Regardless if you think COVID-19 pandemic is over, we need to be prepared to tackle upcoming global health challenges. Our resources will always be limited and treatments will always need the most effective administration. According to scientists from the University of Waterloo, this is where artificial intelligence (AI) could be extremely useful.
Scientists at the University of Waterloo have developed a new sophisticated computer software, which is able to identify the most vulnerable patients by learning from previous COVID-19 patient cases with known clinical outcomes. This information could help healthcare institutions to assign more care, medicine and general attention to those patients who could benefit from it the most.
This is not the first time scientists are trying to utilise the strong parts of AI, such as its ability to analyse huge pools of data quickly and efficiently, during this COVID-19 pandemic. But this new software is a part of an open-source project called COVID-Net.
This new AI-based model analyses takes into account such information as serum ferritin levels, use of therapeutic heparin, heart rate and blood pressure. It then compares those biological markers with patterns it learned by looking at hundreds of similar cases.
This system is based on the so-called explainable AI technology, which allows the model to explain which indicators it relied on for its predictions. This removes some doubts from the system and helps health professionals make the final decisions together with the software if needed.
Alexander Wong, one of the authors of the study, said: “AI models that provide not just predictions but the rationale behind the predictions can greatly improve trust and widespread adoption to support clinicians in their decision-making processes along the entire clinical workflow. That's the power of explainable AI.”
The reason why the world needs such systems is because COVID-19 pandemic is not over. Hospitals are already extremely overburdened by this same pandemic. Resources are limited and doctors need to make quick and rational decisions about which patients should be prioritised. This AI-based model predicts which COVID-19 patients are most at risk of dying or developing kidney injuries during hospitalisation. Ultimately, it could help prevent many of the adverse outcomes.
Even if COVID-19 pandemic is going to be over soon, we need these systems in place to be prepared for what the future might have in store for us. There will be other pandemics and we will be overwhelmed again. Our resources are limited and AI can help utilise them in the most efficient way possible. It is good for doctors as well, because it makes the decision making process easier and allows focusing on treating patients instead.
Source: University of Waterloo