1
|
Tache IA, Hatfaludi CA, Puiu A, Itu LM, Popa-Fotea NM, Calmac L, Scafa-Udriste A. Assessment of the functional severity of coronary lesions from optical coherence tomography based on ensembled learning. Biomed Eng Online 2023; 22:127. [PMID: 38104144 PMCID: PMC10724936 DOI: 10.1186/s12938-023-01192-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/12/2023] [Accepted: 12/07/2023] [Indexed: 12/19/2023] Open
Abstract
BACKGROUND Atherosclerosis is one of the most frequent cardiovascular diseases. The dilemma faced by physicians is whether to treat or postpone the revascularization of lesions that fall within the intermediate range given by an invasive fractional flow reserve (FFR) measurement. The paper presents a monocentric study for lesions significance assessment that can potentially cause ischemia on the large coronary arteries. METHODS A new dataset is acquired, comprising the optical coherence tomography (OCT) images, clinical parameters, echocardiography and FFR measurements collected from 80 patients with 102 lesions, with stable multivessel coronary artery disease. Having the ground truth given by the invasive FFR measurement, the dataset is challenging because almost 40% of the lesions are in the gray zone, having an FFR value between 0.75 and 0.85. Twenty-six features are extracted from OCT images, clinical characteristics, and echocardiography and the most relevant are identified by examining the models' accuracy. An ensembled learning is performed for solving the binary classification problem of lesion significance considering the leave-one-out cross-validation approach. RESULTS Ensemble models are designed from the multi-features voting from 5 features models by prediction aggregation with a maximum accuracy of 81.37% and a maximum area under the curve score (AUC) of 0.856. CONCLUSIONS The proposed explainable supervised learning-based lesion classification is a new method that can be improved by training with a larger multicenter dataset for further designing a tool for guiding the decision making of the clinician for the cases outside the gray zone and for the other situation extra clinical information about the lesion is needed.
Collapse
Affiliation(s)
- Irina-Andra Tache
- Department of Automatic Control and Systems Engineering, University Politehnica of Bucharest, Bucharest, Romania.
- Siemens Advanta SRL, 15 Noiembrie Bvd, 500097, Brasov, Romania.
- Romanian Academy of Scientists, Bucharest, Romania.
| | - Cosmin-Andrei Hatfaludi
- Siemens Advanta SRL, 15 Noiembrie Bvd, 500097, Brasov, Romania
- Department of Automation and Information Technology, Transilvania University of Brasov, Mihai Viteazu Nr. 5, 5000174, Brasov, Romania
| | - Andrei Puiu
- Siemens Advanta SRL, 15 Noiembrie Bvd, 500097, Brasov, Romania
- Department of Automation and Information Technology, Transilvania University of Brasov, Mihai Viteazu Nr. 5, 5000174, Brasov, Romania
| | - Lucian Mihai Itu
- Siemens Advanta SRL, 15 Noiembrie Bvd, 500097, Brasov, Romania
- Department of Automation and Information Technology, Transilvania University of Brasov, Mihai Viteazu Nr. 5, 5000174, Brasov, Romania
- Romanian Academy of Scientists, Bucharest, Romania
| | - Nicoleta-Monica Popa-Fotea
- Department of Cardiology, Emergency Clinical Hospital, 8 Calea Floreasca, 014461, Bucharest, Romania
- Department Cardio-Thoracic, University of Medicine and Pharmacy "Carol Davila", 8 Eroii Sanitari, 050474, Bucharest, Romania
| | - Lucian Calmac
- Department of Cardiology, Emergency Clinical Hospital, 8 Calea Floreasca, 014461, Bucharest, Romania
- Department Cardio-Thoracic, University of Medicine and Pharmacy "Carol Davila", 8 Eroii Sanitari, 050474, Bucharest, Romania
| | - Alexandru Scafa-Udriste
- Department of Cardiology, Emergency Clinical Hospital, 8 Calea Floreasca, 014461, Bucharest, Romania
- Department Cardio-Thoracic, University of Medicine and Pharmacy "Carol Davila", 8 Eroii Sanitari, 050474, Bucharest, Romania
| |
Collapse
|
2
|
Hashemi J, Patel B, Chatzizisis YS, Kassab GS. Real time reduced order model for angiography fractional flow reserve. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2022; 216:106674. [PMID: 35134596 PMCID: PMC8920778 DOI: 10.1016/j.cmpb.2022.106674] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/21/2021] [Revised: 01/19/2022] [Accepted: 01/30/2022] [Indexed: 05/05/2023]
Abstract
BACKGROUND AND OBJECTIVES Fractional flow reserve (FFR) is the gold standard for quantification of coronary stenosis and pressure wire is the gold standard for measuring FFR. Recently, computational fluid dynamics (CFD) methods have been used to compute FFR less invasively using images obtained from coronary angiography. This approach is, however, computationally intensive and solutions to reduce computation time are clearly required. METHODS We hypothesized that FFR can be calculated instantly using a reduced order model (ROM) derived using response surface method (RSM) for simulation modeling in lieu of the computationally intensive CFD. Specifically, eleven physiological and anatomical factors known to affect FFR were selected as input variables, and Plackett-Burman analysis was performed in conjunction with CFD on model arteries to identify set of variables affecting FFR the most. Based on the Box-Behnken design, a mathematical model was developed to compute FFR using the retained set of variables. RESULTS The model fidelity was tested on a cohort of 90 patients (100 coronary arteries) with known pressure-wire FFR. FFR derived from this ROM had a strong correlation with pressure-wire FFR with sensitivity of 89.4%, specificity of 100% and area under curve of 0.947 (p < 0.05). CONCLUSIONS The ROM method can be used to reliably calculate FFR in patients with coronary stenosis and able to replace time-consuming CFD-based FFR estimation and provide instead a real-time calculation method.
Collapse
Affiliation(s)
- Javad Hashemi
- California Medical Innovation Institute, San Diego, CA, USA
| | - Bhavesh Patel
- California Medical Innovation Institute, San Diego, CA, USA
| | | | | |
Collapse
|