OR WAIT null SECS
The tech start-up has developed machine learning algorithms to identify sepsis and other health issues. CEO Andrew Pucher is excited for the future but knows it’s a long road.
As healthcare evolves, providers aren’t just aiming to treat disease. They’re aiming to prevent disease and predict health complications as soon as possible.
Dascena, a start-up based in San Francisco, aims to give providers artificial intelligence tools to detect certain health issues much earlier and allow providers to act sooner. The company utilizes machine learning algorithms, programs which evaluate enormous amounts of data, to predict healthcare complications.
The company’s tagline is “we bring data to life,” and Dascena has achieved some notable successes.
Dascena has developed a machine learning algorithm for the early detection of sepsis, the body’s extreme reaction to an infection. Without proper treatment, sepsis can lead to organ failure and death. About 270,000 people die of sepsis in the U.S. each year, according to the U.S. Centers for Disease Control and Prevention.
The Food and Drug Administration has awarded “breakthrough device designation” for two of Dascena’s products that have shown success in early detection of early kidney injury and for gastrointestinal bleeding.
Andrew Pucher, the CEO of Dascena, said in an interview with Chief Healthcare Executive that it’s an exciting time for the company. Pucher was named CEO in July 2021, succeeding company founder Ritankar Das, who remains on the board.
“It’s been a lot of fun,” Pucher said. “It’s been really energizing.”
“We are at the forefront, us and a small number of other companies, of this emerging space, the application of machine learning algorithms for different diseases in the in-patient setting,” he said.
In 2020, Dascena received emergency use authorization from the FDA for COViage, an algorithm used to help determine if COVID-19 patients were likely to experience unstable blood pressure or declines in their respiratory system.
Machine learning algorithms are showing increasing promise in the ability to detect health issues.
In a peer-reviewed study published in the American Journal of Infection Control Thursday, Dascena researchers found that machine learning algorithms showed success in predicting which hospital patients will develop infections. The study examined the use of such programs in predicting Clostridiodes difficile (C-diff) infections in hospitals, which can lead to excessive diarrhea and can be life-threatening.
Linda Dickey, president of the Association for Professionals in Infection Control and Epidemiology, said that study backs earlier research showing the promise of machine learning algorithms in predicting infections.
Dickey said in a statement the research suggests machine learning algorithms “provide reliable infection-risk prediction that can empower clinical teams to implement appropriate infection control measures at earlier time points and thereby improve healthcare outcomes.”
Dascena’s flagship algorithm, InSight, is being used in some hospitals. InSight uses vital sign data to predict the onset of sepsis sooner. The software takes data from electronic health records and sends alerts to the patients’ care team that a patient is at risk of sepsis.
The company customizes its software with its hospitals.
“We work directly in close partnership in hospital systems when we are deploying the product,” Pucher said. “We work to determine what the right threshold is to send an alert for that hospital.”
In addition, Dascena also customizes the way healthcare professionals get the alerts about warning signs of sepsis. Some hospitals want text messages, while others want phone calls.
Clinical studies, including research published in BMJ Open Respiratory Research, have shown InSight’s value, he said.
“We haven’t just built a piece of technology,” Pucher said. “We've proven to physicians that it does work.”
Pucher sees huge potential in developing tools that can help predict health problems or at least detect them earlier.
“This is a natural progression, the integration of machine learning and artificial intelligence tools in healthcare,” Pucher said.
Those tools won’t replace the roles of doctors anytime soon, he said. But he added, “Will AI be supplementing what physicians are doing, and make better decisions? Certainly.”
With the COVID-19 pandemic, Pucher said it can be challenging connecting with hospitals to talk about the company’s products. Healthcare systems are treating an enormous number of COVID-19 patients and systems are dealing with both high patient volume and staffing shortages. Understandably, some hospital leaders have less time for pitches on new products.
Still, Pucher said he is very optimistic for Dascena’s prospects. Before joining the company, he was chief corporate development officer at Tilray, a medical cannabis firm, and he was formerly a managing director at Goldman Sachs.
Dascena now employs about 100 workers and is actively hiring. While Pucher is bullish on the company’s future, he’s also patient. Pucher said success will take time and he’s working to build the business around the company’s products.
“It’s a very exciting time in this space,” he said. There’s growing recognition of the value of machine learning as a tool in decision-making in healthcare.
But he also said, “You need to play the long game here, especially with the products we’re building.”