1
|
Nolin Lapalme A, Corbin D, Tastet O, Avram R, Hussin JG. Advancing Fairness in Cardiac Care: Strategies for Mitigating Bias in Artificial Intelligence Models within Cardiology. Can J Cardiol 2024:S0828-282X(24)00357-X. [PMID: 38735528 DOI: 10.1016/j.cjca.2024.04.026] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/01/2024] [Revised: 04/03/2024] [Accepted: 04/22/2024] [Indexed: 05/14/2024] Open
Abstract
In the dynamic field of medical artificial intelligence (AI), cardiology stands out as a key area for its technological advancements and clinical application. This review explores the complex issue of data bias, specifically addressing those encountered during the development and implementation of AI tools in cardiology. We dissect the origins and impacts of these biases, which challenge their reliability and widespread applicability in healthcare. Using a case study, we highlight the complexities involved in addressing these biases from a clinical viewpoint. The goal of this review is to equip researchers and clinicians with the practical knowledge needed to identify, understand, and mitigate these biases, advocating for the creation of AI solutions that are not just technologically sound, but also fair and effective for all patient demographics.
Collapse
Affiliation(s)
- Alexis Nolin Lapalme
- Department of Medicine, Montreal Heart Institute, Montreal, Canada; Faculté de Médecine, Université de Montréal, Montreal, Canada; Mila - Québec AI Institute, Montreal, Canada; Heartwise (heartwise.ai), Montreal Heart Institute, Montreal, Canada.
| | - Denis Corbin
- Department of Medicine, Montreal Heart Institute, Montreal, Canada
| | - Olivier Tastet
- Department of Medicine, Montreal Heart Institute, Montreal, Canada
| | - Robert Avram
- Department of Medicine, Montreal Heart Institute, Montreal, Canada; Faculté de Médecine, Université de Montréal, Montreal, Canada; Heartwise (heartwise.ai), Montreal Heart Institute, Montreal, Canada
| | - Julie G Hussin
- Department of Medicine, Montreal Heart Institute, Montreal, Canada; Faculté de Médecine, Université de Montréal, Montreal, Canada; Heartwise (heartwise.ai), Montreal Heart Institute, Montreal, Canada
| |
Collapse
|
2
|
Petek BJ, Drezner JA, Churchill TW. The International Criteria for Electrocardiogram Interpretation in Athletes: Common Pitfalls and Future Directions. Card Electrophysiol Clin 2024; 16:35-49. [PMID: 38280813 DOI: 10.1016/j.ccep.2023.09.006] [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] [Indexed: 01/29/2024]
Abstract
Preparticipation cardiovascular screening (PPCS) in young athletes is performed to detect conditions associated with sudden cardiac death. Many medical societies and sports governing bodies support the addition of a 12-lead electrocardiogram (ECG) to the history and physical to improve PPCS sensitivity. The current standard for ECG interpretation in athletes, the International Criteria, was developed to distinguish physiologic from pathologic ECG findings in athletes. Although application of the International Criteria has reduced the PPCS false-positive rate, interpretative challenges and potential areas of improvement remain. This review provides an overview of common pitfalls and future directions for ECG interpretation in athletes.
Collapse
Affiliation(s)
- Bradley J Petek
- Division of Cardiology, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA; Cardiovascular Performance Program, Massachusetts General Hospital, Yawkey Suite 5B, 55 Fruit Street, Boston, MA 02114, USA
| | - Jonathan A Drezner
- University of Washington Medical Center for Sports Cardiology, Massachusetts General Hospital, 3800 Montlake Boulevard Northeast, Box 354060, Seattle, WA 98195, USA
| | - Timothy W Churchill
- Division of Cardiology, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA; Cardiovascular Performance Program, Massachusetts General Hospital, Yawkey Suite 5B, 55 Fruit Street, Boston, MA 02114, USA.
