1
|
Hadida Barzilai D, Cohen-Shelly M, Sorin V, Zimlichman E, Massalha E, Allison TG, Klang E. Machine learning in cardiac stress test interpretation: a systematic review. EUROPEAN HEART JOURNAL. DIGITAL HEALTH 2024; 5:401-408. [PMID: 39081945 PMCID: PMC11284008 DOI: 10.1093/ehjdh/ztae027] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 12/24/2023] [Revised: 03/13/2024] [Accepted: 04/01/2024] [Indexed: 08/02/2024]
Abstract
Coronary artery disease (CAD) is a leading health challenge worldwide. Exercise stress testing is a foundational non-invasive diagnostic tool. Nonetheless, its variable accuracy prompts the exploration of more reliable methods. Recent advancements in machine learning (ML), including deep learning and natural language processing, have shown potential in refining the interpretation of stress testing data. Adhering to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we conducted a systematic review of ML applications in stress electrocardiogram (ECG) and stress echocardiography for CAD prognosis. Medical Literature Analysis and Retrieval System Online, Web of Science, and the Cochrane Library were used as databases. We analysed the ML models, outcomes, and performance metrics. Overall, seven relevant studies were identified. Machine-learning applications in stress ECGs resulted in sensitivity and specificity improvements. Some models achieved rates of above 96% in both metrics and reduced false positives by up to 21%. In stress echocardiography, ML models demonstrated an increase in diagnostic precision. Some models achieved specificity and sensitivity rates of up to 92.7 and 84.4%, respectively. Natural language processing applications enabled the categorization of stress echocardiography reports, with accuracy rates nearing 98%. Limitations include a small, retrospective study pool and the exclusion of nuclear stress testing, due to its well-documented status. This review indicates the potential of artificial intelligence applications in refining CAD stress testing assessment. Further development for real-world use is warranted.
Collapse
Affiliation(s)
- Dor Hadida Barzilai
- Sami Sagol AI Hub, ARC, Sheba Medical Center, 31 Emek Ha'ela, Ramat Gan 5262000, Israel
| | - Michal Cohen-Shelly
- Sami Sagol AI Hub, ARC, Sheba Medical Center, 31 Emek Ha'ela, Ramat Gan 5262000, Israel
- Leviev Heart Center, Sheba Medical Center, 31 Emek Ha'ela, Ramat Gan 5262000, Israel
| | - Vera Sorin
- School of Medicine, Tel Aviv University, Tel Aviv, Ramat Aviv 69978, Israel
- Department of Diagnostic Radiology, Sheba Medical Center, 31 Emek Ha'ela, Ramat Gan 5262000, Israel
| | - Eyal Zimlichman
- Sami Sagol AI Hub, ARC, Sheba Medical Center, 31 Emek Ha'ela, Ramat Gan 5262000, Israel
- School of Medicine, Tel Aviv University, Tel Aviv, Ramat Aviv 69978, Israel
- Sheba Medical Center, The Sheba Talpiot Medical Leadership Program, 31 Emek Ha'ela, Ramat Gan 5262000, Israel
| | - Eias Massalha
- Leviev Heart Center, Sheba Medical Center, 31 Emek Ha'ela, Ramat Gan 5262000, Israel
| | - Thomas G Allison
- Department of Cardiovascular Medicine, Mayo Clinic, 21 2nd St SW Suite 30, Rochester, MN 55905, USA
- Division of Pediatric Cardiology, Department of Pediatric and Adolescent Medicine, Mayo Clinic, 200 1st St SW, Rochester, MN 55905, USA
| | - Eyal Klang
- Sami Sagol AI Hub, ARC, Sheba Medical Center, 31 Emek Ha'ela, Ramat Gan 5262000, Israel
- School of Medicine, Tel Aviv University, Tel Aviv, Ramat Aviv 69978, Israel
- Department of Diagnostic Radiology, Sheba Medical Center, 31 Emek Ha'ela, Ramat Gan 5262000, Israel
- Sheba Medical Center, The Sheba Talpiot Medical Leadership Program, 31 Emek Ha'ela, Ramat Gan 5262000, Israel
- Division of Data-Driven and Digital Medicine (D3M), Icahn School of Medicine at Mount Sinai, 1 Gustave L. Levy Place, New York, NY 10029-5674, USA
| |
Collapse
|
2
|
