Detection of Anaemia using Image Processing

Abstract

Anaemia is a widespread hematological condition characterized by a reduction in the number of red blood cells (RBCs) or a decrease in haemoglobin concentration, which leads to insufficient oxygen transport to body tissues. Early and accessible screening is particularly important in rural and resource-limited regions where conventional laboratory infrastructure such as automated haematology analysers is unavailable. This paper presents an automated image-processing-based system for detecting anaemia from peripheral blood smear images and a simple graphical user interface (GUI) to facilitate practical clinical use. High-resolution microscopic images acquired at 400× magnification are processed using a two-stage preprocessing approach consisting of gamma correction for contrast enhancement and Wiener filtering for noise suppression. Red blood cells are segmented and detected using the Circular Hough Transform (CHT). The total number of detected RBCs within a field of view is calculated and compared with a clinically motivated threshold of 500 cells per image. Images with RBC counts below this threshold are classified as anaemic, while higher counts are labelled as normal. A Python-based Tkinter GUI enables users to load images, visualize detected cells, and obtain diagnostic results in a single step. The proposed system offers a low-cost, portable, and user-friendly solution suitable for rural health centres, mobile diagnostic units, and telemedicine services. The framework can be further extended with machine learning methods and advanced feature analysis for improved diagnostic accuracy and classification of anaemia subtypes.