Thalassaemia is one of the most common single-gene disorder in which the production of haemoglobin is impaired. this autosomal recessive disorder is highly prevalent in indian populations accounting to ~10% of the world’s thalassaemia carriers. the beta-thalassaemia carrier state resulting from heterozygous mutation in beta globin gene, is clinically asymptomatic and thus remain undiagnosed. at present, the gold standard method that is used for carrier detection in hospitals requires expensive instruments, skilled manpower and time, thus, making it difficult to be used as an onsite method. a rapid, portable and automated technology for thalassaemia carrier screening is hence of significant importance. the present study has conclusively proven that distinct patterns are observed on drying of whole blood droplets for carrier and normal samples. length of the radial cracks is significantly shorter for carrier samples as compared to normal ones. a sample whose average crack length is less than 800 microns can be classified as a carrier sample, whereas the ones with larger radial cracks are considered as normal. identifying the carrier samples by this image analysis technique gave zero false negative results. these patterns can be further utilized to create a databank for automated classification of carrier samples, by employing appropriate techniques from computational pattern recognition and deep learning. proposed method will examine the images of dried blood drops, extract its distinctive features and categorize as normal or carrier samples by comparing it with reference images stored in databank. this automated process will remarkably increase the number of total population screened for thalassaemia per year in the country and will reduce the burden in the state run advanced health facilities along with a significant reduction in costs incurred for screening of each sample.