Yu Z, Rahman A, Laforest R, Schindler TH, Gropler RJ, Wahl RL, Siegel BA, Jha AK. Need for objective task-based evaluation of deep learning-based denoising methods: A study in the context of myocardial perfusion SPECT. Med Phys 2023; 50:4122-4137. [PMID: 37010001 PMCID: PMC10524194 DOI: 10.1002/mp.16407] [Citation(s) in RCA: 8] [Impact Index Per Article: 8.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/05/2022] [Revised: 01/20/2023] [Accepted: 03/01/2023] [Indexed: 04/04/2023] Open
Abstract
BACKGROUND Artificial intelligence-based methods have generated substantial interest in nuclear medicine. An area of significant interest has been the use of deep-learning (DL)-based approaches for denoising images acquired with lower doses, shorter acquisition times, or both. Objective evaluation of these approaches is essential for clinical application. PURPOSE DL-based approaches for denoising nuclear-medicine images have typically been evaluated using fidelity-based figures of merit (FoMs) such as root mean squared error (RMSE) and structural similarity index measure (SSIM). However, these images are acquired for clinical tasks and thus should be evaluated based on their performance in these tasks. Our objectives were to: (1) investigate whether evaluation with these FoMs is consistent with objective clinical-task-based evaluation; (2) provide a theoretical analysis for determining the impact of denoising on signal-detection tasks; and (3) demonstrate the utility of virtual imaging trials (VITs) to evaluate DL-based methods. METHODS A VIT to evaluate a DL-based method for denoising myocardial perfusion SPECT (MPS) images was conducted. To conduct this evaluation study, we followed the recently published best practices for the evaluation of AI algorithms for nuclear medicine (the RELAINCE guidelines). An anthropomorphic patient population modeling clinically relevant variability was simulated. Projection data for this patient population at normal and low-dose count levels (20%, 15%, 10%, 5%) were generated using well-validated Monte Carlo-based simulations. The images were reconstructed using a 3-D ordered-subsets expectation maximization-based approach. Next, the low-dose images were denoised using a commonly used convolutional neural network-based approach. The impact of DL-based denoising was evaluated using both fidelity-based FoMs and area under the receiver operating characteristic curve (AUC), which quantified performance on the clinical task of detecting perfusion defects in MPS images as obtained using a model observer with anthropomorphic channels. We then provide a mathematical treatment to probe the impact of post-processing operations on signal-detection tasks and use this treatment to analyze the findings of this study. RESULTS Based on fidelity-based FoMs, denoising using the considered DL-based method led to significantly superior performance. However, based on ROC analysis, denoising did not improve, and in fact, often degraded detection-task performance. This discordance between fidelity-based FoMs and task-based evaluation was observed at all the low-dose levels and for different cardiac-defect types. Our theoretical analysis revealed that the major reason for this degraded performance was that the denoising method reduced the difference in the means of the reconstructed images and of the channel operator-extracted feature vectors between the defect-absent and defect-present cases. CONCLUSIONS The results show the discrepancy between the evaluation of DL-based methods with fidelity-based metrics versus the evaluation on clinical tasks. This motivates the need for objective task-based evaluation of DL-based denoising approaches. Further, this study shows how VITs provide a mechanism to conduct such evaluations computationally, in a time and resource-efficient setting, and avoid risks such as radiation dose to the patient. Finally, our theoretical treatment reveals insights into the reasons for the limited performance of the denoising approach and may be used to probe the effect of other post-processing operations on signal-detection tasks.
