský, Petr; Felix, Stephan B; Braun-Dullaeus, Ruediger C; Rottbauer, Wolfgang; Hasenfuß, Gerd; Pieske, Burkert M; Schunkert, Heribert; Budnik, Monika; Opolski, Grzegorz; Thiele, Holger; Bauersachs, Johann; Horowitz, John D; Di Mario, Carlo; Bruno, Francesco; Kong, William; Dalakoti, Mayank; Imori, Yoichi; Münzel, Thomas; Crea, Filippo; Lüscher, Thomas F; Bax, Jeroen J; Ruschitzka, Frank; de Ferrari, Gaetano Maria; Fariselli, Piero; Ghadri, Jelena R; Citro, Rodolfo; D'Ascenzo, Fabrizio; Templin, Christian; InterTAK-ML.

In: EUR J HEART FAIL, Vol. 25, No. 12, 12.2023, p. 2299-2311.

Research output: SCORING: Contribution to journal › SCORING: Journal article › Research › peer-review

Harvard

De Filippo, O, Cammann, VL, Pancotti, C, Di Vece, D, Silverio, A, Schweiger, V, Niederseer, D, Szawan, KA, Würdinger, M, Koleva, I, Dusi, V, Bellino, M, Vecchione, C, Parodi, G, Bossone, E, Gili, S, Neuhaus, M, Franke, J, Meder, B, Jaguszewski, M, Noutsias, M, Knorr, M, Jansen, T, Dichtl, W, von Lewinski, D, Burgdorf, C, Kherad, B, Tschöpe, C, Sarcon, A, Shinbane, J, Rajan, L, Michels, G, Pfister, R, Cuneo, A, Jacobshagen, C, Karakas, M, Koenig, W, Pott, A, Meyer, P, Roffi, M, Banning, A, Wolfrum, M, Cuculi, F, Kobza, R, Fischer, TA, Vasankari, T, Airaksinen, KEJ, Napp, LC, Dworakowski, R, MacCarthy, P, Kaiser, C, Osswald, S, Galiuto, L, Chan, C, Bridgman, P, Beug, D, Delmas, C, Lairez, O, Gilyarova, E, Shilova, A, Gilyarov, M, El-Battrawy, I, Akin, I, Poledniková, K, Toušek, P, Winchester, DE, Massoomi, M, Galuszka, J, Ukena, C, Poglajen, G, Carrilho-Ferreira, P, Hauck, C, Paolini, C, Bilato, C, Kobayashi, Y, Kato, K, Ishibashi, I, Himi, T, Din, J, Al-Shammari, A, Prasad, A, Rihal, CS, Liu, K, Schulze, PC, Bianco, M, Jörg, L, Rickli, H, Pestana, G, Nguyen, TH, Böhm, M, Maier, LS, Pinto, FJ, Widimský, P, Felix, SB, Braun-Dullaeus, RC, Rottbauer, W, Hasenfuß, G, Pieske, BM, Schunkert, H, Budnik, M, Opolski, G, Thiele, H, Bauersachs, J, Horowitz, JD, Di Mario, C, Bruno, F, Kong, W, Dalakoti, M, Imori, Y, Münzel, T, Crea, F, Lüscher, TF, Bax, JJ, Ruschitzka, F, de Ferrari, GM, Fariselli, P, Ghadri, JR, Citro, R, D'Ascenzo, F, Templin, C & InterTAK-ML 2023, 'Machine-learning based prediction of in-hospital death for patients with takotsubo syndrome: the InterTAK-ML model', EUR J HEART FAIL, vol. 25, no. 12, pp. 2299-2311. https://doi.org/10.1002/ejhf.2983

APA

De Filippo, O., Cammann, V. L., Pancotti, C., Di Vece, D., Silverio, A., Schweiger, V., Niederseer, D., Szawan, K. A., Würdinger, M., Koleva, I., Dusi, V., Bellino, M., Vecchione, C., Parodi, G., Bossone, E., Gili, S., Neuhaus, M., Franke, J., Meder, B., ... InterTAK-ML (2023). Machine-learning based prediction of in-hospital death for patients with takotsubo syndrome: the InterTAK-ML model. EUR J HEART FAIL, 25(12), 2299-2311. https://doi.org/10.1002/ejhf.2983

