1
|
Huang D, Yang X, Ruan H, Zhuo Y, Yuan K, Ruan B, Li F. Enhancing Prediction of Myocardial Recovery After Coronary Revascularization: Integrating Radiomics from Myocardial Contrast Echocardiography with Machine Learning. Int J Gen Med 2024; 17:2539-2555. [PMID: 38841127 PMCID: PMC11151281 DOI: 10.2147/ijgm.s465023] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/20/2024] [Accepted: 05/23/2024] [Indexed: 06/07/2024] Open
Abstract
Introduction Chronic coronary artery disease (CAD) management often relies on myocardial contrast echocardiography (MCE), yet its effectiveness is limited by subjective interpretations and difficulty in distinguishing hibernating from necrotic myocardium. This study explores the integration of machine learning (ML) with radiomics to predict functional recovery in dyskinetic myocardial segments in CAD patients undergoing revascularization, aiming to overcome these limitations. Methods This prospective study enrolled 55 chronic CAD patients, dividing into training (39 patients, 205 segments) and testing sets (16 patients, 68 segments). Dysfunctional myocardial segments were identified by initial wall motion scores (WMS) of ≥2 (hypokinesis or higher). Functional recovery was defined as a decrease of ≥1 grade in WMS during follow-up echocardiography. Radiomics features were extracted from dyssynergic segments in end-systolic phase MCE images across five cardiac cycles post- "flash" impulse and processed through a five-step feature selection. Four ML classifiers were trained and compared using these features and MCE parameters, to identify the optimal model for myocardial recovery prediction. Results Functional improvement was noted in 139 out of 273 dyskinetic segments (50.9%) following revascularization. Receiver Operating Characteristic (ROC) analysis determined that myocardial blood flow (MBF) was the most precise clinical predictor of recovery, with an area under the curve (AUC) of 0.770. Approximately 1.34 million radiomics features were extracted, with nine features identified as key predictors of myocardial recovery. The random forest (RF) model, integrating MBF values and radiomics features, demonstrated superior predictive accuracy over other ML classifiers. Validation of the RF model on the testing dataset demonstrated its effectiveness, evidenced by an AUC of 0.821, along with consistent calibration and clinical utility. Conclusion The integration of ML with radiomics from MCE effectively predicts myocardial recovery in CAD. The RF model, combining radiomics and MBF values, presents a non-invasive, precise approach, significantly enhancing CAD management.
Collapse
Affiliation(s)
- Deyi Huang
- Department of Ultrasound, The People’s Hospital of Yuhuan, Yuhuan City, Zhejiang Province, People’s Republic of China
| | - Xingan Yang
- Department of Ultrasound, Taizhou Hospital of Zhejiang Province, Linhai City, Zhejiang Province, People’s Republic of China
| | - Hongbiao Ruan
- Department of Cardiology, The People’s Hospital of Yuhuan, Yuhuan City, Zhejiang Province, People’s Republic of China
| | - Yushui Zhuo
- Department of Ultrasound, The People’s Hospital of Yuhuan, Yuhuan City, Zhejiang Province, People’s Republic of China
| | - Kai Yuan
- Department of Clinical Laboratory, The People’s Hospital of Yuhuan, Yuhuan City, Zhejiang Province, People’s Republic of China
| | - Bowen Ruan
- Department of Ultrasound, The People’s Hospital of Yuhuan, Yuhuan City, Zhejiang Province, People’s Republic of China
| | - Fang Li
- Department of Ultrasound, The People’s Hospital of Yuhuan, Yuhuan City, Zhejiang Province, People’s Republic of China
| |
Collapse
|
2
|
Lee SH, Lee SR, Rhee KS, Chae JK, Kim WH. Usefulness of Myocardial Longitudinal Strain in Prediction of Heart Failure in Patients with Successfully Reperfused Anterior Wall ST-segment Elevation Myocardial Infarction. Korean Circ J 2019; 49:960-972. [PMID: 31074229 PMCID: PMC6753029 DOI: 10.4070/kcj.2018.0421] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/21/2018] [Revised: 03/04/2019] [Accepted: 03/27/2019] [Indexed: 01/25/2023] Open
Abstract
Background and Objectives Acute myocardial infarction-related heart failure (HF) is associated with poor outcome. This study was designed to investigate the usefulness of global longitudinal strain (GLS), global circumferential strain (GCS) and mean longitudinal strain of left anterior descending artery territory (LSant) measured by 2-dimensional speckle tracking echocardiography (2D STE) in prediction of acute anterior wall ST-segment elevation myocardial infarction (ant-STEMI)-related HF. Methods A total of 171 patients with ant-STEMI who underwent successful primary coronary intervention and had available 2D STE data were enrolled. Patients were divided into 3 groups: in-hospital HF, post-discharge HF, and no-HF groups. Results In-hospital and post-discharge HF developed in 39 (22.8%) and 13 (7.6%) of patients, respectively and 113 patients (69.6%) remained without HF. Multivariate analysis showed that GLS was the only factor significantly associated with the development of in-hospital HF. For post-discharge HF, LSant was the only independent predictor. Other echocardiographic or laboratory parameters did not show independent association with the development of ant-STEMI-related HF. Conclusions GLS is a powerful echocardiographic parameter related to development of in-hospital HF and LSant was significantly associated with post-discharge HF in patients with successfully reperfused ant-STEMI.
