The inspiration of the design is that one of the people in our team, Yutao Hu, caught a cold last month. He went to hospital. He waited 30 minus in Purdue Pharmacy Center to wait for the medicine. He think the efficiency of the medical system is too low. Our design is a vending machine that can help enhance the efficiency and accuracy of the medicine receiving process. The design applied the facial recognition technology to recognize patient quickly. The average reacting time is less than 1 minutes each person.
What it does
We will build a folder for each of the patients in the hospital server. When a patient comes to our hospital, doctor would make a database for new incoming patient, or load previous database for previous patient. To simulate the process of interrogation，a GUI is created. The doctors can input the basic personal information and photo easily using the GUI. The prescription will be stored together with the information of the patients.
The medicine distribution machine has a camera that used to take pictures of patients. Then the data is sent to the server of the system and the return value will be used to drive the machine to distributing the corresponding medicine to the patients. The patients can get their medicine by using our machine by following the instructions on another GUI based on Raspberry and LCD display.
How I built it
The first step is to step up the machine. Raspberry pi and Arduino boards are used as the controller in the machine. The raspberry pi take charge of controlling the machine and communicating with other device. LCD display is used to provide information to patients and receive instruction. Arduino boards are applied to drive the servo. The GUI of the system is created by Kivy library. The connection of the net work is created by the ssh in the local net work.
Challenges I ran into
The biggest challenge I met was that how to use hardware interrupts with software. The communication between the software and hardware is hard. The problems with hardware also gives me a lot of trouble. The torque of my servo is not stability. It is hard for me to vending the machine with that. I used the Raspberry pi, Arduino and some other sensors. The supply voltages are different. It is hard to implement them together.
Accomplishments that I'm proud of
We successfully made the design in the given time. We still up all nights in the two days to finish the design. I finished hardware part which I have never tried before.
What I learned
It is my first time to do a project on Microprocessor and I never use python before. I got some experience in digital design and I learnt a little bit python. Besides that, I learnt how to use the ssh to setup local network between different computers.
What's next for Express Medication
Firstly, other method should be used to make the design more accurate. Now, we use higher threshold value for the face detect, but one method is not enough for biomedical application in which no mistake is allowed.
Second, we need to use better data analysis system. The system should contains more information about the patients,
which have a high potential in salvage. In the emergence situation, the doctor can save more life if they do not have to spend time on inspection. Third, the machine parts of the design should be improved. None of us has experience in ME. We need find better parts for our design.