Advertisement
Mayo Clinic Proceedings Home

Introducing Artificial Intelligence into the Preventive Medicine Visit

      To the Editor:
      The artificial intelligence–enhanced electrocardiogram (AI-ECG) has been validated for the identification of multiple cardiac pathologies.
      • Attia Z.I.
      • Harmon D.M.
      • Behr E.R.
      • Friedman P.A.
      Application of artificial intelligence to the electrocardiogram.
      • Siontis K.C.
      • Noseworthy P.A.
      • Attia Z.I.
      • Friedman P.A.
      Artificial intelligence-enhanced electrocardiography in cardiovascular disease management.
      • Yao X.
      • Rushlow D.R.
      • Inselman J.W.
      • et al.
      Artificial intelligence-enabled electrocardiograms for identification of patients with low ejection fraction: a pragmatic, randomized clinical trial.
      We developed an AI-ECG algorithm that closely predicts chronologic age and demonstrated that the difference between AI-ECG age and chronologic age, or delta age (the first minus the latter), predicts long-term survival. Furthermore, AI-ECG age has been associated with cardiovascular diseases and risk factors such as hypertension and dyslipidemia.
      • Attia Z.I.
      • Friedman P.A.
      • Noseworthy P.A.
      • et al.
      Age and sex estimation using artificial intelligence from standard 12-lead ECGs.
      ,
      • Ladejobi A.O.
      • Medina-Inojosa J.R.
      • Shelly Cohen M.
      • et al.
      The 12-lead electrocardiogram as a biomarker of biological age.
      As this AI tool reflects potentially valuable information about overall fitness and cardiovascular health, we discuss an exemplary case where the AI-ECG showed a change towards youth following risk factor modification.
      A 36-year-old woman presented to an outpatient appointment in June 2021. She had no significant cardiac history. She weighed 106 kg at that visit (body mass index [BMI]: 39.2 kg/m2); 6 months prior, she had altered her diet and exercise regimen in attempt to lose weight and improve fitness. An ECG was obtained (Figure 1A) showing sinus bradycardia. At a follow-up visit 7 months later in January 2022, the patient reported 40 pounds of intentional weight loss in the past year. Her weight was now 96.4 kg (BMI: 36.15 kg/m2). An ECG showed normal sinus rhythm with no significant change compared with her prior ECG (Figure 1B). Interestingly, the patient’s AI-ECG predicted age had significantly dropped over the past 6 months (Figure 1C). Her June 2021 AI-ECG age was 40.76 years (actual age: 35.44 years), and her January 2022 AI-ECG age was 36.45 years (actual age: 36.05 years). These results were shared with the patient both as an encouragement and reinforcement that her healthy lifestyle changes resulted in measurable physiologic improvement.
      Figure thumbnail gr1
      FigureA,B, Electrocardiograms (ECGs) obtained in (A) June 2021 and (B) January 2022. C, AI-ECG physiologic age (y-axis) vs the patient’s chronologic age (x-axis). The two collected ECGs from (A) June 2021 and (B) January 2022 are represented by the labeled red circles and correlate with the ECGs in the Figure with the same letter. Exact artificial intelligence (AI)–ECG predicted age for the respective ECGs is given in parenthesis below each red circle. The patient’s chronologic aging over time is represented by the red dotted line.
      Algorithms using AI-ECG have been validated as promising screening tools for specific cardiac pathologies,
      • Attia Z.I.
      • Harmon D.M.
      • Behr E.R.
      • Friedman P.A.
      Application of artificial intelligence to the electrocardiogram.
      ,
      • Siontis K.C.
      • Noseworthy P.A.
      • Attia Z.I.
      • Friedman P.A.
      Artificial intelligence-enhanced electrocardiography in cardiovascular disease management.
      and the clinical application of AI continues to evolve as this area of research grows exponentially. It can be expected that AI tools will become a routine part of patient care. Here we offer an example of how AI-ECG results may be introduced in the physician-patient encounter both as a quantifiable marker of physical well-being and as a reinforcement of positive lifestyle changes.
      Previously, we found that patients with an AI-ECG age greater than chronologic age (a positive age gap) more frequently had pre-existing comorbidities including hypertension, coronary disease, and low ejection fraction with greater exposure to cardiovascular risk factors.
      • Attia Z.I.
      • Friedman P.A.
      • Noseworthy P.A.
      • et al.
      Age and sex estimation using artificial intelligence from standard 12-lead ECGs.
      ,
      • Ladejobi A.O.
      • Medina-Inojosa J.R.
      • Shelly Cohen M.
      • et al.
      The 12-lead electrocardiogram as a biomarker of biological age.
      Patients with an AI-ECG predicted below their chronologic age had significantly less comorbidities than their “older” AI-ECG counterparts.
      • Attia Z.I.
      • Friedman P.A.
      • Noseworthy P.A.
      • et al.
      Age and sex estimation using artificial intelligence from standard 12-lead ECGs.
      In the present case, a young woman with obesity but no other cardiovascular comorbidities exhibited a decreasing AI-ECG age over 6-months’ time correlating with improved physical fitness and weight loss stemming from alterations in diet and exercise. The AI-ECG results were discussed with the patient to demonstrate that her lifestyle changes had a measurable, physiologic impact recorded by her heart, beyond simple weight change identified by a scale. Although significant physiologic changes may be represented by variation in AI-ECG age, further prospective study is needed to validate that “fitness” interventions result in a reduction of physiologic age.
      • Attia Z.I.
      • Friedman P.A.
      • Noseworthy P.A.
      • et al.
      Age and sex estimation using artificial intelligence from standard 12-lead ECGs.
      ,
      • Ladejobi A.O.
      • Medina-Inojosa J.R.
      • Shelly Cohen M.
      • et al.
      The 12-lead electrocardiogram as a biomarker of biological age.

      Acknowledgments

      This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. EKO, a maker of digital stethoscopes with embedded electrocardiogram electrodes, and Anumana. Drs Friedman and Lopez-Jimenez and other Mayo inventors may also receive financial benefit from this agreement through Mayo Clinic policies. Dr Harmon is currently supported by an NIH resident investigator grant (StARR NIH 5R38HL150086-02 ). This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The research reported in this article adhered to the CARE case report guidelines. Patient consent was obtained before the submission of this case report.

      References

        • Attia Z.I.
        • Harmon D.M.
        • Behr E.R.
        • Friedman P.A.
        Application of artificial intelligence to the electrocardiogram.
        Eur Heart J. 2021; 42: 4717-4730
        • Siontis K.C.
        • Noseworthy P.A.
        • Attia Z.I.
        • Friedman P.A.
        Artificial intelligence-enhanced electrocardiography in cardiovascular disease management.
        Nat Rev Cardiol. 2021; 18: 465-478
        • Yao X.
        • Rushlow D.R.
        • Inselman J.W.
        • et al.
        Artificial intelligence-enabled electrocardiograms for identification of patients with low ejection fraction: a pragmatic, randomized clinical trial.
        Nat Med. 2021; 27: 815-819
        • Attia Z.I.
        • Friedman P.A.
        • Noseworthy P.A.
        • et al.
        Age and sex estimation using artificial intelligence from standard 12-lead ECGs.
        Circ Arrhythm Electrophysiol. 2019; 12e007284
        • Ladejobi A.O.
        • Medina-Inojosa J.R.
        • Shelly Cohen M.
        • et al.
        The 12-lead electrocardiogram as a biomarker of biological age.
        Eur Heart J Digital Health. 2021; 2: 379-389