Collapse
Affiliation(s)
- Sun Hwa Lee
- Division of Cardiology, Department of Internal Medicine, Chonbuk National University Medical School, Jeonju, Korea.,Research Institute of Clinical Medicine of Chonbuk National University, Jeonju, Korea.,Biomedical Research Institute of Chonbuk National University Hospital, Jeonju, Korea.
| | - Sang Rok Lee
- Division of Cardiology, Department of Internal Medicine, Chonbuk National University Medical School, Jeonju, Korea.,Research Institute of Clinical Medicine of Chonbuk National University, Jeonju, Korea.,Biomedical Research Institute of Chonbuk National University Hospital, Jeonju, Korea
| | - Kyoung Suk Rhee
- Division of Cardiology, Department of Internal Medicine, Chonbuk National University Medical School, Jeonju, Korea.,Research Institute of Clinical Medicine of Chonbuk National University, Jeonju, Korea.,Biomedical Research Institute of Chonbuk National University Hospital, Jeonju, Korea
| | - Jei Keon Chae
- Division of Cardiology, Department of Internal Medicine, Chonbuk National University Medical School, Jeonju, Korea.,Research Institute of Clinical Medicine of Chonbuk National University, Jeonju, Korea.,Biomedical Research Institute of Chonbuk National University Hospital, Jeonju, Korea
| | - Won Ho Kim
- Division of Cardiology, Department of Internal Medicine, Chonbuk National University Medical School, Jeonju, Korea.,Research Institute of Clinical Medicine of Chonbuk National University, Jeonju, Korea.,Biomedical Research Institute of Chonbuk National University Hospital, Jeonju, Korea
| |
Collapse
|
3
|
Orloff E, Fournier P, Bouisset F, Moine T, Cournot M, Elbaz M, Carrié D, Galinier M, Lairez O, Cognet T. Myocardial multilayer strain does not provide additional value for detection of myocardial viability assessed by SPECT imaging over and beyond standard strain. Echocardiography 2018; 35:1300-1309. [PMID: 29756381 DOI: 10.1111/echo.14022] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/29/2022] Open
Abstract
BACKGROUND The aim of this study was to evaluate the value of multilayer strain analysis to the assessment of myocardial viability (MV) through the comparison of both speckle tracking echocardiography and single-photon emission computed tomography (SPECT) imaging. We also intended to determine which segmental longitudinal strain (LS) cutoff value would be optimal to discriminate viable myocardium. METHODS We included 47 patients (average age: 61 ± 11 years) referred to our cardiac imaging center for MV evaluation. All patients underwent transthoracic echocardiography with measures of LS, SPECT, and coronary angiography. RESULTS In all, 799 segments were analyzed. We correlated myocardial tracer uptake by SPECT with sub-endocardial, sub-epicardial, and mid-segmental LS values with r = .514 P < .0001, r = .501 P < .0001, and r = .520 P < .0001, respectively. The measurements of each layer strain (sub-endocardial, sub-epicardial, and mid) had the same performance to predict MV viability as defined by SPECT with areas under curve of 0.819 [0.778-0.861, P < .0001], 0.809 [0.764-0.854, P < .0001], and 0.817 [0.773-0.860, P < .0001], respectively. The receiver-operating characteristic analysis yielded a cutoff value of -6.5% for mid-segmental LS with a sensitivity of 76% and specificity of 76% to predict segmental MV as defined by SPECT. CONCLUSIONS Multilayer strain analysis does not evaluate MV with more accuracy than standard segmental LS analysis.
