Machine learning and physical based modeling for cardiac hypertrophy
Background and objective: Predicting the long-term expansion and remodeling of the left ventricle in patients is challenging task but it has the potential to be clinically very useful. Methods: In our study, we present machine learning models based on random forests, gradient boosting, and neural ne...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-06-01
|
| Series: | Heliyon |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405844023039312 |
| _version_ | 1797812014890876928 |
|---|---|
| author | Bogdan Milićević Miljan Milošević Vladimir Simić Andrej Preveden Lazar Velicki Đorđe Jakovljević Zoran Bosnić Matej Pičulin Bojan Žunkovič Miloš Kojić Nenad Filipović |
| author_facet | Bogdan Milićević Miljan Milošević Vladimir Simić Andrej Preveden Lazar Velicki Đorđe Jakovljević Zoran Bosnić Matej Pičulin Bojan Žunkovič Miloš Kojić Nenad Filipović |
| author_sort | Bogdan Milićević |
| collection | DOAJ |
| description | Background and objective: Predicting the long-term expansion and remodeling of the left ventricle in patients is challenging task but it has the potential to be clinically very useful. Methods: In our study, we present machine learning models based on random forests, gradient boosting, and neural networks, used to track cardiac hypertrophy. We collected data from multiple patients, and then the model was trained using the patient's medical history and present level of cardiac health. We also demonstrate a physical-based model, using the finite element procedure to simulate the development of cardiac hypertrophy. Results: Our models were used to forecast the evolution of hypertrophy over six years. The machine learning model and finite element model provided similar results. Conclusions: The finite element model is much slower, but it's more accurate compared to the machine learning model since it's based on physical laws guiding the hypertrophy process. On the other hand, the machine learning model is fast but the results can be less trustworthy in some cases. Both of our models, enable us to monitor the development of the disease. Because of its speed machine learning model is more likely to be used in clinical practice. Further improvements to our machine learning model could be achieved by collecting data from finite element simulations, adding them to the dataset, and retraining the model. This can result in a fast and more accurate model combining the advantages of physical-based and machine learning modeling. |
| first_indexed | 2024-03-13T07:32:17Z |
| format | Article |
| id | doaj.art-3623697cb55743d699402dcdd5abfb8c |
| institution | Directory Open Access Journal |
| issn | 2405-8440 |
| language | English |
| last_indexed | 2024-03-13T07:32:17Z |
| publishDate | 2023-06-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Heliyon |
| spelling | doaj.art-3623697cb55743d699402dcdd5abfb8c2023-06-04T04:24:10ZengElsevierHeliyon2405-84402023-06-0196e16724Machine learning and physical based modeling for cardiac hypertrophyBogdan Milićević0Miljan Milošević1Vladimir Simić2Andrej Preveden3Lazar Velicki4Đorđe Jakovljević5Zoran Bosnić6Matej Pičulin7Bojan Žunkovič8Miloš Kojić9Nenad Filipović10Faculty of Engineering, University of Kragujevac, Kragujevac 34000, Serbia; Bioengineering Research and Development Center (BioIRC), Kragujevac 34000, SerbiaBioengineering Research and Development Center (BioIRC), Kragujevac 34000, Serbia; Institute for Information Technologies, University of Kragujevac, Kragujevac 34000, Serbia; Belgrade Metropolitan University, Belgrade 11000, SerbiaBioengineering Research and Development Center (BioIRC), Kragujevac 34000, Serbia; Institute for Information Technologies, University of Kragujevac, Kragujevac 34000, SerbiaFaculty of Medicine, University of Novi Sad, Serbia and Institute of Cardiovascular Diseases Vojvodina, Sremska Kamenica, SerbiaFaculty of Medicine, University of Novi Sad, Serbia and Institute of Cardiovascular Diseases Vojvodina, Sremska Kamenica, SerbiaTranslational and Clinical Research Institute, Faculty of Medical Sciences, Newcastle University, Newcastle Upon Tyne, UK; Faculty of Health and Life Sciences, Coventry University, Coventry, UKUniversity of Ljubljana, Faculty of Computer and Information Science, Večna Pot 113, Ljubljana, SloveniaUniversity