| |
Collapse
|
3
|
Williamson MD, Grubic N, Moulson N, Johri AM. Comparative evaluation of an electrocardiogram workflow model for athletic cardiovascular screening: Primary care network versus sports cardiology interpretation. J Electrocardiol 2023; 81:36-40. [PMID: 37517199 DOI: 10.1016/j.jelectrocard.2023.07.007] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/29/2023] [Revised: 07/09/2023] [Accepted: 07/18/2023] [Indexed: 08/01/2023]
Abstract
BACKGROUND Electrocardiogram (ECG) testing in pre-participation screening (PPS) remains controversial due to its cost, resource dependency, and the potential for inaccurate interpretations. At most centres, ECGs are conducted internally by providers trained in athletic ECG interpretation. Outsourcing ECG requisitions to an athlete's primary care network (PCN) may reduce institutional demands. This study compared PCN-conducted athletic ECG interpretation to expert sports cardiology interpretation. METHODS This was a retrospective, single-centre chart-review study of all athletes who underwent cardiovascular PPS between 2017 and 2021. All athletes submitted an ECG with their screening package, which was conducted and interpreted within their PCN. All ECGs were reinterpreted by a sports cardiologist using the International Criteria (IC) for electrocardiographic interpretation in athletes. Overall, positive, and negative percent agreement were used to compare PCN-conducted ECG interpretation with IC interpretation. RESULTS A total of 740 athletes submitted a screening package with a valid ECG (mean age: 18.5 years, 39.6% female). PCN-conducted ECGs were interpreted by 181 unique physicians. Among 41 (5.5%) PCN-conducted ECGs that were initially interpreted as abnormal, only 5 (0.7%) were classified as abnormal according to the IC. All PCN-conducted ECGs reported as normal were also classified as normal according to the IC. The overall agreement between PCN-conducted and IC ECG interpretation was 95.1% (positive percent agreement: 100%, negative percent agreement: 95.1%). CONCLUSIONS Normal PCN-conducted athletic ECGs are interpreted with high agreement to the IC. Majority of PCN-conducted ECGs interpreted as abnormal are indeed normal as per the IC. These findings suggest that a PPS workflow model that outsources ECG requisitions to a PCN may be a reliable approach to PPS, all while reducing screening-related institutional costs and resource requirements.
Collapse
Affiliation(s)
| | - Nicholas Grubic
- Department of Medicine, Queen's University, Kingston, Ontario, Canada
| | - Nathaniel Moulson
- Division of Cardiology and Sports Cardiology BC, University of British Columbia, Vancouver, British Columbia, Canada
| | - Amer M Johri
- Department of Medicine, Queen's University, Kingston, Ontario, Canada.
| |
Collapse
|
4
|
Petek BJ, Drezner JA, Churchill TW. The International Criteria for Electrocardiogram Interpretation in Athletes: Common Pitfalls and Future Directions. Cardiol Clin 2023; 41:35-49. [PMID: 36368810 PMCID: PMC10292923 DOI: 10.1016/j.ccl.2022.08.003] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/02/2022]
Abstract
Preparticipation cardiovascular screening (PPCS) in young athletes is performed to detect conditions associated with sudden cardiac death. Many medical societies and sports governing bodies support the addition of a 12-lead electrocardiogram (ECG) to the history and physical to improve PPCS sensitivity. The current standard for ECG interpretation in athletes, the International Criteria, was developed to distinguish physiologic from pathologic ECG findings in athletes. Although application of the International Criteria has reduced the PPCS false-positive rate, interpretative challenges and potential areas of improvement remain. This review provides an overview of common pitfalls and future directions for ECG interpretation in athletes.
Collapse
Affiliation(s)
- Bradley J Petek
- Division of Cardiology, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA; Cardiovascular Performance Program, Massachusetts General Hospital, Yawkey Suite 5B, 55 Fruit Street, Boston, MA 02114, USA
| | - Jonathan A Drezner
- University of Washington Medical Center for Sports Cardiology, Massachusetts General Hospital, 3800 Montlake Boulevard Northeast, Box 354060, Seattle, WA 98195, USA
| | - Timothy W Churchill
- Division of Cardiology, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA; Cardiovascular Performance Program, Massachusetts General Hospital, Yawkey Suite 5B, 55 Fruit Street, Boston, MA 02114, USA.