Vancouver

De Filippo O, Cammann VL, Pancotti C, Di Vece D, Silverio A, Schweiger V et al. Machine-learning based prediction of in-hospital death for patients with takotsubo syndrome: the InterTAK-ML model. EUR J HEART FAIL. 2023 Dec;25(12):2299-2311. https://doi.org/10.1002/ejhf.2983

Bibtex

@article{bf75c2a00b124ba08430334be65ca704,
title = "Machine-learning based prediction of in-hospital death for patients with takotsubo syndrome: the InterTAK-ML model",
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.",
author = "{De Filippo}, Ovidio and Cammann, {Victoria L} and Corrado Pancotti and {Di Vece}, Davide and Angelo Silverio and Victor Schweiger and David Niederseer and Szawan, {Konrad A} and Michael W{\"u}rdinger and Iva Koleva and Veronica Dusi and Michele Bellino and Carmine Vecchione and Guido Parodi and Eduardo Bossone and Sebastiano Gili and Michael Neuhaus and Jennifer Franke and Benjamin Meder and Mi{\l}osz Jaguszewski and Michel Noutsias and Maike Knorr and Thomas Jansen and Wolfgang Dichtl and {von Lewinski}, Dirk and Christof Burgdorf and Behrouz Kherad and Carsten Tsch{\"o}pe and Annahita Sarcon and Jerold Shinbane and Lawrence Rajan and Guido Michels and Roman Pfister and Alessandro Cuneo and Claudius Jacobshagen and Mahir Karakas and Wolfgang Koenig and Alexander Pott and Philippe Meyer and Marco Roffi and Adrian Banning and Mathias Wolfrum and Florim Cuculi and Richard Kobza and Fischer, {Thomas A} and Tuija Vasankari and Airaksinen, {K E Juhani} and Napp, {L Christian} and Rafal Dworakowski and Philip MacCarthy and Christoph Kaiser and Stefan Osswald and Leonarda Galiuto and Christina Chan and Paul Bridgman and Daniel Beug and Cl{\'e}ment Delmas and Olivier Lairez and Ekaterina Gilyarova and Alexandra Shilova and Mikhail Gilyarov and Ibrahim El-Battrawy and Ibrahim Akin and Karolina Polednikov{\'a} and Petr Tou{\v s}ek and Winchester, {David E} and Michael Massoomi and Jan Galuszka and Christian Ukena and Gregor Poglajen and Pedro Carrilho-Ferreira and Christian Hauck and Carla Paolini and Claudio Bilato and Yoshio Kobayashi and Ken Kato and Iwao Ishibashi and Toshiharu Himi and Jehangir Din and Ali Al-Shammari and Abhiram Prasad and Rihal, {Charanjit S} and Kan Liu and Schulze, {P Christian} and Matteo Bianco and Lucas J{\"o}rg and Hans Rickli and Gon{\c c}alo Pestana and Nguyen, {Thanh H} and Michael B{\"o}hm and Maier, {Lars S} and Pinto, {Fausto J} and Petr Widimsk{\'y} and Felix, {Stephan B} and Braun-Dullaeus, {Ruediger C} and Wolfgang Rottbauer and Gerd Hasenfu{\ss} and Pieske, {Burkert M} and Heribert Schunkert and Monika Budnik and Grzegorz Opolski and Holger Thiele and Johann Bauersachs and Horowitz, {John D} and {Di Mario}, Carlo and Francesco Bruno and William Kong and Mayank Dalakoti and Yoichi Imori and Thomas M{\"u}nzel and Filippo Crea and L{\"u}scher, {Thomas F} and Bax, {Jeroen J} and Frank Ruschitzka and {de Ferrari}, {Gaetano Maria} and Piero Fariselli and Ghadri, {Jelena R} and Rodolfo Citro and Fabrizio D'Ascenzo and Christian Templin and InterTAK-ML",
note = "This article is protected by copyright. All rights reserved.",
year = "2023",
month = dec,
doi = "10.1002/ejhf.2983",
language = "English",
volume = "25",
pages = "2299--2311",
journal = "EUR J HEART FAIL",
issn = "1388-9842",
publisher = "Oxford University Press",
number = "12",

}

RIS

TY - JOUR

T1 - Machine-learning based prediction of in-hospital death for patients with takotsubo syndrome: the InterTAK-ML model

