Researchers have developed a deep learning network capable of accurately predicting a person’s risk of adverse cardiac events, presenting their findings virtually at the Society of Nuclear Medicine and Molecular Imaging (SNMMI) 2021 Annual Meeting.

The new analysis included data from more than 20,000 patients who underwent single photon emission CT (SPECT) myocardial perfusion imaging (MPI). The advanced algorithm used those SPECT MPI results to determine each patient’s risk of a major adverse cardiac event—myocardial infarctions or cardiovascular deaths, for example—and then patients were followed for an average of nearly five years to test the algorithm’s accuracy.

Overall, the authors found, the annual rate of major adverse cardiac events among patients with the highest deep learning scores was 9.7%. This represented a 10.2-fold increase compared to the annual rate among patients with the lowest scores.

“These findings show that artificial intelligence could be incorporated in standard clinical workstations to assist physicians in accurate and fast risk assessment of patients undergoing SPECT MPI scans,” Ananya Singh, MS, a research software engineer in the Slomka Lab at Cedars-Sinai Medical Center in Los Angeles, said in a prepared statement. “This work signifies the potential advantage of incorporating artificial intelligence techniques in standard imaging protocols to assist readers with risk stratification.”

The post Combining AI with cardiac imaging helps predict heart attacks, cardiovascular deaths appeared first on Artificial Intelligence.

Heart disease and cancer are the leading causes of death in the United States, and it’s increasingly understood that they share common risk factors, including tobacco use, diet, blood pressure, and obesity. Thus, a diagnostic tool that could screen for cardiovascular disease while a patient is already being screened for cancer, has the potential to expedite a diagnosis, accelerate treatment, and improve patient outcomes. 

In research published today in Nature Communications, a team of engineers from Rensselaer Polytechnic Institute and clinicians from Massachusetts General Hospital developed a deep learning algorithm that can help assess a patient’s risk of cardiovascular disease with the same low-dose computerized tomography (CT) scan used to screen for lung cancer. This approach paves the way for more efficient, more cost-effective, and lower radiation diagnoses, without requiring patients to undergo a second CT scan. 

“In this paper, we demonstrate very good performance of a deep learning algorithm in identifying patients with cardiovascular diseases and predicting their mortality risks, which shows promise in converting lung cancer screening low-dose CT into a dual screening tool,” said Pingkun Yan, an assistant professor of biomedical engineering and member of the Center for Biotechnology and Interdisciplinary Studies (CBIS) at Rensselaer.

Numerous hurdles had to be overcome in order to make this dual screening possible. Low-dose CT images tend to have lower image quality and higher noise, making the features within an image harder to see. Using a large dataset from the National Lung Screening Trial (NLST), Yan and his team used data from more than 30,000 low-dose CT images to develop, train, and validate a deep learning algorithm capable of filtering out unwanted artifacts and noise, and extracting features needed for diagnosis. Researchers validated the algorithm using an additional 2,085 NLST images.

The Rensselaer team also partnered with Massachusetts General Hospital, where researchers were able to test this deep learning approach against state-of-the-art scans and the expertise of the hospital’s radiologists. The Rensselaer-developed algorithm, Yan said, not only proved to be highly effective in analyzing the risk of cardiovascular disease in high-risk patients using low-dose CT scans, but it also proved to be equally effective as radiologists in analyzing those images. In addition, the algorithm closely mimicked the performance of dedicated cardiac CT scans when it was tested on an independent dataset collected from 335 patients at Massachusetts General Hospital.

“This innovative research is a prime example of the ways in which bioimaging and artificial intelligence can be combined to improve and deliver patient care with greater precision and safety,” said Deepak Vashishth, the director of CBIS.

Yan was joined in this work by Ge Wang, an endowed chair professor of biomedical engineering at Rensselaer and fellow member of CBIS. The Rensselaer team was joined by Dr. Mannudeep K. Kalra, an attending radiologist at Massachusetts General Hospital and professor of radiology with Harvard Medical School. This research was funded by the National Institutes of Health National Heart, Lung, and Blood Institute. 

The post Deep Learning Enables Dual Screening for Cancer and Cardiovascular Disease appeared first on Artificial Intelligence.

September 11, 2020 – A machine learning model was able to detect different clusters of gut bacteria that could potentially identify individuals with existing cardiovascular disease (CVD), according to a study published in the journal Hypertension.

Recent studies have found a link between gut microbiota, the microorganisms in human digestive tracts, and CVD, the leading cause of mortality worldwide. Gut microbiota is highly variable between individuals, and differences in gut microbial compositions between people with and without CVD have been reported.

“Based on our previous research linking gut microbiota to CVD in animal models, we designed this study to test whether it is possible to screen for CVD in humans using artificial intelligence screening of stool samples,” said Bina Joe, PhD, FAHA, the study director, Distinguished University Professor and Chairwoman of the department of physiology and pharmacology at the University of Toledo in Toledo, Ohio.

“Gut microbiota has a profound effect on cardiovascular function, and this could be a potential new strategy for evaluation of cardiovascular health.”

Researchers used data from the American Gut Project, an open platform for microbiome research based in the US. The team leveraged innovative machine learning methods to analyze microbial composition of nearly 1,000 stool samples. Approximately half of the samples were from people with CVD.

Researchers found that the model was able to identify different clusters of gut bacteria that could potentially help identify individuals with and without CVD. The results show the potential for machine learning and artificial intelligence to improve CVD screening.

“Despite the fact that gut microbiomes are highly variable among individuals, we were surprised by the promising level of accuracy obtained from these preliminary results, which indicate fecal microbiota composition could potentially serve as a convenient diagnostic screening method for CVD,” Joe said.

“It is conceivable that one day, maybe without even assessing detailed cardiovascular function, clinicians could analyze the gut microbiome of patients’ stool samples with an artificial machine learning method to screen patients for heart and vascular diseases.”

Researchers have previously used machine learning to improve cardiovascular disease treatment and detection. A study published in Cardiovascular Research showed that machine learning tools could predict patients’ long-term risk of heart attacks and cardiac deaths better than standard methods used by cardiologists.

In that study, researchers found that subjects’ predicted machine learning scores aligned accurately with the actual distribution of observed events.

“In this prospective trial, machine learning demonstrated high performance in risk assessment for myocardial infarction and cardiac death in asymptomatic subjects,” the researchers stated.  

“By objectively combining clinical data and quantitative CT measures, machine learning provided significantly superior risk prediction compared with the coronary artery calcium score. These promising results suggest that machine learning has a potential for clinical implementation to improve risk assessment.”

In a separate study published in 2018, a team from Stanford University were able to create a personal health management tool that combined EHR data and machine learning algorithms to accurately diagnose the heart condition known as abdominal aortic aneurysm (AAA).

Researchers used the hierarchical estimate from agnostic learning (HEAL) health management tool to analyze patients’ genome sequences and EHRs to identify individuals with the cardiovascular disease.

“For each individual, HEAL accurately predicted his/her AAA risk using personal genome and EHR data. On the other hand, for the same individual with newly adopted lifestyles resulting in physiological changes (e.g., from a high cholesterol to a low cholesterol diet), HEAL can immediately update his/her AAA risk upon corresponding changes conditioned on the person’s genome baseline,” Stanford researchers said.

“This allows us to further investigate the interplay between personal genomes and lifestyles underlying disease predisposition.”

As machine learning and AI become more widely used in healthcare, these tools could make the transition from research to clinical care for improved patient health and outcomes. 

The post Machine Learning Could Improve Cardiovascular Disease Screening appeared first on Artificial Intelligence.