| |
Collapse
|
5
|
Nijpels G, van der Heijden AAWA, Elders P, Beulens JWJ, de Vet HCW. The interobserver agreement of ECG abnormalities using Minnesota codes in people with type 2 diabetes. PLoS One 2021; 16:e0255466. [PMID: 34383817 PMCID: PMC8360582 DOI: 10.1371/journal.pone.0255466] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/23/2020] [Accepted: 07/09/2021] [Indexed: 12/02/2022] Open
Abstract
Objectives To assess the interobserver agreement in categories of electrocardiogram (ECG) abnormalities using the Minnesota Code criteria. Methods We used a random sample of 180 ECGs from people with type 2 diabetes. ECG abnormalities were classified and coded using the Minnesota ECG Classification. Each ECG was independently rated on several abnormalities by an experienced rater (rater 1) and by two cardiologists (raters 2 and 3) trained to apply the Minnesota codes on four Minnesota codes; 1-codes as an indication for myocardial infarction, 4 en 5-codes as an indication for ischemic abnormalities, 3-codes as an indication for left ventricle hypertrophy, 7-1-codes as an indication for ventricular conduction abnormalities, and 8-3-codes as an indication for atrial fibrillation / atrial flutter. After all pairwise tables were summed, the overall agreement, the specific positive and negative agreement were calculated with a 95% confidence interval (CI) for each abnormality. Also, Kappa’s with a 95% CI were calculated. Results The overall agreement (with 95% CI) were for myocardial infarction, ischemic abnormalities, left ventricle hypertrophy, conduction abnormalities and atrial fibrillation/atrial flutter respectively: 0.87 (0.84–0.91), 0.79 (0.74–0.84), 0.81 (0.76–0.85), 0.93 (0.90–0.95), 0.96 (0.93–0.97). Conclusion This study shows that the overall agreement of the Minnesota code is good to excellent.
Collapse
Affiliation(s)
- Giel Nijpels
- Department of General Practice and Elderly Care Medicine, Amsterdam University Medical Center, Location VU, Amsterdam, The Netherlands
- * E-mail:
| | - Amber A. W. A. van der Heijden
- Department of General Practice and Elderly Care Medicine, Amsterdam University Medical Center, Location VU, Amsterdam, The Netherlands
| | - Petra Elders
- Department of General Practice and Elderly Care Medicine, Amsterdam University Medical Center, Location VU, Amsterdam, The Netherlands
| | - Joline W. J. Beulens
- Department of Epidemiology and Data Science, Amsterdam University Medical Center, Location VU, Amsterdam, The Netherlands
| | - Henrica C. W. de Vet
- Department of Epidemiology and Data Science, Amsterdam University Medical Center, Location VU, Amsterdam, The Netherlands
| |
Collapse
|
6
|
Bird K, Chan G, Lu H, Greeff H, Allen J, Abbott D, Menon C, Lovell NH, Howard N, Chan WS, Fletcher RR, Alian A, Ward R, Elgendi M. Assessment of Hypertension Using Clinical Electrocardiogram Features: A First-Ever Review. Front Med (Lausanne) 2020; 7:583331. [PMID: 33344473 PMCID: PMC7746856 DOI: 10.3389/fmed.2020.583331] [Citation(s) in RCA: 10] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/14/2020] [Accepted: 10/16/2020] [Indexed: 12/31/2022] Open
Abstract
Hypertension affects an estimated 1.4 billion people and is a major cause of morbidity and mortality worldwide. Early diagnosis and intervention can potentially decrease cardiovascular events later in life. However, blood pressure (BP) measurements take time and require training for health care professionals. The measurements are also inconvenient for patients to access, numerous daily variables affect BP values, and only a few BP readings can be collected per session. This leads to an unmet need for an accurate, 24-h continuous, and portable BP measurement system. Electrocardiograms (ECGs) have been considered as an alternative way to measure BP and may meet this need. This review summarizes the literature published from January 1, 2010, to January 1, 2020, on the use of only ECG wave morphology to monitor BP or identify hypertension. From 35 articles analyzed (9 of those with no listed comorbidities and confounders), the P wave, QTc intervals and TpTe intervals may be promising for this purpose. Unfortunately, with the limited number of articles and the variety of participant populations, we are unable to make conclusions about the effectiveness of ECG-only BP monitoring. We provide 13 recommendations for future ECG-only BP monitoring studies and highlight the limited findings in pregnant and pediatric populations. With the advent of convenient and portable ECG signal recording in smart devices and wearables such as watches, understanding how to apply ECG-only findings to identify hypertension early is crucial to improving health outcomes worldwide.
