Researchers develop new algorithm to minimize diagnostic errors

Researchers at the University of Copenhagen have developed an algorithm to identify patients who may have been misdiagnosed. Using the digital disease history, the algorithm is able to record disease trajectories that differ so much from normal trajectories that there may be a misdiagnosis. The algorithm was developed on the basis of data from several hundred thousand patients with COPD.

This does not happen often. But on rare occasions, doctors make mistakes and can misdiagnose. Patients may have multiple illnesses at once, where it may be difficult to distinguish symptoms of one illness from another, or there may be an absence of symptoms.

Misdiagnosis can lead to incorrect treatment or a lack of treatment. Therefore, everyone in the healthcare system tries to minimize errors as much as possible.

Now researchers at the University of Copenhagen have developed an algorithm that can help with this.

Our new algorithm can find patients who have such an unusual disease trajectory that they may indeed not be suffering from the disease they were diagnosed with. It is hoped that it can become a supporting tool for doctors. “

Isabella Friis Jørgensen, postdoc, Novo Nordisk Foundation Protein Research Center

Algorithm revealed possible lung cancer

The researchers developed the algorithm based on disease trajectories for 284,000 patients with chronic obstructive pulmonary disease (COPD), from 1994 to 2015. Based on these data, they came up with approximately 69,000 typical disease trajectories. .

“In the national patient registry, we were able to map what you might call the typical trajectory of the disease. And if a patient presents with a very unusual disease trajectory, the patient may simply be suffering from a different disease. Our tool can help detect this, ”says Søren Brunak, professor at the Novo Nordisk Foundation Center for Protein Research.

For example, researchers found a small group of 2,185 COPD patients who died very soon after being diagnosed with COPD. According to the researchers, it was a sign that something else could have been wrong, possibly something even more serious.

“When we took a closer look at the laboratory values ​​of these patients, we found that they deviated from normal values ​​for patients with COPD. Instead, the values ​​looked like something seen in patients with lung cancer. Only 10% of these patients were diagnosed with lung cancer, but we are reasonably convinced that most, if not all of these patients actually had lung cancer, ”says Søren Brunak.

Data that will bring immediate benefit

Although the algorithm has been validated by data from patients with COPD, it can be used for many other diseases. The principle is the same: the algorithm uses data from the registry to map typical disease trajectories and can detect if the disease trajectory of certain patients stands out so much that something is wrong.

“Of course, our most important goal is to get patients value for money when it comes to their health care. And we believe that in the future this algorithm may end up becoming a supporting tool for physicians. Once the algorithm has mapped the typical trjaectories of the disease, it only takes 10 seconds to compare a single patient to everyone, ”says Søren Brunak.

He stresses that the algorithm needs to be further validated and tested in clinical trials before it can be implemented in Danish hospitals. But he hopes it’s something that can be started soon.

“In Denmark, we often praise our good health registries because they contain valuable data for researchers. We use them in our research because it may benefit other people in the future in the form of better treatment. But this is actually an example of how your own health data can benefit you right away, ”says Søren Brunak.