Collapse
Affiliation(s)
- Zitong Yu
- Department of Biomedical Engineering, Washington University in St. Louis, St. Louis, Missouri, USA
| | - Ashequr Rahman
- Department of Biomedical Engineering, Washington University in St. Louis, St. Louis, Missouri, USA
| | - Richard Laforest
- Mallinckrodt Institute of Radiology, Washington University in St. Louis, St. Louis, Missouri, USA
| | - Thomas H. Schindler
- Mallinckrodt Institute of Radiology, Washington University in St. Louis, St. Louis, Missouri, USA
| | - Robert J. Gropler
- Mallinckrodt Institute of Radiology, Washington University in St. Louis, St. Louis, Missouri, USA
| | - Richard L. Wahl
- Mallinckrodt Institute of Radiology, Washington University in St. Louis, St. Louis, Missouri, USA
| | - Barry A. Siegel
- Mallinckrodt Institute of Radiology, Washington University in St. Louis, St. Louis, Missouri, USA
| | - Abhinav K. Jha
- Department of Biomedical Engineering, Washington University in St. Louis, St. Louis, Missouri, USA
- Mallinckrodt Institute of Radiology, Washington University in St. Louis, St. Louis, Missouri, USA
| |
Collapse
|
3
|
Hage FG, Einstein AJ, Ananthasubramaniam K, Bourque JM, Case J, DePuey EG, Hendel RC, Henzlova MJ, Shah NR, Abbott BG, Al Jaroudi W, Better N, Doukky R, Duvall WL, Malhotra S, Pagnanelli R, Peix A, Reyes E, Saeed IM, Sanghani RM, Slomka PJ, Thompson RC, Veeranna V, Williams KA, Winchester DE. Quality metrics for single-photon emission computed tomography myocardial perfusion imaging: an ASNC information statement. J Nucl Cardiol 2023; 30:864-907. [PMID: 36607538 DOI: 10.1007/s12350-022-03162-7] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/17/2022] [Accepted: 11/17/2022] [Indexed: 01/07/2023]
Affiliation(s)
- Fadi G Hage
- Section of Cardiology, Birmingham VA Medical Center, Birmingham, AL, USA.
- Division of Cardiovascular Disease, Department of Medicine, University of Alabama at Birmingham, 446 GSB, 520 19Th Street South, Birmingham, AL, 35294, USA.
| | - Andrew J Einstein
- Seymour, Paul and Gloria Milstein Division of Cardiology, Department of Medicine and Department of Radiology, Columbia University Irving Medical Center and NewYork-Presbyterian Hospital, New York, NY, USA
| | | | - Jamieson M Bourque
- Department of Medicine (Cardiology), University of Virginia Health System, Charlottesville, VA, USA
- Department of Radiology and Medical Imaging, University of Virginia Health System, Charlottesville, VA, USA
| | - James Case
- Cardiovascular Imaging Technologies, Kansas City, MO, USA
| | - E Gordon DePuey
- Mount Sinai Morningside Hospital, New York, NY, USA
- Bay Ridge Medical Imaging, Brooklyn, NY, USA
| | - Robert C Hendel
- Department of Medicine, Division of Cardiology, Tulane University School of Medicine, New Orleans, LA, USA
| | | | - Nishant R Shah
- Warren Alpert Medical School, Brown University, Providence, RI, USA
| | - Brian G Abbott
- Warren Alpert Medical School, Brown University, Providence, RI, USA
| | - Wael Al Jaroudi
- Division of Cardiovascular Medicine, Medical College of Georgia, Augusta University, Augusta, GA, USA
| | - Nathan Better
- Department of Nuclear Medicine and Cardiology, Royal Melbourne Hospital and University of Melbourne, Melbourne, Australia
| | - Rami Doukky
- Division of Cardiology, Cook County Health and Hospitals System, Chicago, IL, USA
| | - W Lane Duvall
- Heart and Vascular Institute, Hartford Hospital, Hartford, CT, USA
| | - Saurabh Malhotra
- Division of Cardiology, Cook County Health and Hospitals System, Chicago, IL, USA
| | | | - Amalia Peix
