Considering the rapid spread of Coronavirus, our project aims to limit direct interactions between the frontline medical staff and suspected patients. We will use a robot to reduce the manual work for nursing staff, which can also reduce the risk of cross-infection. This will accelerate the overall medical screen procedure in hospitals and possibly improve it.

What it does

  • Check patient body temperature prior to entering the hospital and alerts the staff if the patient has a fever
  • Guide the patient to complete registration
  • Answer the patient’s questions and ask them to fill out the COVID-19 Health Declaration Form
  • Send patient’s data to doctors immediately and keep them informed before visit begins

How we built it

  • We developed our system with Android for frontend and Laravel framework for backend
  • We used Azure Facial Recognition API service to detect patients
  • We used another hardware to measure vital signs and then transfer the data to our robot via Bluetooth technology

Challenges we ran into

We need additional hardware to measure patient vital signs. It is also difficult to program the robot to do accurate speech recognition.

Accomplishments that we’re proud of

  • The system is now capable of processing the COVID-19 Health Declaration Form, and will soon be deployed to help the frontline medical staff at the National Cheng Kung University (NCKU) Hospital's quarantine station soon
  • The software for our robot was developed using official API approved by NCKU Hospital

What We Learned

We learned that new technologies, such as artificial intelligence (AI) and Internet of Things (IoT) can really be utilized in many aspects of our lives. We also learned that our profession can help others, especially in a pandemic.

What’s next for our project

We will keep updating the system with more functionality and improve existing ones. At the same time, we will gather feedback from the medical staff at the NCKU Hospital to actively improve the product’s usability.

+ 27 more
Share this project: