– Researchers from Mount Sinai have been awarded a four-year, $3 million grant from the Nationwide Coronary heart, Lung, and Blood Institute of the Nationwide Institutes of Well being (NIH) to develop synthetic intelligence (AI)-driven prediction fashions to flag danger of heart problems occasions in sufferers with obstructive sleep apnea.
The American Coronary heart Affiliation (AHA) signifies that obstructive sleep apnea will increase sufferers’ danger of heart problems, together with coronary artery illness, hypertension and stroke. Using steady constructive airway stress (CPAP) machines is usually prescribed to deal with sleep apnea, however proof to counsel the advantages of CPAP use in relation to cardiovascular occasion charges is proscribed.
To bridge this hole, the analysis workforce will construct machine studying (ML) instruments to determine obstructive sleep apnea sufferers at excessive danger for atherosclerosis development and cardiovascular occasions like stroke and coronary heart assault.
The researchers underscored that this method might assist forecast cardiovascular remedy effectiveness of CPAP use in sufferers thought of “non-sleepy” based mostly on their responses to a medical take a look at. Doing so could make clear which sufferers would profit most from CPAP remedy and which ought to keep away from the remedy.
This work relies on earlier Mount Sinai analysis highlighting the potential hurt of CPAP use in non-sleepy sufferers, necessitating an elevated give attention to personalised sleep apnea remedy.
READ MORE: $4M Grant to Fund Growth of Sleep Apnea End result Prediction Instruments
“We’re impressed by the transformative potential of machine studying methods in well being care, notably in analyzing huge quantities of advanced information to personalize remedy methods,” said principal investigator Girish Nadkarni, MD, MPH, the Irene and Dr. Arthur M. Fishberg Professor of Medication, director of The Charles Bronfman Institute of Personalised Medication, and system chief of Knowledge-Pushed and Digital Medication at Icahn Mount Sinai, within the press launch. “Our research has the potential to revolutionize the administration of obstructive sleep apnea by providing choice assist instruments that optimize remedy plans, enhance affected person outcomes, and scale back the burden of sleep apnea-related heart problems occasions on each people and well being care techniques.”
To construct their predictive fashions, the researchers will make the most of information from two cohorts: the Sleep Apnea Cardiovascular Endpoints (SAVE) randomized medical trial – containing data from over 2,500 non-sleepy members with reasonable to extreme obstructive sleep apnea and established heart problems – and the Multi-Ethnic Examine of Atherosclerosis (MESA) cohort – made up of information from over 6,000 non-sleepy members who’re usually wholesome and are available from ethnically various backgrounds.
These information shall be used to flag key predictors of atherosclerosis development and cardiovascular occasions alongside figuring out risk-based affected person subgroups with differential CPAP remedy outcomes for cardiovascular occasions.
The predictive fashions shall be validated utilizing digital well being document (EHR) information from Mount Sinai.
“By precision medication, we’re prioritizing rigorous intervention to reinforce heart problems danger discount,” mentioned principal investigator Mayte Suarez-Farinas, PhD, affiliate director of the Heart for Biostatistics, and professor of Inhabitants Well being Science and Coverage, and Genetics and Genomic Sciences at Icahn Mount Sinai. “Well being care suppliers shall be outfitted with progressive instruments to determine sufferers at heightened danger for coronary heart assault or stroke and have the ability to predict remedy outcomes of CPAP remedy in sleep apnea sufferers. This personalised method will allow clinicians to tailor remedy methods to particular person affected person wants, optimizing CPAP adherence and efficacy.”
READ MORE: Explaining the Fundamentals of Affected person Danger Scores in Healthcare
This analysis furthers Mount Sinai’s efforts to enhance sleep apnea care by means of superior analytics applied sciences.
“Supported by a transformative grant, I’m thrilled to guide a venture that stands on the intersection of cutting-edge synthetic intelligence and sleep medication,” said main principal investigator Neomi Shah, MD, MPH, MSc, affiliate dean for School Profession Development, vice chair for School Affairs within the Mount Sinai Well being System Division of Medication, and professor of Medication (Pulmonary, Crucial Care and Sleep Medication) on the Icahn Faculty of Medication at Mount Sinai.
“Our work will epitomize the wealth of experience and collaborative effort throughout the Mount Sinai Well being System to each enrich our understanding of the situation and enhance affected person care, impacting tens of millions in america. We’re dedicated to validating our AI instruments inside Mount Sinai’s medical dataset to translate our analysis into real-world apply, thereby, successfully bridging the analysis to apply hole,” Shah continued.
Final month, a Mount Sinai workforce was awarded a $4.1 million NIH grant to develop AI-driven fashions to foretell adversarial outcomes in sleep apnea sufferers.
The researchers famous that present diagnostic instruments for sleep apnea – which generally depend on counting the variety of respiratory disturbances a affected person experiences throughout sleep – will not be correct predictors of short- and long-term adversarial outcomes like extreme daytime sleepiness, cardio-cerebrovascular morbidity and neurocognitive impairment.
READ MORE: Utilizing Danger Scores, Stratification for Inhabitants Well being Administration
The AI fashions are designed to handle the shortcomings of those instruments by analyzing hypoxic, arousal and ventilatory variables to generate a personalised affected person danger rating.
Replace 2024/03/14: This text has been up to date with a quote from Dr. Neomi Shah.