- Nuclear Medicine Department, Institute of Cardiology and Cardiovascular Surgery, La Habana, Cuba
| | - Eliana Reyes
- Nuclear Medicine Department, Royal Brompton and Harefield NHS Foundation Trust, London, UK
| | - Ibrahim M Saeed
- Virginia Heart, Falls Church, VA, USA
- INOVA Heart and Vascular Institute, Falls Church, VA, USA
- University of Missouri, Kansas City, MO, USA
| | - Rupa M Sanghani
- Division of Cardiology, Department of Medicine, Rush University Medical Center, Chicago, IL, USA
| | | | - Randall C Thompson
- Saint Luke's Mid America Heart Institute, University of Missouri-Kansas City, Kansas City, MO, USA
| | - Vikas Veeranna
- Division of Cardiology, Department of Medicine, New England Heart and Vascular Institute, Manchester, NH, USA
| | - Kim A Williams
- Department of Medicine, University of Louisville Department of Medicine, Louisville, KY, USA
| | - David E Winchester
- Malcom Randall VA Medical Center, Gainesville, FL, USA
- Division of Cardiovascular Medicine, University of Florida College of Medicine, Gainesville, FL, USA
| |
Collapse
|
4
|
Case JA. Deep Learning and Artificial Intelligence: What Does the Cardiologist Really Need to Know? Circ Cardiovasc Imaging 2022; 15:e014744. [PMID: 36126127 DOI: 10.1161/circimaging.122.014744] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 12/27/2022]
Affiliation(s)
- James A Case
- Cardiovascular Imaging Technologies, University of Missouri Kansas City
| |
Collapse
|
5
|
Martinez DSL, Noseworthy PA, Akbilgic O, Herrmann J, Ruddy KJ, Hamid A, Maddula R, Singh A, Davis R, Gunturkun F, Jefferies JL, Brown SA. Artificial intelligence opportunities in cardio-oncology: Overview with spotlight on electrocardiography. AMERICAN HEART JOURNAL PLUS : CARDIOLOGY RESEARCH AND PRACTICE 2022; 15:100129. [PMID: 35721662 PMCID: PMC9202996 DOI: 10.1016/j.ahjo.2022.100129] [Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 12/30/2021] [Revised: 03/20/2022] [Accepted: 03/21/2022] [Indexed: 01/21/2023]
Abstract
Cardiovascular disease is a leading cause of death among cancer survivors, second only to cancer recurrence or development of new tumors. Cardio-oncology has therefore emerged as a relatively new specialty focused on prevention and management of cardiovascular consequences of cancer therapies. Yet challenges remain regarding precision and accuracy with predicting individuals at highest risk for cardiotoxicity. Barriers such as access to care also limit screening and early diagnosis to improve prognosis. Thus, developing innovative approaches for prediction and early detection of cardiovascular illness in this population is critical. In this review, we provide an overview of the present state of machine learning applications in cardio-oncology. We begin by outlining some factors that should be considered while utilizing machine learning algorithms. We then examine research in which machine learning has been applied to improve prediction of cardiac dysfunction in cancer survivors. We also highlight the use of artificial intelligence (AI) in conjunction with electrocardiogram (ECG) to predict cardiac malfunction and also atrial fibrillation (AF), and we discuss the potential role of wearables. Additionally, the article summarizes future prospects and critical takeaways for the application of machine learning in cardio-oncology. This study is the first in a series on artificial intelligence in cardio-oncology, and complements our manuscript on echocardiography and other forms of imaging relevant to cancer survivors cared for in cardiology clinical practice.
