COVID-19 is a life threatening outbreak since December 2019. Physicians, nurses and other heroes working on the front-line to cure infected people. On the other side, scientists and researchers trying to find a faster, cheaper, better options for COVID-19 diagnosis and treatment. One part of their researches is absolutely artificial intelligence and deep learning.

I listed on PyMed and below, peer-reviewed & published researches that using artificial intelligence for COVID-19. If you find any other published research, you can send me.

Diagnosis with Radiology

1. COVIDiagnosis-Net: Deep Bayes-SqueezeNet based diagnosis of the coronavirus disease 2019 (COVID-19) from X-ray images

The Coronavirus Disease 2019 (COVID-19) outbreak has a tremendous impact on global health and the daily life of people…

I used Crestle before for deep learning and image processing. It was charging $0.59 per hour and it was using Nvidia Tesla K80. As the dollar rises in Turkey(6.01 TRY = $1 now), hourly cost increases. As physical GPU prices have increased in the same way, I think that the best solution in the short term is still a cloud. That’s why I started looking for a cheaper cloud system.

Already the first to come to mind when you mention the cloud GPU is AWS and Google Cloud. I looked into the price of these two and I saw it…

We have just finished the June. So, I put together some exciting researches from I read in June. Please comment your favorite researches in June except these :)

Long-term reduction in hyperglycemia in advanced type 1 diabetes: the value of induced aerobic glycolysis with BCG vaccinations

This paper is published in Nature by 12 researchers. According to this research BCG vaccine leads to long-term blood sugar improvement in type 1 diabetes patients.

They reported on a randomized 8-year long prospective examination of type 1 diabetic subjects with long-term disease who received two doses of the BCG vaccine. After year 3, BCG lowered hemoglobin A1c to near normal levels for the next 5 years. The BCG impact on blood…

Research is a sum of asking questions and then trying to answer them. But it is not easy as somebody think. Because there are lot of questions from creation of man’s until today. Some of them has been answered but most of them still need answer. So researcher should know which one is answered or not, which one needs more explanation, which one needs more trials and another things.

From Past to Today Making a Research

The Knick

According to the past, accessing the true knowledge and information is so easy today. In past, researchers communicated with mails, discussed in congresses, published their articles by printed journals. But today…

I’ve been working on machine learning algorithms recently, especially deep learning. It is really successful method for unresolved problems in the computer science. I believe that we will talk more about ‘deep learning’ in upcoming days.

There is no doubt that deep learning and neural networks are effective to solve problems, but there are some problems when you try to fit a deep learning model with a stack of the data. One of them is computing power which is supplied by GPUs generally.

Finding a GPU to test your neural network is a painful process, especially if you are newbie…

Abdüssamet Aslan

MD, emergency physician, indie software developer, ai researcher, blogger

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store