TY  - JOUR
AU  - Smole, Tim
AU  - Žunkovič, Bojan
AU  - Pičulin, Matej
AU  - Kokalj, Enja
AU  - Robnik-Šikonja, Marko
AU  - Kukar, Matjaž
AU  - Fotiadis, Dimitrios I.
AU  - Pezoulas, Vasileios C.
AU  - Tachos, Nikolaos S.
AU  - Barlocco, Fausto
AU  - Mazzarotto, Francesco
AU  - Popović, Dejana
AU  - Maier, Lars
AU  - Velicki, Lazar
AU  - MacGowan, Guy A.
AU  - Olivotto, Iacopo
AU  - Filipović, Nenad
AU  - Jakovljević, Đorđe G.
AU  - Bosnić, Zoran
PY  - 2021
UR  - https://farfar.pharmacy.bg.ac.rs/handle/123456789/3928
AB  - Background: Machine learning (ML) and artificial intelligence are emerging as important components of precision medicine that enhance diagnosis and risk stratification. Risk stratification tools for hypertrophic cardiomyopathy (HCM) exist, but they are based on traditional statistical methods. The aim was to develop a novel machine learning risk stratification tool for the prediction of 5-year risk in HCM. The goal was to determine if its predictive accuracy is higher than the accuracy of the state-of-the-art tools. Method: Data from a total of 2302 patients were used. The data were comprised of demographic characteristics, genetic data, clinical investigations, medications, and disease-related events. Four classification models were applied to model the risk level, and their decisions were explained using the SHAP (SHapley Additive exPlanations) method. Unwanted cardiac events were defined as sustained ventricular tachycardia occurrence (VT), heart failure (HF), ICD activation, sudden cardiac death (SCD), cardiac death, and all-cause death. Results: The proposed machine learning approach outperformed the similar existing risk-stratification models for SCD, cardiac death, and all-cause death risk-stratification: it achieved higher AUC by 17%, 9%, and 1%, respectively. The boosted trees achieved the best performing AUC of 0.82. The resulting model most accurately predicts VT, HF, and ICD with AUCs of 0.90, 0.88, and 0.87, respectively. Conclusions: The proposed risk-stratification model demonstrates high accuracy in predicting events in patients with hypertrophic cardiomyopathy. The use of a machine-learning risk stratification model may improve patient management, clinical practice, and outcomes in general.
PB  - Elsevier Ltd
T2  - Computers in Biology and Medicine
T1  - A machine learning-based risk stratification model for ventricular tachycardia and heart failure in hypertrophic cardiomyopathy
VL  - 135
DO  - 10.1016/j.compbiomed.2021.104648
ER  - 

@article{
author = "Smole, Tim and Žunkovič, Bojan and Pičulin, Matej and Kokalj, Enja and Robnik-Šikonja, Marko and Kukar, Matjaž and Fotiadis, Dimitrios I. and Pezoulas, Vasileios C. and Tachos, Nikolaos S. and Barlocco, Fausto and Mazzarotto, Francesco and Popović, Dejana and Maier, Lars and Velicki, Lazar and MacGowan, Guy A. and Olivotto, Iacopo and Filipović, Nenad and Jakovljević, Đorđe G. and Bosnić, Zoran",
year = "2021",
abstract = "Background: Machine learning (ML) and artificial intelligence are emerging as important components of precision medicine that enhance diagnosis and risk stratification. Risk stratification tools for hypertrophic cardiomyopathy (HCM) exist, but they are based on traditional statistical methods. The aim was to develop a novel machine learning risk stratification tool for the prediction of 5-year risk in HCM. The goal was to determine if its predictive accuracy is higher than the accuracy of the state-of-the-art tools. Method: Data from a total of 2302 patients were used. The data were comprised of demographic characteristics, genetic data, clinical investigations, medications, and disease-related events. Four classification models were applied to model the risk level, and their decisions were explained using the SHAP (SHapley Additive exPlanations) method. Unwanted cardiac events were defined as sustained ventricular tachycardia occurrence (VT), heart failure (HF), ICD activation, sudden cardiac death (SCD), cardiac death, and all-cause death. Results: The proposed machine learning approach outperformed the similar existing risk-stratification models for SCD, cardiac death, and all-cause death risk-stratification: it achieved higher AUC by 17%, 9%, and 1%, respectively. The boosted trees achieved the best performing AUC of 0.82. The resulting model most accurately predicts VT, HF, and ICD with AUCs of 0.90, 0.88, and 0.87, respectively. Conclusions: The proposed risk-stratification model demonstrates high accuracy in predicting events in patients with hypertrophic cardiomyopathy. The use of a machine-learning risk stratification model may improve patient management, clinical practice, and outcomes in general.",
publisher = "Elsevier Ltd",
journal = "Computers in Biology and Medicine",
title = "A machine learning-based risk stratification model for ventricular tachycardia and heart failure in hypertrophic cardiomyopathy",
volume = "135",
doi = "10.1016/j.compbiomed.2021.104648"
}