Collapse
Affiliation(s)
- Daniel Sierra-Lara Martinez
- Coronary Care Unit, National Institute of Cardiology/Instituto Nacional de Cardiologia, Ciudad de Mexico, Mexico
| | | | - Oguz Akbilgic
- Department of Health Informatics and Data Science, Parkinson School of Health Sciences and Public Health, Loyola University Chicago, Maywood, IL, USA
- Section of Cardiovascular Medicine, Department of Internal Medicine, Wake Forest School of Medicine, Wake Forest, NC, USA
| | - Joerg Herrmann
- Department of Cardiovascular Diseases, Mayo Clinic, Rochester, MN, USA
| | | | | | | | - Ashima Singh
- Institute of Health and Equity, Medical College of Wisconsin, Milwaukee, WI, USA
| | - Robert Davis
- Center for Biomedical Informatics, University of Tennessee Health Sciences Center, USA
| | - Fatma Gunturkun
- Center for Biomedical Informatics, University of Tennessee Health Sciences Center, USA
| | - John L. Jefferies
- Division of Cardiovascular Diseases, University of Tennessee Health Sciences Center, USA
- Department of Epidemiology, St. Jude Children's Research Hospital, USA
| | - Sherry-Ann Brown
- Department of Cardiovascular Diseases, Mayo Clinic, Rochester, MN, USA
- Cardio-Oncology Program, Division of Cardiovascular Medicine, Medical College of Wisconsin, Milwaukee, WI, USA
| |
Collapse
|
6
|
Baseline intracardiac echocardiography predicts haemodynamic changes and Doppler velocity patterns during follow-up after percutaneous pulmonary valve implantation. Cardiol Young 2022; 32:444-450. [PMID: 34140059 DOI: 10.1017/s1047951121002365] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/06/2022]
Abstract
BACKGROUND Intracardiac echocardiography Doppler-derived gradients have previously been shown to correlate with post-procedure echocardiographic evaluations when compared with invasive gradients measured during percutaneous pulmonary valve implantation, suggesting that intracardiac echocardiography could offer an accurate and predictable starting point to estimate valve function after percutaneous pulmonary valve implantation. METHODS We performed a retrospective chart review of 51 patients who underwent percutaneous pulmonary valve implantation between September 2018 and December 2019 in whom intracardiac echocardiography was performed immediately after valve implantation. We evaluated the correlation between intracardiac echocardiography gradients and post-procedural Doppler-derived gradients. Among the parameters assessed, those which demonstrated the strongest correlation were used to create a predictive model of expected echo-derived gradients after percutaneous pulmonary valve implantation. The equation was validated on the same sample data along with a subsequent cohort of 25 consecutive patients collected between January 2020 and July 2020. RESULTS All the assessed correlation models between intracardiac echocardiography evaluation and post-procedure transthoracic echocardiographic assessments were statistically significant, presenting moderate to strong correlations. The strongest relationship was found between intracardiac echocardiography mean gradients and post-procedural transthoracic echocardiographic mean gradients. Therefore, an equation was created based on the intracardiac echocardiography-derived mean gradient, to allow prediction of the post-procedural and follow-up transthoracic echocardiographic-derived mean gradients within a range of ±5 mmHg from the observed value in more than 80% of cases. CONCLUSIONS There is a strong correlation between intracardiac echocardiography and post-procedure transthoracic echocardiographic. This allowed us to derive a predictive equation that defines the expected transthoracic echocardiographic Doppler-derived gradient following the procedure and at out-patient follow-up after percutaneous pulmonary valve implantation.
Collapse
|
7
|
Pyslar N, Doukky R. Myocardial perfusion imaging and coronary calcium score: A marriage made in heaven. J Nucl Cardiol 2021; 28:2097-2099. [PMID: 31797318 DOI: 10.1007/s12350-019-01966-8] [Citation(s) in RCA: 7] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/12/2019] [Accepted: 11/12/2019] [Indexed: 12/13/2022]
Affiliation(s)
- Nataliya Pyslar
- Division of Cardiology, Cook County Health, Chicago, IL, USA
| | - Rami Doukky
- Division of Cardiology, Cook County Health, Chicago, IL, USA.
- Division of Cardiology, Rush University Medical Center, Chicago, IL, USA.
| |
Collapse
|
8
|
Manco L, Maffei N, Strolin S, Vichi S, Bottazzi L, Strigari L. Basic of machine learning and deep learning in imaging for medical physicists. Phys Med 2021; 83:194-205. [DOI: 10.1016/j.ejmp.2021.03.026] [Citation(s) in RCA: 12] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/02/2020] [Revised: 03/07/2021] [Accepted: 03/16/2021] [Indexed: 02/08/2023] Open
|