Back to blog

Artificial Intelligence against COVID-19: easy to use solutions for healthcare organizations

April 08, 2022

9 minutes read

Can technology help us in the fight against the virus? There are already a lot of solutions developed to protect people from getting infected. For example, the “Stop Corona” app from the Austrian Red Cross society. It keeps an anonymous contact diary that logs personal encounters using a “digital handshake.” If somebody has symptoms of COVID-19, all those who “have digitally shaken hands” with that person will be automatically informed about one of their anonymous contacts showing signs of an infection. All these people are asked to isolate themselves to reduce the infection’s risk. This way, fewer people will be exposed to potential diseases. Additionally, this system supports and relieves doctors who would otherwise have to go through this process manually.

It is a classic example of a maintenance strategy that can let machines help you to avoid danger. But can devices act reactively as well?

What’s wrong with the existing testing approach?

Imagine that you already have some alarming symptoms. What can you do besides stay at home and pray? The most popular way to diagnose the virus is a test. But is testing the best, quickest, and most effective method? The first problem with tests is that they are not always available. The second problem is that it simply takes too long.

What are the alternatives to RT-PCR testing?

One of the first researches of possible alternatives was made at the very beginning of the pandemic, in January-February 2020 at Tongji Hospital in Wuhan, China. About 1000 patients were not only RT-PCR tested but had chest x-ray and CTs as well. It found that the sensitivity of CT for COVID-19 infection was 27 percent higher than the sensitivity of RT-PCR testing.

This led us to assume that CT might offer an alternative way to help diagnose COVID-19 positive patients. The low sensitivity of RT-PCR implies that many COVID-19 patients can not be identified quickly enough and may not receive needed treatment.

“Compared to RT-PCR, chest CT imaging may be a more reliable, practical, and rapid method to diagnose and assess COVID-19, especially in the epidemic area”, – the Radiology study authors wrote.

How can artificial intelligence help fight COVID-19?

The symptoms of COVID-19 are very similar to other lung diseases. It is often confusing, but machines can help us to distinguish.

Machine learning solutions that can detect COVID-19 and other lung diseases help to improve the diagnostic process. With these solutions, doctors can recognize a lot of conditions very quickly. Such applications can analyze x-rays and CT/MRT images in png, jpeg, or DICOM files format. They help the doctors to classify about 14 different diseases such as ARDS, Bacterial Pneumonia, COVID-19, Chlamydophila, Fungal Pneumonia, Klebsiella, Legionella, Pneumocystis, Pneumonia, Streptococcus, Viral Pneumonia. VITech can develop such products for our potential clients.

Effectively detecting COVID-19 and other lung diseases, doctors can make the process of diagnosis easier, faster, and more precise. The ML-based apps offer the doctor better visibility of infected areas because they are highlighted. Also, he helps to identify a particular disease more quickly, suggesting diagnoses from the database. It is an important assistant in situations like now when a lot of patients need medical help immediately, and time for each patient has significantly decreased.

How to be safe visiting a doctor?

Visit medical institutions during a time of the pandemic means entering a high risks area. We need to do more to reduce risks to medical personnel. “Stop Corona” app would recognize the visit to the doctor as a “digital handshake” and send an alarm signal to you and your contact persons. Is it possible to reduce the risk for both sides: patients and doctors?

Acting preventive

Theoretically, everybody knows the necessity of wearing a medical mask or respirator, gloves and eyeglasses, washing hands, and staying at a distance from other people. But how to make sure that everybody follows the rules? How to control all people in the medical institution all day long? Is it possible? Without machine help, it cannot be easy.

Medical apps capable of helping by protection from spreading COVID-19 already exist. How does it work? The application analyzes the video stream from surveillance cameras and automatically detects faces without protecting equipment (masks/respirators, gloves, glasses, medical gown) with 95% accuracy, recognizes them, and reports to the controller. Additionally, the app can track the body temperature of a person. If the temperature is higher than normal the solution alarms medical staff. The keeping of safe distance can be tracked as well.

Related: How is AI used in healthcare: Top Cases

The solution helps to control persons who violate the rules and track all contacts and persons in the risk group. Such solutions can be used for controlling not only the medical service organizations but in other institutions, helping them to improve safety in the workplace as well.

More articles

prev
next

More articles

Leave us a message

You’re in a good company: