{"id":36559,"date":"2019-05-02T17:16:33","date_gmt":"2019-05-02T21:16:33","guid":{"rendered":"https:\/\/sciencebusiness.technewslit.com\/?p=36559"},"modified":"2019-05-03T11:11:57","modified_gmt":"2019-05-03T15:11:57","slug":"kidney-diagnostics-receive-fda-breakthrough-tag","status":"publish","type":"post","link":"https:\/\/technewslit.com\/sciencebusiness\/?p=36559","title":{"rendered":"Kidney Diagnostics Receive FDA Breakthrough Tag"},"content":{"rendered":"
\"Data<\/a>
(Gerd Altmann, Pixabay)<\/figcaption><\/figure>\n

2 May 2019. An algorithm-driven diagnostic test for chronic kidney disease is receiving a breakthrough device designation from the Food and Drug Administration. The designation is given to the test called KidneyIntelX developed by the start-up company Renalytix AI<\/a> in New York.<\/p>\n

Renalytix AI<\/a> is a spin-off enterprise from Mount Sinai medical center in New York that provides diagnostics for chronic kidney disease to detect the condition earlier, allowing for more focused care before the disorder progresses and causes irreparable damage. Damage to the kidneys can occur as a result of type 2 diabetes and high blood pressure, but often has few noticeable symptoms on its own, and as a result is not diagnosed until complications such as anemia occur.<\/p>\n

National Kidney Foundation says some 30 million Americans have chronic kidney disease<\/a>, with many more at risk, including people with diabetes and in some ethnic groups including African-Americans, Hispanics, Pacific Islanders, and American Indians. The company cites data showing chronic and end-stage kidney disease cost the U.S. health care system some $114 billion each year.<\/p>\n

KidneyIntelX is an upgrade of a program called KidneyTrack<\/a>, written by Mount Sinai kidney specialists Steven Coca<\/a> and Girish Nadkarni<\/a>, also the scientific founders of Renalytix AI. The program looks for characteristic blood-based genetic biomarkers for chronic kidney disease, but also combs through electronic health records for clinical indicators of kidney damage or disease over time.<\/p>\n

These data are then fed into machine-learning algorithms, known as random forest algorithms, that predict the probability of kidney function decline some 5 years before the disease progresses. Random forest algorithms<\/a> build data into decision-trees, then merge the data from the trees together, thus the \u201cforest\u201d reference, for more accurate and stable predictions.<\/p>\n

A paper<\/a> by Coca, Nadkarni, and colleagues posted in March 2019 demonstrates the KidneyIntelX algorithms in action. The authors sampled 1,369 non-identified individuals from Mount Sinai’s BioMe health records database<\/a> either with type 2 diabetes or with African heritage and expressing a high-risk gene for kidney disease. The results show the algorithms better predicted rapid chronic kidney decline than current standard clinical models, as well as people already with kidney disease but not likely to experience further decline in kidney function, about 1 in 3 of those sampled. (The BioMe database was also used to train the algorithm.)<\/p>\n

Coca and Nadkarni founded Renalytix AI last year and are now scientific advisors<\/a> to the company. In May 2018, Renalytix AI licensed<\/a> the technology developed by Coca and Nadkarni, but also gained access to the medical center’s BioMe and general electronic health record data sets for algorithm training. The license gives the company exclusive access to the technology to develop A.I. solutions for kidney disease prediction and detection among at-risk populations.<\/p>\n

The breakthrough device<\/a> designation began in December 2017, building on an earlier accelerated review process called Expedited Access Pathway. FDA says the designation recognizes new devices that offer meaningful advantages over currently cleared technologies, addressing unmet medical needs, and where no other current alternatives exist. Breakthrough devices are given higher priority and more attention from FDA staff to reduce the time needed for review under the agency’s medical device pathways.<\/p>\n

More from Science & Enterprise:<\/p>\n