New research from Mayo Clinic suggests that artificial intelligence (AI) could improve the diagnosis of peripartum cardiomyopathy, a potentially life-threatening and treatable condition that weakens the heart muscle of women during pregnancy or in the months after giving birth. Researchers used an AI-enabled digital stethoscope that captures electrocardiogram (ECG) data and heart sounds to identify twice as many cases of peripartum cardiomyopathy as compared to regular care, according to a news release from the American Heart Association.
Identifying a weak heart pump caused by pregnancy is important because the symptoms, such as shortness of breath when lying down, swelling of hands and feet, weight gain, and rapid heartbeat, can be confused with normal symptoms of pregnancy.
Dr Demilade Adedinsewo, a cardiologist at Mayo Clinic, shared research insights during a late-breaking science presentation at the American Heart Association’s Scientific Sessions 2023.
Women in Nigeria have the highest reported incidence of peripartum cardiomyopathy. The randomised pragmatic clinical trial enrolled 1195 women receiving pregnancy care in Nigeria. Approximately half were evaluated with AI-guided screening using the digital stethoscope, and half received usual obstetric care in addition to a clinical ECG. An echocardiogram was used to confirm when the AI-enabled digital stethoscope predicted peripartum cardiomyopathy. Overall, 4% of the pregnant and postpartum women in the intervention arm of the clinical trial had cardiomyopathy compared to 2% in the control arm, suggesting that half are likely undetected with usual care.
The exercise stress test, which involves treadmill exercise test with electrocardiogram (ECG), is one of the most familiar tests in medicine. While exercise testing typically is focused on diagnosing coronary artery disease, a recent study from Mayo Clinic finds that exercise test abnormalities, such as low functional aerobic capacity, predicted non-cardiovascular causes of death such as cancer in addition to cardiovascular-related deaths. These new findings are published in Mayo Clinic Proceedings.
The exercise stress test is noninvasive, easily available and provides important diagnostic information. In addition to the ECG itself, the test produces data on functional aerobic capacity, heart rate recovery and chronotropic index, the standardised measure of heart rate during exercise that reflects age, resting heart rate and fitness.
“In our exercise testing cohort, non-cardiovascular deaths were more frequently observed than cardiovascular deaths,” says Thomas Allison, PhD, MPH, director of Mayo Clinic’s Integrated Stress Testing Center and the study’s senior author. “Though this was a cardiac stress test, we found that cancer was the leading cause of death, at 38%, whereas only 19% of deaths were cardiovascular. Exercise test results including low exercise capacity, low peak heart rate, and a slow recovery of the heart rate after exercise test were associated with increased mortality.”
The study looked at 13 382 patients who had no baseline cardiovascular issues or other serious diseases and who had completed exercise tests at Mayo Clinic between 1993 and 2010, then were followed closely for a median period of 12.7 years.
The findings suggest that clinicians should focus not only on ECG results but on data in the exercise test results such as low functional aerobic capacity, low chronotropic index and abnormal heart rate recovery. Patients should be encouraged to increase their physical activity if these results are atypical, even if the ECG results show no significant cardiovascular-related risk, Dr Allison says.
A smartwatch ECG can accurately detect heart failure (HF) in nonclinical environments, according to a study published in Nature Medicine. Researchers analysed Apple Watch ECG recordings with AI to identify patients with ventricular dysfunction. Study participants were able to remotely record their smartwatch ECGs at any time, with the data automatically and securely uploaded to their electronic health records via a smartphone app.
“Currently, we diagnose ventricular dysfunction – a weak heart pump – through an echocardiogram, CT scan or an MRI, but these are expensive, time consuming and at times inaccessible. The ability to diagnose a weak heart pump remotely, from an ECG that a person records using a consumer device, such as a smartwatch, allows a timely identification of this potentially life-threatening disease at massive scale,” says senior study author Paul Friedman, MD, chair of the Department of Cardiovascular Medicine at Mayo Clinic.
Ventricular dysfunction might not cause symptoms, but affects about 2% of the population and 9% of people over 60. Symptoms may develop with a low ejection fraction, including shortness of breath, a rapid heart rate and swelling in the legs. Early diagnosis is important because once identified, there are numerous treatments to improve quality of life and decrease the risks of heart failure and death.
Mayo researchers interpreted Apple Watch single-lead ECGs by modifying an earlier algorithm developed for 12-lead ECGs that is proven to detect a low ejection fraction.
While the data are early, the modified AI algorithm using single-lead ECG data had an area under the curve of 0.88 to detect low ejection fraction. By comparison, this measure of accuracy is as good as or slightly better than a medical treadmill diagnostic test.
“These data are encouraging because they show that digital tools allow convenient, inexpensive, scalable screening for important conditions. Through technology, we can remotely gather useful information about a patient’s heart in an accessible way that can meet the needs of people where they are,” says first author Zachi Attia, PhD, the lead AI scientist in the Department of Cardiovascular Medicine at Mayo Clinic.
“Building the capability to ingest data from wearable consumer electronics and provide analytic capabilities to prevent disease or improve health remotely in the manner demonstrated by this study can revolutionize health care. Solutions like this not only enable prediction and prevention of problems, but also will eventually help diminish health disparities and the burden on health systems and clinicians,” says co-author Bradley Leibovich, MD, the medical director for the Mayo Clinic Center for Digital Health.
Approximately 420 of the 2454 participants had an echocardiogram within 30 days of logging an Apple Watch ECG in the app. Of those, 16 patients had low ejection fraction confirmed by the echocardiogram, which provided a comparison for accuracy.
By chance, neuroscientists were able to record the activity of a dying human brain and discovered brain wave patterns similar to dreaming, memory recall, and meditation. An analysis of this case, reported in Frontiers in Aging Neurosciencesuggests a possible explanation for near-death experiences.
Imagine reliving your entire life in the space of seconds. Like a flash of lightning, you are outside of your body, watching memorable moments you lived through. This process, known as ‘life recall’, can be similar to what it’s like to have a near-death experience. What happens inside your brain during these experiences and after death are questions that have puzzled neuroscientists for centuries. However, the present study suggests that your brain may remain active and coordinated during and even after the transition to death, and may in fact be programmed to orchestrate the whole ordeal.
When an 87-year-old patient developed epilepsy, Dr Raul Vicente of the University of Tartu, Estonia and colleagues used continuous electroencephalography (EEG) to detect the seizures and treat the patient. During these recordings, the patient had a heart attack and passed away. This unexpected event allowed the scientists to record the activity of a dying human brain for the first time ever.
“We measured 900 seconds of brain activity around the time of death and set a specific focus to investigate what happened in the 30 seconds before and after the heart stopped beating,” said Dr Ajmal Zemmar, a neurosurgeon at the University of Louisville, US, who organised the study.
“Just before and after the heart stopped working, we saw changes in a specific band of neural oscillations, so-called gamma oscillations, but also in others such as delta, theta, alpha, and beta oscillations.”
Brain oscillations (aka ‘brain waves’) are patterns of rhythmic brain activity normally present in living human brains. These different types of oscillations, including gamma, are involved in high-cognitive functions, such as concentrating, dreaming, meditation, memory retrieval, information processing, and conscious perception, just like those associated with memory flashbacks.
“Through generating oscillations involved in memory retrieval, the brain may be playing a last recall of important life events just before we die, similar to the ones reported in near-death experiences,” Dr Zemmar speculated. “These findings challenge our understanding of when exactly life ends and generate important subsequent questions, such as those related to the timing of organ donation.”
Though this is the first study to ever measure live brain activity during the process of dying in humans, similar changes in gamma oscillations have been previously recorded in rats kept in controlled environments. This raises the possibility that, during death, the brain organises and executes a biological response that could be conserved across species.
The interepretation of this however is complicated by the fact that these measurements are based on a single case and stem from the brain of a patient who had suffered injury, seizures and swelling. Nonetheless, Dr Zemmar plans to investigate more cases and sees these results as a source of hope.
“As a neurosurgeon, I deal with loss at times. It is indescribably difficult to deliver the news of death to distraught family members,” he said.
“Something we may learn from this research is: although our loved ones have their eyes closed and are ready to leave us to rest, their brains may be replaying some of the nicest moments they experienced in their lives.”
Specific and dynamic changes on electrocardiograms (ECGs) of hospitalised COVID patients with COVID or influenza can help predict a timeframe for worsening health and death, according to a new Mount Sinai study.