Collapse
Affiliation(s)
- Kathleen Bird
- Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada
| | - Gabriel Chan
- Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada
| | - Huiqi Lu
- Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Oxford, United Kingdom
| | - Heloise Greeff
- Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Oxford, United Kingdom
| | - John Allen
- Research Center for Intelligent Healthcare, Coventry University, Coventry, United Kingdom
| | - Derek Abbott
- School of Electrical and Electronic Engineering, The University of Adelaide, Adelaide, SA, Australia.,Center for Biomedical Engineering, The University of Adelaide, Adelaide, SA, Australia
| | - Carlo Menon
- School of Mechatronic Systems Engineering, Simon Fraser University, Burnaby, BC, Canada
| | - Nigel H Lovell
- Graduate School of Biomedical Engineering, UNSW Sydney, Sydney, NSW, Australia
| | - Newton Howard
- Nuffield Department of Surgical Sciences, University of Oxford, Oxford, United Kingdom
| | - Wee-Shian Chan
- Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada
| | - Richard Ribon Fletcher
- D-Lab, Massachusetts Institute of Technology, Cambridge, MA, United States.,Department of Psychiatry, University of Massachusetts Medical School, Worcester, MA, United States
| | - Aymen Alian
- Yale School of Medicine, Yale University, New Haven, CT, United States
| | - Rabab Ward
- School of Electrical and Computer Engineering, University of British Columbia, Vancouver, BC, Canada
| | - Mohamed Elgendi
- Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada.,School of Mechatronic Systems Engineering, Simon Fraser University, Burnaby, BC, Canada.,Nuffield Department of Surgical Sciences, University of Oxford, Oxford, United Kingdom.,School of Electrical and Computer Engineering, University of British Columbia, Vancouver, BC, Canada.,BC Children's & Women's Hospital, Vancouver, BC, Canada
| |
Collapse
|
7
|
Electrocardiogram interpretation in college athletes: Local institution versus sports cardiology center interpretation. J Electrocardiol 2020; 62:49-56. [DOI: 10.1016/j.jelectrocard.2020.08.002] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/18/2020] [Revised: 07/31/2020] [Accepted: 08/03/2020] [Indexed: 02/04/2023]
|
8
|
Zorzi A, Vio R, Bettella N, Corrado D. Criteria for interpretation of the athlete's ECG: A critical appraisal. PACING AND CLINICAL ELECTROPHYSIOLOGY: PACE 2020; 43:882-890. [PMID: 32602144 DOI: 10.1111/pace.14001] [Citation(s) in RCA: 10] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 03/27/2020] [Revised: 05/18/2020] [Accepted: 06/28/2020] [Indexed: 01/27/2023]
Abstract
The electrocardiogram (ECG) is cheap and widely available but its use as a screening tool for early identification of athletes with a cardiac disease at risk of sudden cardiac death is controversial because of presumed low specificity. In the last decade, several efforts have been made to improve the distinction between physiological and pathological ECG findings in the athlete, leading to continuous evolution of the interpretation criteria. The most recent 2017 International criteria grouped ECG changes into three categories: normal, borderline, and abnormal. Borderline findings warrant further investigations only when two or more are present while abnormal changes should always be considered as the sign of a possible underlying disease. This review encompasses the evolution of the athlete's ECG interpretation criteria and highlights areas of uncertainty that will need to be addressed by further studies.
Collapse
Affiliation(s)
- Alessandro Zorzi
- Department of Cardiac, Thoracic and Vascular Sciences and Public Health, University of Padova, Padova, Italy
| | - Riccardo Vio
- Department of Cardiac, Thoracic and Vascular Sciences and Public Health, University of Padova, Padova, Italy
| | - Natascia Bettella
- Department of Cardiac, Thoracic and Vascular Sciences and Public Health, University of Padova, Padova, Italy
| | - Domenico Corrado
- Department of Cardiac, Thoracic and Vascular Sciences and Public Health, University of Padova, Padova, Italy
| |
Collapse
|