Mini-μScope for healthcare in resource-limited area

Elevator pitch

Contaminated drinking water is estimated to cause 485, 000 diarrheal deaths worldwide each year. According to WHO, globally, at least 2 billion people use a drinking water source contaminated with faeces, 1 in 3 people do not have access to clean drinking water. A microscope is the most intuitive tool for observing microorganisms. However, laboratory microscopes are bulky and expensive, not readily accessible in underdeveloped countries. This project aims to develop an AI-enhanced low-cost and portable microscopy system, Mini-μScope, to monitor water quality, disease diagnosis, like Phytoplankton detection and red blood cell recognition.Mini-μScope is user friendly and can be easily operated by lay users. It uses a simple webcam as an objective lens and has an embedded AI module to realize real-time analysis such as parasite recognition in drinking water or sickle red blood cells. Mini-μScope aims to safeguard healthcare conditions for living in underdeveloped countries.

What is the project about?

This project is about developing an intelligent portable and low-cost microscopy system (called Mini-μScope) in rural areas where access to laboratory microscopes is impossible.Mini-μScope can analyze water quality and blood smear in real-time.

Why did you decide to make it?

The lack of environmental and medical monitoring devices in underdeveloped countries is a well-known problem. It is desirable to realize low-cost devices for fast water quality detection or disease diagnosis. A microscope is a fundamental tool. However, traditional microscopes are bulky and expensive and can only be operated by experienced users, not suitable for on-site or point-of-care applications.

To safeguard healthcare conditions in resource-limited areas, we propose the Mini-μScope project. Mini-μScope is low-cost, portable, and user-friendly. Users do not need to have programming or scientific backgrounds; even a child can use it easily.

How does it work? What are the main features? Please specify how you will use the Xilinx Kria KV260, VCK5000 or Big Data Analytics Card in your solution

The Kria KV260 module serves as a powerful microcomputer in Mini-μScope. We can achieve fast hardware prototyping using the full-stack tools of Vitis. Multi-purpose neural network architecture will be efficiently implemented on KV260 to perform real-time data analysis, such as recognizing water parasites, cell types, or immunofluorescence diseases. Besides, we will design a compact mechanical part that can assemble an optical imaging system. By using high-performance peripherals on KV260, the optical system can be efficiently coupled with the hardware computing engine. This system is very compact, providing a more economical solution and the potential to replace bulky microscope systems. A user-friendly software interface will be designed for general users without a technical background.