Smole, T., Žunkovič, B., Pičulin, M., Kokalj, E., Robnik-Šikonja, M., Kukar, M., Fotiadis, D. I., Pezoulas, V. C., Tachos, N. S., Barlocco, F., Mazzarotto, F., Popović, D., Maier, L., Velicki, L., MacGowan, G. A., Olivotto, I., Filipović, N., Jakovljević, Đ. G.,& Bosnić, Z.. (2021). A machine learning-based risk stratification model for ventricular tachycardia and heart failure in hypertrophic cardiomyopathy. in Computers in Biology and Medicine
Elsevier Ltd., 135.
https://doi.org/10.1016/j.compbiomed.2021.104648

Smole T, Žunkovič B, Pičulin M, Kokalj E, Robnik-Šikonja M, Kukar M, Fotiadis DI, Pezoulas VC, Tachos NS, Barlocco F, Mazzarotto F, Popović D, Maier L, Velicki L, MacGowan GA, Olivotto I, Filipović N, Jakovljević ĐG, Bosnić Z. A machine learning-based risk stratification model for ventricular tachycardia and heart failure in hypertrophic cardiomyopathy. in Computers in Biology and Medicine. 2021;135.
doi:10.1016/j.compbiomed.2021.104648 .

Smole, Tim, Žunkovič, Bojan, Pičulin, Matej, Kokalj, Enja, Robnik-Šikonja, Marko, Kukar, Matjaž, Fotiadis, Dimitrios I., Pezoulas, Vasileios C., Tachos, Nikolaos S., Barlocco, Fausto, Mazzarotto, Francesco, Popović, Dejana, Maier, Lars, Velicki, Lazar, MacGowan, Guy A., Olivotto, Iacopo, Filipović, Nenad, Jakovljević, Đorđe G., Bosnić, Zoran, "A machine learning-based risk stratification model for ventricular tachycardia and heart failure in hypertrophic cardiomyopathy" in Computers in Biology and Medicine, 135 (2021),
https://doi.org/10.1016/j.compbiomed.2021.104648 . .

TY  - JOUR
AU  - Pandhita, Bashar A.W.
AU  - Okwose, Nduka C.
AU  - Koshy, Aaron
AU  - Fernández, Óscar G.
AU  - Cruz, Noelia B.
AU  - Eggett, Christopher
AU  - Velicki, Lazar
AU  - Popović, Dejana
AU  - MacGowan, Guy A.
AU  - Jakovljević, Đorđe G.
PY  - 2021
UR  - https://farfar.pharmacy.bg.ac.rs/handle/123456789/3923
AB  - Objectives: The aim of the present study was to assess the validity and trending ability of the bioreactance method in estimating cardiac output at rest and in response to stress in advanced heart failure patients and heart transplant candidates. Design: This was a prospective single-center study. Setting: This study was conducted at the heart transplant center at the Freeman Hospital, Newcastle upon Tyne, UK. Participants: Eighteen patients with advanced chronic heart failure due to reduced left ventricular ejection fraction (19 ± 7%), and peak oxygen consumption 12.3 ± 3.9 mL/kg/min. Interventions: Participants underwent right heart catheterization using the Swan-Ganz catheter. Measurements and Main Results: Cardiac output was measured simultaneously using thermodilution and bioreactance at rest and during active straight leg raise test to volitional exertion. There was no significant difference in cardiac index values obtained by the thermodilution and bioreactance methods (2.26 ± 0.59 v 2.38 ± 0.50 L/min, p > 0.05) at rest and peak straight leg raise test (2.92 ± 0.77 v 3.01 ± 0.66 L/min, p > 0.05). In response to active leg raise test, thermodilution cardiac output increased by 22% and bioreactance by 21%. There was also a strong relationship between cardiac outputs from both methods at rest (r = 0.88, p < 0.01) and peak straight leg raise test (r = 0.92, p < 0.01). Cartesian plot analysis showed good trending ability of bioreactance compared with thermodilution (concordance rate = 93%) Conclusions: `Cardiac output measured by the bioreactance method is comparable to that from the thermodilution method. Bioreactance method may be used in clinical practice to assess hemodynamics and improve management of advanced heart failure patients undergoing heart transplant assessment.
PB  - Elsevier B.V.
T2  - Journal of Cardiothoracic and Vascular Anesthesia
T1  - Noninvasive Assessment of Cardiac Output in Advanced Heart Failure and Heart Transplant Candidates Using the Bioreactance Method
VL  - 35
IS  - 6
SP  - 1776
EP  - 1781
DO  - 10.1053/j.jvca.2020.09.109
ER  - 

