How a Data Lake Could Help Italy Rebound from COVID-19
Scenes from Italy at the dawn of the pandemic seemed otherworldly. Patients, often struggling to breathe, lined cramped hallways and waiting rooms in hospitals that had run out of beds. The death toll leaped into the thousands, warning the rest of the world of COVID-19’s propensity for ruin.
Italy has made considerable ground in its fight against the coronavirus, and now researchers there believe they’re months away from getting a firmer grip on the COVID-19. Milan’s largest hospital, Ospedale Niguarda, recently launched a project whose use of data (from EPR and PACS) and artificial intelligence (AI) could shore up our understanding of treatments and prevent future deaths and outbreaks. The initiative, CoviLake Niguarda, is designed to identify biomarkers that predict how the mysterious disease will develop in patients, unlocking stronger and quicker diagnoses and treatments for a rapidly reopening world.
The project’s backbone is a data lake from InterSystems and a vendor-neutral archive, which Niguarda will soon feed 1,000 patients’ anonymized medical images and clinical data, including X-rays, blood tests, and more. The data lake, an open container, will then correlate the data for use in training AI algorithms and developing a clinical decision support system. In six months, researchers expect CoviLake Niguarda to answer pressing clinical questions.
“Current observations and predictions can be complemented with data from years ago, if needed,” write Prof. Angelo Vanzulli, MD, Ospedale Niguarda’s director of advanced technologies department, and Alberto Torresin, PhD, director of medical physics. “Current imaging and clinical data, whether the patient had been in the hospital before, what their treatment was, whether there are any underlying conditions—all this information can be extracted, added to the data lake, and used to support the decision-making process.”
While swab tests take hours to produce a result, the algorithms developed through CoviLake Niguarda could determine almost instantly whether a patient likely has COVID-19. The technology is also poised to promptly decipher between COVID-19 cases and similar diseases, which researchers say could help prevent future outbreaks.
Radiologists, emergency department physicians, anesthesiologists, infectious disease specialists, medical physicists, and other physicians stand to benefit from CoviLake Niguarda. The work could deliver more accurate prognoses, which would aid hospitals in stratifying patients by disease risk. Patients, meanwhile, could receive more effective treatments based on their biomarkers and active monitoring during and after hospitalization.
“COVID-19 is a complex disease, and detailed information about the virus and its mechanics is still unknown,” Vanzulli and Torresin note. “This adds to the burden of clinicians on the front line.”
But CoviLake Niguarda could unearth the insights required to fuel the next stage of the global COVID-19 response—with less burden on people and more faith in cutting-edge technology.
This story originally appeared the July 10, 2020 – HealthShare Connections News Flash No.3: COVID-19 Pandemic newsletter