Our team follows the news on AI and software developments related to COVID-19, so we thought we would share some of the latest with the readers of our blog.

AI research in Cambridge to help guide COVID-19 diagnosis and prognosis for Addenbrooke’s patients

Artificial intelligence has been applied to guide diagnosis and prognosis of COVID-19 for Addenbrooke’s patients. The AI helps to diagnose patients with COVID-19 before they are actually subjected to actual tests. In so doing, it gives the medical practitioners an opportunity to gain more understanding of their prognosis hence facilitating guided treatment. The system involves an algorithm with an ability to collect and analyze images from National COVID-19 Chest Imaging Database (NCCID). NCCID database comprises of 40,000 X-rays, MRIs and CT scans collected from over 10,000 patients from different areas within the UK. The AI system therefore has the ability to track patterns and establish trends observed from the images. The main feature of interest in the scans are the visual signatures specific to the COVID-19 virus through which diagnosis will be made and a prognosis be predicted from. By being able to predict a prognosis, any appropriate interventions can be implemented early enough to reduce chances of developing complications when the virus has affected the body. By early prognosis the patient is able to get medication and oxygen early enough before the critical stage of infection kicks in. This technology also helps medical facilities to predict the number of patients that may need ICU services hence make early preparations such expansion of ICU capacity.

As SARS-CoV-2 Mutates, AI Algorithms Try to Keep Pace

Artificial intelligence has proved to be a remedy toward the continuing springing up of Coronavirus variants. Machine learning algorithms can effectively be used to monitor the mutation sequence of COVID-19 virus. Through an AI system it is possible to study and isolate the various mutations that would evade immune systems when a body is vaccinated or an instant where a person had earlier been infected. Through the AI system, the Scientists will predict the sequences of mutation and alter the vaccines accordingly to prevent the virus from escaping the immunity gained from vaccination. The incorporation of the AI algorithm in the vaccine development process would eliminate the incidences of viral escape by ensuring the scientists developing the vaccines keep up with the mutations. AI helps the researchers to identify the various segments and genetic codes of the virus including how the mutations would affect the physical structure of the virus. The AI model is trained to detect changes in the genetic sequences of SARS-CoV-2 spike protein, which forms the different forms of coronaviruses. The AI model is expected, in the long run, to be helpful in predicting future mutations in the COVID-19 virus which will therefore provide an angle through which researchers can develop treatment.

AI helps to prevent suicides during COVID-19 epidemic

Yang Bingxiang, a consultant and a volunteer for the Tree Hole project applied AI to monitor suicidal comments of people on online platforms. The project uses Artificial intelligence to collect and analyze the comments of netizens. It thereafter provide, reports on netizens whose comments appear to be suicidal. When a suicidal comment is detected, the project has a connected network of volunteers who locate the individual and offer them counselling and other forms of therapeutic assistance. Bingxiang’s project has been very instrumental having assisted 11,715 people at high risk of committing suicide and prevented 3,629 suicides. During the COVID-19 pandemic, the project through the AI robot uses semantic analysis to sort out content that is being posted and identifying possible suicidal risks in the process. The artificial intelligence robot provides the necessary information and the channels for intervention for people facing depression. The project guarantees a 24-hour real-time monitoring to comfort netizens who are in need of psychological help during the COVID-19 pandemic. In 2020, being the year that was most affected by the Covid-19 global pandemic, the project conducted 14 training seminars and trained over 800 volunteers to help in racing and helping victims of depression.

AI database used to improve treatment of UK COVID-19 patients

NHSX developed an AI database that is used to enhance treatment of COVID-19 patients in the UK. It is national imaging database that is used to identify variations and patterns in chest images of patients manifesting symptoms of COVID-19 in the UK. The AI database helps in the diagnosis of patients with COVID-19 before they are actually subjected to actual tests. NCCID database comprises of 40,000 X-rays, MRIs and CT scans collected from over 10,000 patients from different regions within the UK. In so doing, it give the medical practitioners and opportunity to gain more understanding of their prognosis hence facilitating guided treatment. The system involves an algorithm with an ability to collect and analyze images from National COVID-19 Chest Imaging Database (NCCID). The AI system therefore have the ability to track patterns and establish trends observed from the images. The main feature of interest in the scans are the visual signatures specific to the COVID-19 virus through which diagnosis will be made and a prognosis be predicted from. By being able to predict a prognosis, any appropriate interventions can be implemented to reduce chances of developing complications when the virus has affected the body. By early prognosis the patient is able to get medication and oxygen early enough before the critical stage of infection kicks in. This technology also helps medical facilities to predict the number of patients that may need ICU services hence make early preparations such expansion of ICU capacity.

We Helped Greece Build an AI System to Make COVID-19 Testing More Efficient. Here’s What We Learned

Two scientist created an AI system named Eva the enabled Greece to effectively test COVID-19. The AI system enables the Greek healthcare ministry to determine which foreign visitors to test and which not to be subjected to testing. Demographic information of visitors including the nations they recently visited are recorded and the AI will use machine algorithm to process and hence predict the risk of the visitor having been infected with COVID-19; this determined whether the visitor should be subjected to testing at the border or not. Those who are predict to present low degree of risk were released and allowed passage into the country. Through Eva, Greece was not only able to save time that would have been wasted on testing the thousands of visitors arriving but also was able to save up on the testing resources. The Eva AI system also played a significant role of producing more timely COVID-19 prevalence information. Eva system enables target testing in Greece which in the process made it easy to facilitate testing among the visitors entering the country. By using Eva AI software properly the effectiveness of testing operations improved hence reducing the spread of COVID-19 virus.