of Ljubljana, Faculty of Computer and Information Science, Večna Pot 113, Ljubljana, SloveniaUniversity of Ljubljana, Faculty of Computer and Information Science, Večna Pot 113, Ljubljana, SloveniaBioengineering Research and Development Center (BioIRC), Kragujevac 34000, Serbia; Serbian Academy of Sciences and Arts, Belgrade 11000, Serbia; Houston Methodist Research Institute, Houston TX 77030, USAFaculty of Engineering, University of Kragujevac, Kragujevac 34000, Serbia; Bioengineering Research and Development Center (BioIRC), Kragujevac 34000, Serbia; Corresponding author. Faculty of Engineering, University of Kragujevac, Kragujevac 34000, Serbia.Background and objective: Predicting the long-term expansion and remodeling of the left ventricle in patients is challenging task but it has the potential to be clinically very useful. Methods: In our study, we present machine learning models based on random forests, gradient boosting, and neural networks, used to track cardiac hypertrophy. We collected data from multiple patients, and then the model was trained using the patient's medical history and present level of cardiac health. We also demonstrate a physical-based model, using the finite element procedure to simulate the development of cardiac hypertrophy. Results: Our models were used to forecast the evolution of hypertrophy over six years. The machine learning model and finite element model provided similar results. Conclusions: The finite element model is much slower, but it's more accurate compared to the machine learning model since it's based on physical laws guiding the hypertrophy process. On the other hand, the machine learning model is fast but the results can be less trustworthy in some cases. Both of our models, enable us to monitor the development of the disease. Because of its speed machine learning model is more likely to be used in clinical practice. Further improvements to our machine learning model could be achieved by collecting data from finite element simulations, adding them to the dataset, and retraining the model. This can result in a fast and more accurate model combining the advantages of physical-based and machine learning modeling.http://www.sciencedirect.com/science/article/pii/S2405844023039312Finite element analysisMachine learningLeft ventricle modeCardiac hypertrophyDisease progress tracking |
| spellingShingle | Bogdan Milićević Miljan Milošević Vladimir Simić Andrej Preveden Lazar Velicki Đorđe Jakovljević Zoran Bosnić Matej Pičulin Bojan Žunkovič Miloš Kojić Nenad Filipović Machine learning and physical based modeling for cardiac hypertrophy Heliyon Finite element analysis Machine learning Left ventricle mode Cardiac hypertrophy Disease progress tracking |
| title | Machine learning and physical based modeling for cardiac hypertrophy |
| title_full | Machine learning and physical based modeling for cardiac hypertrophy |
| title_fullStr | Machine learning and physical based modeling for cardiac hypertrophy |
| title_full_unstemmed | Machine learning and physical based modeling for cardiac hypertrophy |
| title_short | Machine learning and physical based modeling for cardiac hypertrophy |
| title_sort | machine learning and physical based modeling for cardiac hypertrophy |
| topic | Finite element analysis Machine learning Left ventricle mode Cardiac hypertrophy Disease progress tracking |
| url | http://www.sciencedirect.com/science/article/pii/S2405844023039312 |
| work_keys_str_mv | AT bogdanmilicevic machinelearningandphysicalbasedmodelingforcardiachypertrophy AT miljanmilosevic machinelearningandphysicalbasedmodelingforcardiachypertrophy AT vladimirsimic machinelearningandphysicalbasedmodelingforcardiachypertrophy AT andrejpreveden machinelearningandphysicalbasedmodelingforcardiachypertrophy AT lazarvelicki machinelearningandphysicalbasedmodelingforcardiachypertrophy AT đorđejakovljevic machinelearningandphysicalbasedmodelingforcardiachypertrophy AT zoranbosnic machinelearningandphysicalbasedmodelingforcardiachypertrophy AT matejpiculin machinelearningandphysicalbasedmodelingforcardiachypertrophy AT bojanzunkovic machinelearningandphysicalbasedmodelingforcardiachypertrophy AT miloskojic machinelearningandphysicalbasedmodelingforcardiachypertrophy AT nenadfilipovic machinelearningandphysicalbasedmodelingforcardiachypertrophy |