Image Processing for Microscopic Images

Former students
Annika Wong, T. Suwannaphong
Collaborators
T. Chalidabhongse, K. Naruenatthanaset (Computer Engineering), D. Palasuwan, A. Palasuwan (Clinical Microscopy), R. Chancharoen, G. Phanomchoeng (Mechanical Engineering), Chulalongkorn University, Thailand

An Automated System for Parasitic Worm Eggs Detection and Classification on a Motorised Stage Microscope

Intestinal parasitic infections remain among the leading causes of morbidity worldwide, especially in tropical and sub-tropical areas with more temperate climates. In this context, we aim to develop a novel cost-effective diagnosis system that automates detection and identification of intestinal parasite eggs from faecal smear samples, that can be used by non-experts. This is a highly interdisciplinary project necessitating this collaborative research proposal.

The hardware part, comprising programmable X-Y-Z motorised staging microscopic system, will be developed by researchers from Chulalongkorn University, Mechanical Engineering and Computer Engineering Departments, and the Solimac Automation Co.,Ltd., which is the leading machine vision and automation technology distributor and integrator in Thailand. The algorithmic component, which will automate detection and classification of the parasitic worm eggs from the microscopic images, will be collaboratively performed by researchers from CU's Clinical Microscopy and Computer Engineering Departments, and researchers from Bristol Vision Institute, and Faculty of Life Sciences, University of Bristol who are the experts in medical imaging and machine learning. The system will be tested and validated by personnel at the Phon Phisai Hospital, a district hospital in northeast Thailand.

poster aloe

Posters

poster aloe


Publications
Datasets

Vision-based Abnormal Red Blood Cells (RBC) Detection and Classification

Identification of abnormalities in red blood cells (RBC) is key to diagnosing a range of medical conditions from anaemia to liver disease. Currently this is done manually, a time-consuming and subjective process. This project aims to automate this process utilising the advantages of machine learning to increase capacity and standardisation of cell abnormality detection and cell type classification. To classify multiple classes with deep learning, imbalance problems are common in medical imaging because normal samples are always higher than rare disease samples. This project presents new methods to segment and classify red blood cells from blood smear images, specifically to tackle cell overlapping and data imbalance problems.


Posters

poster annika poster kao


Publications
Dataset