Lunit’s technology not only helps doctors who are experts in reading X-ray images, but also those without extensive radiology skills. This has proven valuable during the pandemic.
“Physicians from various fields such as orthopedics, psychiatry, and family medicine, have been taking part in (COVID-19) treatment. They are not trained radiologists, so they are limited in their ability to read X-ray test results,” Lee said.
“X-rays only appear in black and white, which means that there are cases where lesions aren’t noticeable to the human eye. Lunit’s solution has the advantage of displaying lesions in vivid color, which makes them more noticeable.”
It can help medical staff make informed decisions. But a patient’s fate is not put solely in the hands of AI. Cases that have been flagged by the technology are followed up and the results are double-checked by doctors.
Lee says the processing scale and accuracy of Lunit’s technology can help free up busy radiologists.
“An X-ray is a compressed two-dimensional rendering of three-dimensional human structures. Inevitably, organs and structures overlap in the images, which can make it easy for the human eye to miss lesions,” he said.
“The reality for radiologists, especially in Korea, is that it’s impossible to invest a lot of time in reading each X-ray as they would have to read hundreds or thousands every day.”
A radiologist uses Lunit technology at the Seoul National University Hospital in South Korea. (Photo: Lunit)
Using the cloud computing power of Microsoft Azure, Lunit’s technology generates location information of detected lesions in the form of heatmaps. It also makes abnormality scores that reflect the probability that the detected lesion is abnormal and needs further investigation by a radiologist.
Even seasoned radiologists can sometimes miss vital details when under pressure. Lee recalls a recent case where Lunit’s algorithms found a COVID-19-related lesion in a patient’s lung that had gone undetected by doctors.
If this had remained unnoticed and untreated,
This article is purposely trimmed, please visit the source to read the full article.
The post AI and X-rays: Identifying the many faces of COVID-19 appeared first on Microsoft | The AI Blog.