Computer Aided Diagnosis of Thalassaemias: An Overview

The distribution of thalassaemias is global, with particular incidence in areas affected by malaria as the Mediterranean area (Italy, Greece, Turkey, Cyprus) and in southeast Asia (India, Vietnam, Cambodia). Screening of the heterozygous population is fundamental for keeping thalassaemic pathology diffusion under control. Thalassaemia recognition is based on a first-level analysis performed with haemochromocytometric data and a second-level examination (HbA2 quantification, globin chain synthesis, and genetic analysis). Many of the latter techniques are finalised to a secure diagnosis of the genetic defect and they are time-consuming and expensive, therefore it would be important to have a Computer Aided Diagnosis (CAD) support based mainly on the haemochromocytometric data and on the simple HbA2quantification. Such classification through CAD systems would contribute to the selection of cases that need further examination and could be helpful in laboratory quality control. Several automated expert systems have been proposed to detect thalassaemias. There are different types of α-thalassaemia resulting from different gene mutations and their distribution is dissimilar in different geographic areas. The CAD systems are often optimized on the α-Thalassaemia and β-Thalassaemia types of the area where the software is built. In this work an overview of some automated systems used for supporting the diagnosis of different Thalassaemia traits is presented.