In hospitals, nurses are required to count the instruments (e.g. surgical knives, forceps) lest they be left inside a patient’s body/not enough or wrong instruments during the surgery.
This process is called surgical counting, and the Central Sterile Services Department of the Hospital Authority currently employs manual counting.
The department frequently encounters the following problems:
the number or order of instruments in a tray is wrong,
certain instruments are difficult to count (such as overlapping scissors), and
foreign objects (such as hair clips or stationery)
Object Recognition is a computer vision technique used to identify and locate objects in a photo. In this project, we aim to apply computer vision techniques to help medical professionals count the instruments correctly and rapidly.
For this project, you will need to
plan and execute the object detection workflow,
investigate which algorithms are best suited for these scenarios, and
create a usable user interface for nurses to work on.
| Previous Work
|