AU - De Filippo, Ovidio

AU - Cammann, Victoria L

AU - Pancotti, Corrado

AU - Di Vece, Davide

AU - Silverio, Angelo

AU - Schweiger, Victor

AU - Niederseer, David

AU - Szawan, Konrad A

AU - Würdinger, Michael

AU - Koleva, Iva

AU - Dusi, Veronica

AU - Bellino, Michele

AU - Vecchione, Carmine

AU - Parodi, Guido

AU - Bossone, Eduardo

AU - Gili, Sebastiano

AU - Neuhaus, Michael

AU - Franke, Jennifer

AU - Meder, Benjamin

AU - Jaguszewski, Miłosz

AU - Noutsias, Michel

AU - Knorr, Maike

AU - Jansen, Thomas

AU - Dichtl, Wolfgang

AU - von Lewinski, Dirk

AU - Burgdorf, Christof

AU - Kherad, Behrouz

AU - Tschöpe, Carsten

AU - Sarcon, Annahita

AU - Shinbane, Jerold

AU - Rajan, Lawrence

AU - Michels, Guido

AU - Pfister, Roman

AU - Cuneo, Alessandro

AU - Jacobshagen, Claudius

AU - Karakas, Mahir

AU - Koenig, Wolfgang

AU - Pott, Alexander

AU - Meyer, Philippe

AU - Roffi, Marco

AU - Banning, Adrian

AU - Wolfrum, Mathias

AU - Cuculi, Florim

AU - Kobza, Richard

AU - Fischer, Thomas A

AU - Vasankari, Tuija

AU - Airaksinen, K E Juhani

AU - Napp, L Christian

AU - Dworakowski, Rafal

AU - MacCarthy, Philip

AU - Kaiser, Christoph

AU - Osswald, Stefan

AU - Galiuto, Leonarda

AU - Chan, Christina

AU - Bridgman, Paul

AU - Beug, Daniel

AU - Delmas, Clément

AU - Lairez, Olivier

AU - Gilyarova, Ekaterina

AU - Shilova, Alexandra

AU - Gilyarov, Mikhail

AU - El-Battrawy, Ibrahim

AU - Akin, Ibrahim

AU - Poledniková, Karolina

AU - Toušek, Petr

AU - Winchester, David E

AU - Massoomi, Michael

AU - Galuszka, Jan

AU - Ukena, Christian

AU - Poglajen, Gregor

AU - Carrilho-Ferreira, Pedro

AU - Hauck, Christian

AU - Paolini, Carla

AU - Bilato, Claudio

AU - Kobayashi, Yoshio

AU - Kato, Ken

AU - Ishibashi, Iwao

AU - Himi, Toshiharu

AU - Din, Jehangir

AU - Al-Shammari, Ali

AU - Prasad, Abhiram

AU - Rihal, Charanjit S

AU - Liu, Kan

AU - Schulze, P Christian

AU - Bianco, Matteo

AU - Jörg, Lucas

AU - Rickli, Hans

AU - Pestana, Gonçalo

AU - Nguyen, Thanh H

AU - Böhm, Michael

AU - Maier, Lars S

AU - Pinto, Fausto J

AU - Widimský, Petr

AU - Felix, Stephan B

AU - Braun-Dullaeus, Ruediger C

AU - Rottbauer, Wolfgang

AU - Hasenfuß, Gerd

AU - Pieske, Burkert M

AU - Schunkert, Heribert

AU - Budnik, Monika

AU - Opolski, Grzegorz

AU - Thiele, Holger

AU - Bauersachs, Johann

AU - Horowitz, John D

AU - Di Mario, Carlo

AU - Bruno, Francesco

AU - Kong, William

AU - Dalakoti, Mayank

AU - Imori, Yoichi

AU - Münzel, Thomas

AU - Crea, Filippo

AU - Lüscher, Thomas F

AU - Bax, Jeroen J

AU - Ruschitzka, Frank

AU - de Ferrari, Gaetano Maria

AU - Fariselli, Piero

AU - Ghadri, Jelena R

AU - Citro, Rodolfo

AU - D'Ascenzo, Fabrizio

AU - Templin, Christian

AU - InterTAK-ML

N1 - This article is protected by copyright. All rights reserved.

PY - 2023/12

Y1 - 2023/12

N2 - 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.

AB - 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.

U2 - 10.1002/ejhf.2983

DO - 10.1002/ejhf.2983

M3 - SCORING: Journal article

C2 - 37522520

VL - 25

SP - 2299

EP - 2311

JO - EUR J HEART FAIL

JF - EUR J HEART FAIL

SN - 1388-9842

IS - 12

ER -