Published in the American Journal of Cardiology, the study shows that shrinking waveforms on these tests can be used to help better identify high-risk patients and provide them more aggressive monitoring and treatment.
“Our study shows diminished waveforms on ECGs over the course of COVID illness can be an important tool for health care workers caring for these patients, allowing them to catch rapid clinical changes over their hospital stay and intervene more quickly. […] ECGs may be helpful for hospitals to use when caring for these patients before their condition gets dramatically worse,” said senior author Joshua Lampert, MD, Cardiac Electrophysiology fellow at The Mount Sinai Hospital. “This is particularly useful in overwhelmed systems, as there is no wait for blood work to return and this test can be performed by the majority of health care personnel. Additionally, the ECG can be done at the time of other bedside patient care, eliminating the potential exposure of another health care worker to COVID.”
Researchers did a retrospective analysis of ECGs on 140 hospitalised COVID patients across the Mount Sinai Health System in New York City, and compared them with 281 ECGs from patients with laboratory-confirmed influenza A or B admitted to The Mount Sinai Hospital. For each patient, the researchers compared three ECG time points: a baseline scan done within a year prior to COVID or influenza hospitalisation, a scan taken at hospital admission, and follow-up ECGs performed during hospitalisation.
They manually measured QRS waveform height on all electrocardiograms – changes in this electrical activity can indicate failing ventricles. The researchers analysed follow-up ECGs after hospital admission and analysed changes in the waveforms according to a set of criteria they designed called LoQRS amplitude (LoQRS) to identify a reduced signal. LoQRS was defined by QRS amplitude of less than 5mm measured from the arms and legs or less than 10mm when measured on the chest wall as well as a relative reduction in waveform height in either location by at least 50%.
Fifty-two COVID patients in the study did not survive, and 74% of those had LoQRS. Their ECG QRS waveforms reduced approximately 5.3 days into their hospital admission and they died approximately two days after the first abnormal ECG was observed.
Out of the 281 influenza patients studied, LoQRS was identified in 11 percent of them. Seventeen influenza patients died, and 39% had LoQRS present. Influenza patients met LoQRS criteria a median of 55 days into their hospital admission, and the median time to death was six days from when LoQRS was identified. Overall, these results show influenza patients followed a less virulent course of illness when compared to COVID patients.
“When it comes to caring for COVID patients, our findings suggest it may be beneficial not only for health care providers to check an EKG when the patient first arrives at the hospital, but also follow-up ECGs during their hospital stay to assess for LoQRS, particularly if the patient has not made profound clinical progress. If LoQRS is present, the team may want to consider escalating medical therapy or transferring the patient to a highly monitored setting such as an intensive care unit (ICU) in anticipation of declining health,” added Dr Lampert.
Screening for atrial fibrillation in 75- and 76-year-olds using thumb ECGS could reduce the risk of stroke, severe bleeding and death, according to a large-scale Swedish study.
Atrial fibrillation (AF) is associated with a five-fold increased risk of stroke, and the symptoms are often deleterious since large blood clots can form in the heart, breaking free and posing a stroke risk. Still, countries do not screen the general population for atrial fibrillation, but rather treat those patients who are discovered during routine care. This study by the Karolinska Institutet in Sweden and published in The Lancet, investigated the effectiveness of screening for AF.
“There has never really been a study that examines if it would be beneficial to screen for atrial fibrillation, which is why we wanted to investigate it,” said Emma Svennberg, cardiologist at the Karolinska University Hospital, Huddinge, and researcher at the Department of Medicine, Huddinge, Karolinska Institutet.
The study included almost 28 000 participants aged 75 or 76, randomised to be invited either to screening or to a control group, who received standard care. Of those invited to screening, more than half choose to participate. They completed a health questionnaire and performed a so-called thumb ECG (electrocardiogram), which involves placing one’s thumbs on an ECG device that measures the heart’s electrical activity.
Those without atrial fibrillation were asked to record their heart rhythm twice daily for two weeks using the ECG device which they took home. If the device registered irregular heart rhythms, the participants were referred to a cardiologist for a standardised work-up and, if there were no contra-indications, initiation of oral anticoagulant therapy.
The study’s 28 000 participants were then followed for at least five years. More detections of atrial fibrillation were recorded in the screening group, which also had a slightly lower incidence of death, stroke and severe bleeding than the control group.