Collapse
Affiliation(s)
- Elisabeth Orloff
- Department of Cardiology, Rangueil University Hospital, Toulouse, France
| | - Pauline Fournier
- Department of Cardiology, Rangueil University Hospital, Toulouse, France.,Cardiac Imaging Center, Toulouse University Hospital, Toulouse, France
| | - Frédéric Bouisset
- Department of Cardiology, Rangueil University Hospital, Toulouse, France
| | - Thomas Moine
- Department of Cardiology, Rangueil University Hospital, Toulouse, France
| | - Maxime Cournot
- Department of Cardiology, La Réunion University hospital, Saint-Pierre, France.,Inserm, UMR 1188, Sainte-Clotilde, France
| | - Meyer Elbaz
- Department of Cardiology, Rangueil University Hospital, Toulouse, France.,Medical School of Rangueil, University Paul Sabatier, Toulouse, France
| | - Didier Carrié
- Department of Cardiology, Rangueil University Hospital, Toulouse, France.,Cardiac Imaging Center, Toulouse University Hospital, Toulouse, France.,Medical School of Purpan, University Paul Sabatier, Toulouse, France
| | - Michel Galinier
- Department of Cardiology, Rangueil University Hospital, Toulouse, France.,Cardiac Imaging Center, Toulouse University Hospital, Toulouse, France.,Medical School of Rangueil, University Paul Sabatier, Toulouse, France
| | - Olivier Lairez
- Department of Cardiology, Rangueil University Hospital, Toulouse, France.,Cardiac Imaging Center, Toulouse University Hospital, Toulouse, France.,Medical School of Rangueil, University Paul Sabatier, Toulouse, France.,Department of Nuclear Medicine, Rangueil University Hospital, Toulouse, France
| | - Thomas Cognet
- Department of Cardiology, Rangueil University Hospital, Toulouse, France.,Cardiac Imaging Center, Toulouse University Hospital, Toulouse, France.,Department of Nuclear Medicine, Rangueil University Hospital, Toulouse, France
| |
Collapse
|
4
|
Prediction of Left Ventricular Remodeling after a Myocardial Infarction: Role of Myocardial Deformation: A Systematic Review and Meta-Analysis. PLoS One 2016; 11:e0168349. [PMID: 28036335 PMCID: PMC5201304 DOI: 10.1371/journal.pone.0168349] [Citation(s) in RCA: 34] [Impact Index Per Article: 4.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/22/2016] [Accepted: 11/29/2016] [Indexed: 12/22/2022] Open
Abstract
Aims Left ventricular (LV) adverse or reverse remodeling after ST-segment elevation myocardial infarction (MI) is the best outcome to assess the benefit of revascularization. Speckle tracking echocardiography (STE) may accurately identify early deformation impairment, while also being predictive of LV remodeling during follow-up. This systematic analysis aimed to provide a comprehensive review of current findings on STE as a predictor of LV remodeling after MI. Methods PubMed databases were searched through December 2014 to identify studies in adults targeting the association between LV remodeling and STE. Meta-regression was performed for longitudinal analysis. Results A total of 23 prospective studies (3066 patients) were found eligible. Eleven studies reported an association between STE and adverse remodeling and twelve studies with reverse remodeling. Using peak systolic longitudinal strain, the most accurate cut-off to predict adverse remodeling and reverse remodeling ranged from -12.8% to -10.2% and from -13.7% to -9.5%, respectively. In smaller studies, assessment of circumferential strain and torsion showed additive value in predicting remodeling. Meta-regression analysis revealed that longitudinal STE was associated with adverse remodeling (pooled univariable OR = 1.27, 1.17–1.38, p<0.001; pooled multivariable OR = 1.38, 1.13–1.70, p = 0.002) while pooled ORs of longitudinal STE only tended to predict reverse remodeling (pooled OR = 0.75, 0.54–1.06, p = 0.09). Conclusions This systematic review suggests that STE is associated with changes in LV volume or function regardless of underlying mechanisms and deformation direction. Meta-regression demonstrates a strong association between peak longitudinal systolic strain and adverse remodeling. Added STE predictive value over other clinical, biological and imaging variables remains to be proven.
