AI Vision vs coronavirus: using deep learning augmented computer vision to help amid the current health crisis.
The last few weeks have been increasingly trying to the health systems around the world as the novel coronavirus spread has been declared a pandemic by the WHO officials [1]. The sudden surge in patients needing hospitalisation and assisted respiration has put a tremendous strain on doctors, nurses and other medical staff. So is there any way we could use AI-aided computer vision to help in the current and future situations like this?
Public screening
Considering many people infected with the virus develop a high fever, airports and points of transit around the world have implemented manual temperature checks. Chinese tech giant, Baidu, has another idea. Using deep learning algorithms they have developed a tool for mass screening with thermographic cameras and it is currently being used at Beijing’s Qinghe Railway Station where it can examine up to 200 people at a time. All of this without disturbing natural passenger flow.
Hand washing compliance
Currently, one of the most important aspects of COVID-19 spread prevention is proper hand washing and disinfection. This could be helped by AI too. Researchers at Stanford PAC have shown a successful attempt at creating a computer vision system that can monitor if people walking through the entry use the hand gel dispenser or not. [3] Another study from Leibniz University Hannover proved that you can use a simple camera and a deep learning model to recognise specific hand gestures done while properly washing your hands and tell whether you did the whole routine properly. [4]
Patient monitoring
The last two examples were about preventive measures but what about developed cases?
As the hospitals and Intensive Care Units become more and more overcrowded, there is going to be a need for automating some of the tasks that have to be done by overworked doctors, nurses and other medical professionals. Well, there is still plenty to do. For example, there was a paper published in Nature last year about Intelligent ICU for Autonomous Patient Monitoring showing several data categories that can be gathered by a Computer Vision system such as facial expression, posture and movement analysis and sleep patterns, which can be later used to assess current patient status and his/her treatment. [5]
Another company, Oxehealth, provides cameras with software that can detect basic vital signs from a patient such as pulse rate, breathing rate or changes in activity without the need of a patient’s supervision 24/7. [6]
Conclusions
Apart from analysing available electronic health records, Artificial Intelligence systems can be utilised in both preventive measures and clinical care. As the situation gets more and more severe we should strive for further automation of health-related tasks in order to give medical professionals more time to do the most important part of their job: saving lives.
If you want to read more about AI, check out our site at Tooploox: https://www.tooploox.com/ai
References
[2] https://www.technologyreview.com/s/615342/how-baidu-is-bringing-ai-to-the-fight-against-coronavirus/
[3] https://aicare.stanford.edu/projects/hand_hygiene/
[4] Dietz, A., Pösch, A., Reithmeier, E. (2018): Hand hygiene monitoring based on segmentation of interacting hands with convolutional networks, Proc. SPIE 10579, Medical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications, 1057914 https://doi.org/10.1117/12.2294047
[5] Davoudi A., Malhotra K. R., Shickel B., Siegel S., Williams S., Ruppert M., Bihorac E., Ozrazgat-Baslanti T.,Tighe P. J., Bihorac A., Rashidi P (2018): The Intelligent ICU Pilot Study: Using Artificial Intelligence Technology for Autonomous Patient Monitoring. https://arxiv.org/pdf/1804.10201.pdf