“In total, 31.9 percent of those in the screening group experienced a negative event compared to 33 percent in the control group,” said Johan Engdahl, adjunct lecturer at the Department of Clinical Sciences, Danderyds Hospital, at Karolinska Institutet. “Now, that may sound like a small difference, but you must bear in mind that only about half of those invited to screening participated and it’s possible we would have seen a more pronounced difference had more people turned up for screening. Those who participated in the screening had significantly fewer negative events.”
Based on the findings, the researchers estimated that at least 2300 cases of stroke or death could be avoided per year in Sweden if a national screening of atrial fibrillation in the elderly was introduced.
Researchers at Tomsk Polytechnic University have developed a nanosensor-based system that can detect early abnormalities in the function of cardiac muscle cells, which otherwise can be recorded only with invasive procedures.
The nanosensor-based hardware and software complex can measure cardiac micropotential energies without filtering and averaging-out cardiac cycles in real time. The device allows registering early abnormalities in the function of cardiac muscle cells, which otherwise can be recorded only during open-heart surgery or by inserting an electrode in a cardiac cavity through a vein. Such changes can lead to sudden cardiac death (SCD). Nowadays, there are no alternatives to the Tomsk device for a number of key characteristics in Russia and the world. ).
The main method of detection of electrical pulses in the heart is electrocardiography (ECG). Nevertheless, ECG modern devices detect already critical changes in the function of the myocardium.
“Therefore, there is much concern about the creation of devices for early detection of these disorders, when it is still possible to restore cell function using medication and without surgical intervention. To implement this, it is required to record cardiac micropotential energies, electrical pulses emitted by separate cells. Here, there is a question of how to implement it noninvasive. Our research team have worked on this task for a long time, as a consequence, we jointly with the participation of our colleagues, doctors, have developed a hardware and software complex.
“The core principles of its operation are similar to ECG, however, we changed sensors: we made nanosensors instead of conventional sensors and managed to measure signals of nanovoltage and microvoltage layers without filtering and averaging-out in broadband. The use of nanosensors led to the necessity to apply original circuit solutions, write individual software.
“Ultimately, we gained a tremendous difference in sensitivity,” Diana Avdeeva, Head of the TPU Laboratory for Medical Engineering, a research supervisor of the project, said.
The system consists of a set of sensors, a tiny key device for recording incoming signals from sensors and data processing software. The sensors are fixed on a patient’s chest using a conducting gel, and the monitoring procedure takes about 20 minutes.
Conventional ECG machines operate on frequencies from 0,05 Hz to 150 Hz, while the device of the Tomsk scientists operates on much higher frequencies of up to 10 000 Hz.
“Silver chloride electrodes are usually used for recording ECG of high quality. Our sensors are also silver chloride electrodes, however, we used silver nanoparticles. There are up to 16 thin plates from porous ceramics in every our sensor, silver nanoparticles are placed in these pores. There are millions of particles in one sensor, where every particle is a silver chloride electrode capable to enhance an electric field of the heart. Silver and gold nanoparticles are capable to enhance an electromagnetic field: visible light by 10,000 folds and infrared radiation by 20 folds. We also refused to use filters for rejection network interference and noises, which are usually used in conventional ECG and significantly distort micropotentials,” Diana Avdeeva said.
The published study includes the monitoring data of one volunteer’s heart function, who took part in the research for four years and was monitored every 7-10 days.
“At the beginning of our research, we recorded clear violations of activity of cardiac muscle cells. His attending physician recommended surgery, he gained an inserted stent at the Cardiology Research Institute. Then, he continued to take part in the research and the device recorded the further gradual restoration of cardiac function,” the scientist noted.
“A task to create a sensitive, tiny and affordable complex was set up, in order in a long run, outpatient clinics and patients at home could use it. Moreover, the developed methods and devices can be used not only in cardiology.
“The fields of any electrophysiological research, such as electroencephalography, electromyography and so on are promising. Of course, before applying it to cardiology, we have to pass some essential stages. These are the collection of the required array of statistics, certification of the complex for medical use. All these stages require sponsorship, we are engaged in searching for partners and supporting programs,” said research team member Mikhail Yuzhakov, Engineer at the TPU Laboratory for Medical Engineering.
