View Proposal

---

Proposer

Rob Stewart

Title

Hardware acceleration of skin lesion classification

Goal

Classify skin lesion images in real-time with a smart imaging device

Description

Using smart imaging devices to diagnose skin lesions could automate early clinical diagnosis in developing countries, and provide accessible dermatological care in remote areas with limited healthcare infrastructure or where access to health services is expensive. The main challenges for such imaging technology for skin lesion classification are (1) using a deep learning model small enough to fit on a compact handheld device, (2) fast and sufficiently accurate classification and (3) operating without reliance on internet connectivity. With embedded processing platforms there is a compute spectrum, from general purpose embedded CPUs, to GPU co-processors, all the way to truly custom hardware logic e.g. with FPGAs. Although challenging to program, the main advance of configurable FPGA hardware is you often get both high throughput and low power use, rather than a compromise between the two. This goal of this project is to develop a skin lesion classifier deep learning model for FPGAs. It will target the Pynq FPGA platform, using AMD's Brevitas quantisation framework for compressing the model and the AMD's FINN framework for deploying to the Pynq architecture. The classification accuracy, inference time and energy use of the developed system will be compared with a similar Raspberry Pi based skin lesion classifier. This would be in collaboration with PhD student Tess Watt, whose research is closely aligned to this proposed project, and Dr Christos Chrysoulas.

Resources

AMD Pynq development board.

Background

Url

Difficulty Level

Moderate

Ethical Approval

None

Number Of Students

1

Supervisor

Rob Stewart

Keywords

Degrees

Bachelor of Science in Computer Science
Bachelor of Science in Computer Systems
Master of Engineering in Software Engineering
Master of Science in Artificial Intelligence
Master of Science in Artificial Intelligence with SMI
Master of Science in Computer Science for Cyber Security
Master of Science in Computer Systems Management
Master of Science in Computing (2 Years)
Master of Science in Data Science
Master of Science in Human Robot Interaction
Master of Science in Information Technology (Software Systems)
Master of Science in Network Security
Master of Science in Robotics
Master of Science in Software Engineering
Bachelor of Science in Computing Science
Bachelor of Engineering in Robotics
Bachelor of Science in Computer Science (Cyber Security)
Master of Science in Robotics with Industrial Application
Bachelor of Science in Statistical Data Science
BSc Data Sciences