Computer Vision News - February 2019

to develop AI algorithms that can enable better healthcare. At the same time, I also want to study and to formulate important challenges in real world applications like healthcare that can then help motivate useful research directions in AI. One of my main focuses has been creating AI assisted healthcare spaces where we have computer vision that’s infused in physical spaces like hospitals in order to understand what is happening in an environment like a hospital at every moment in time and use that to assist in delivering quality care. Activity recognition is a huge part of that. It’s something that I have worked on quite a lot in computer vision. In general, there are many challenges that you realize you need to solve when you want to build something like a smart hospital in the real world. What will be the major benefit of your work in health care? I understand that it might help nurses have more time to treat patients or improve the quality of the care. I think there are a number of things. At the first level, if you just have something like an automated system that can recognize and record the actual care activities and what’s happening at every moment in time, all of a sudden you have a lot of data to actually understand what’s happening, study correlations with outcomes, and develop best care practices. This is a kind of data that hasn’t previously been available. I think that’s actually at the first level of what I’m really excited that we can use this data and computer vision for. At the second level, as you said, once we’re able to recognize what is happening, we can use that feedback to assist clinicians and nurses in real time. That means being able to document automatically what is happening, allowing clinicians and nurses to focus more on taking care of patients. We can also help them know what steps have happened so far, whether these are the intended steps of different types of care protocols. Pauline Luc 19 Computer Vision News Serena Yeung Women in Science “I think dissemination and communication are also an important part of research!” photo: Dan Taylor/Heisenberg Media