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

How technology can 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 risk of spreading the infection. This way fewer people will be exposed to a potential infection. Additionally, this system supports and relieves doctors, who would otherwise have to go through this process manually.

 

It is the classical example of the maintenance strategy that can let machines help you to avoid danger. But can machines 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 common 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 that they are not always affordable. The third 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 the chest x-ray and CT 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. In the current public health emergency, the low sensitivity of RT-PCR implies that a large number of COVID-19 patients can not be identified quickly enough and may not receive appropriate treatment or isolation.

“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.

Can it help us to improve the process of diagnosis?

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

Detector for COVID-19 and other lung diseases is a new machine learning solution that can help to improve the diagnostic process. With this solution, doctors can recognize a lot of diseases very quickly. Detector for COVID-19 and other lung diseases is an application that can analyze x-rays and CT/MRT images in png, jpeg or DICOM files format. The solution helps 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.  

With a detector for COVID-19 and other lung diseases, doctors can make the process of diagnosis easier, faster, and more precise. The solution offers the doctor better visibility of infected areas because they are highlighted. Detector helps to identify a particular disease more quickly suggesting diagnoses from the database. It is an important assistant in the situation 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 in a time of pandemic means to enter a high risks area. Not all countries are reporting COVID-19’s impact on their health care workers. But even those statistics we have, are very alarming. In the USA it was reported that infected care workers make up at least from 16% to 28% of all cases, in Spain – 14.4%, in Italy and Ukraine – 20%. It shows clearly that we need to do more to reduce risks to medical personnel. Your “Stop Corona” app would definitely 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: for patients and for doctors?

Acting preventive

Theoretically, everybody knows about the necessity of wearing a medical mask or respirator, gloves and eyeglasses, washing hands, and staying on 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 can be difficult.

Medical PPE detector is a machine learning solution that helps to protect medical, and other institutions from spreading COVID-19. 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 Medical PPE detector 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.

The solution helps not only to control persons who violate the rules but also to track all contacts and persons in the risk group. Using the Medical PPE detector can significantly decrease cases of infected patients and doctors. 

The Medical PPE detector can be used for controlling not only for the medical service organizations but in other institutions helping them to improve safety in the workplace as well.

Need more details or want to try?

Thank you for reading! 

Feel free to rate the article here: