Why Epic’s Algorithm in Medicine is a Failure

Epic Systems is the largest company with electronic health records in the United States and keeps a record of information of over 180 million patients across the country. The data accounts for almost 56% of the entire population.

The company uses the slogan “with the patient at the heart” and consists of a portfolio of twenty proprietary artificial intelligence (AI) Algorithm designed to detect illnesses and predict the length of all the hospital stays.

There are several other proprietary Algorithm in the healthcare sector and other industries where the users have no provision for understanding the programs of Epic and whether they are reliable or just as some marketing trick by the company. The details given under the black boxes are confidential, and the independent tests are low in number. Therefore, there is considerable gaps in the Algorithm.

There are many deaths caused by one of the most important Algorithm of medicine for the prediction of sepsis. Sepsis is a condition that occurs when the body of a patient overreacts to an infection and then sends the chemicals in the blood, which are likely to cause the failure of several organs or even tissue damage. Early detection of the disease is likely to be of great advantage to the patient. However, the early detection of sepsis is rarely possible. The company claims that their predictions made by the Epic Sepsis model hold an accuracy rate of 76% to 83%.