@article{
author = "Pandhita, Bashar A.W. and Okwose, Nduka C. and Koshy, Aaron and Fernández, Óscar G. and Cruz, Noelia B. and Eggett, Christopher and Velicki, Lazar and Popović, Dejana and MacGowan, Guy A. and Jakovljević, Đorđe G.",
year = "2021",
abstract = "Objectives: The aim of the present study was to assess the validity and trending ability of the bioreactance method in estimating cardiac output at rest and in response to stress in advanced heart failure patients and heart transplant candidates. Design: This was a prospective single-center study. Setting: This study was conducted at the heart transplant center at the Freeman Hospital, Newcastle upon Tyne, UK. Participants: Eighteen patients with advanced chronic heart failure due to reduced left ventricular ejection fraction (19 ± 7%), and peak oxygen consumption 12.3 ± 3.9 mL/kg/min. Interventions: Participants underwent right heart catheterization using the Swan-Ganz catheter. Measurements and Main Results: Cardiac output was measured simultaneously using thermodilution and bioreactance at rest and during active straight leg raise test to volitional exertion. There was no significant difference in cardiac index values obtained by the thermodilution and bioreactance methods (2.26 ± 0.59 v 2.38 ± 0.50 L/min, p > 0.05) at rest and peak straight leg raise test (2.92 ± 0.77 v 3.01 ± 0.66 L/min, p > 0.05). In response to active leg raise test, thermodilution cardiac output increased by 22% and bioreactance by 21%. There was also a strong relationship between cardiac outputs from both methods at rest (r = 0.88, p < 0.01) and peak straight leg raise test (r = 0.92, p < 0.01). Cartesian plot analysis showed good trending ability of bioreactance compared with thermodilution (concordance rate = 93%) Conclusions: `Cardiac output measured by the bioreactance method is comparable to that from the thermodilution method. Bioreactance method may be used in clinical practice to assess hemodynamics and improve management of advanced heart failure patients undergoing heart transplant assessment.",
publisher = "Elsevier B.V.",
journal = "Journal of Cardiothoracic and Vascular Anesthesia",
title = "Noninvasive Assessment of Cardiac Output in Advanced Heart Failure and Heart Transplant Candidates Using the Bioreactance Method",
volume = "35",
number = "6",
pages = "1776-1781",
doi = "10.1053/j.jvca.2020.09.109"
}

Pandhita, B. A.W., Okwose, N. C., Koshy, A., Fernández, Ó. G., Cruz, N. B., Eggett, C., Velicki, L., Popović, D., MacGowan, G. A.,& Jakovljević, Đ. G.. (2021). Noninvasive Assessment of Cardiac Output in Advanced Heart Failure and Heart Transplant Candidates Using the Bioreactance Method. in Journal of Cardiothoracic and Vascular Anesthesia
Elsevier B.V.., 35(6), 1776-1781.
https://doi.org/10.1053/j.jvca.2020.09.109

Pandhita BA, Okwose NC, Koshy A, Fernández ÓG, Cruz NB, Eggett C, Velicki L, Popović D, MacGowan GA, Jakovljević ĐG. Noninvasive Assessment of Cardiac Output in Advanced Heart Failure and Heart Transplant Candidates Using the Bioreactance Method. in Journal of Cardiothoracic and Vascular Anesthesia. 2021;35(6):1776-1781.
doi:10.1053/j.jvca.2020.09.109 .

Pandhita, Bashar A.W., Okwose, Nduka C., Koshy, Aaron, Fernández, Óscar G., Cruz, Noelia B., Eggett, Christopher, Velicki, Lazar, Popović, Dejana, MacGowan, Guy A., Jakovljević, Đorđe G., "Noninvasive Assessment of Cardiac Output in Advanced Heart Failure and Heart Transplant Candidates Using the Bioreactance Method" in Journal of Cardiothoracic and Vascular Anesthesia, 35, no. 6 (2021):1776-1781,
https://doi.org/10.1053/j.jvca.2020.09.109 . .