Collapse
|
5
|
Valuckiene Z, Ovsianas J, Ablonskyte-Dudoniene R, Mizariene V, Melinyte K, Jurkevicius R. Left Ventricular Mechanics in Functional Ischemic Mitral Regurgitation in Acute Inferoposterior Myocardial Infarction. Echocardiography 2016; 33:1131-42. [PMID: 27350141 PMCID: PMC5094560 DOI: 10.1111/echo.13240] [Citation(s) in RCA: 6] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/19/2023] Open
Abstract
Ischemic mitral regurgitation (MR) is an established adverse prognostic factor after myocardial infarction (MI). Functional ischemic mitral regurgitation in acute phase of MI remains under‐investigated due to its often transient and dynamic nature. We aimed to assess left ventricular (LV) mechanics by speckle‐tracking echocardiography in acute inferoposterior MI and ischemic mitral regurgitation (MR). Methods Sixty‐nine patients with no structural cardiac valve abnormalities and first acute inferoposterior MI were prospectively enrolled into the study. Two‐dimensional transthoracic echocardiography for regional myocardial function and valve assessment was performed within 48 hours of presentation after reperfusion therapy (percutaneous coronary intervention). Based on degree of MR, patients were divided into no significant MR (NMR) group (N = 34, with no or mild (grade 0–I) MR) and ischemic MR (IMR) group (N = 35, with grade ≥2 MR). Thirty‐five age‐ and gender‐matched healthy individuals served as a normal reference group. Offline 2D speckle tracking analysis was performed with GE EchoPAC software. Results LV ejection fraction and longitudinal myocardial deformation parameters were significantly better in healthy subjects, but did not differ between both study groups. All circumferential myocardial deformation parameters were significantly worse in IMR group compared to healthy subjects and NMR group. Global, basal, and mid‐ventricular radial strain was significantly lower in IMR group compared to both—healthy subjects and NMR group. Conclusion Ischemic mitral regurgitation in acute inferoposterior MI is associated with worse radial and circumferential LV deformation parameters assessed by 2D speckle tracking echocardiography.
Collapse
Affiliation(s)
- Zivile Valuckiene
- Department of Cardiology, Lithuanian University of Health Sciences, Eiveniu 2, LT-50009, Kaunas, Lithuania
| | - Justas Ovsianas
- Department of Cardiology, Lithuanian University of Health Sciences, Eiveniu 2, LT-50009, Kaunas, Lithuania.,Klinikum Gutersloh, Reckenberger Straße 19, 33332, Gutersloh, Germany
| | - Ruta Ablonskyte-Dudoniene
- Department of Cardiology, Lithuanian University of Health Sciences, Eiveniu 2, LT-50009, Kaunas, Lithuania
| | - Vaida Mizariene
- Department of Cardiology, Lithuanian University of Health Sciences, Eiveniu 2, LT-50009, Kaunas, Lithuania
| | - Karolina Melinyte
- Department of Cardiology, Lithuanian University of Health Sciences, Eiveniu 2, LT-50009, Kaunas, Lithuania
| | - Renaldas Jurkevicius
- Department of Cardiology, Lithuanian University of Health Sciences, Eiveniu 2, LT-50009, Kaunas, Lithuania
| |
Collapse
|
6
|
Koshy SKG, George LK. Predicting left ventricular function recovery after reperfusion in ST elevation myocardial infarction: can we balance cost and accuracy? Echocardiography 2014; 32:613-4. [PMID: 25418115 DOI: 10.1111/echo.12776] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/30/2022] Open
Affiliation(s)
- Santhosh K G Koshy
- Regional One Health, Memphis, Tennessee; University of Tennessee Health Sciences Center, Memphis, Tennessee
| | | |
Collapse
|