METHODOLOGIES AND APPLICATION A metaheuristic segmentation framework for detection of retinal disorders from fundus images using a hybrid ant colony optimization D. Devarajan 1 • S. M. Ramesh 1 • B. Gomathy 2 Ó Springer-Verlag GmbH Germany, part of Springer Nature 2020 Abstract Imaging modalities play a major role in early detection and diagnosis of various medical conditions related to the patient. Retinal image segmentation has been taken up for investigation in this research paper to efficiently detect the presence of eye disorder which could be indicators of major onset of conditions like hypertension, cataracts, diabetic retinopathy, age- related macular disorders, etc. A machine learning method for classification of given pixels in the search space into regions containing blood vessels and those that do not contain blood vessels is implemented using a three-stage neural classifier in this paper. Prior to classification, an optimization algorithm namely ant colony optimization derived from nature-inspired phenomena is used to provide an optimal feature vector set to set high standards for the neural network based classification approach. The novelty and merits of the paper lie in back tracing of the segmentation process in which optimization is done first on the preprocessed features followed by classification for segmented output on the optimized features. This results in elimination of redundant feature vectors which tend to occupy much memory as well increase the computational overhead on the process. The entire implemented system is automated by the machine learning process and tested on 30 samples, 15 each on DRIVE and STARE databases. Classification rates of nearly 98% on an average scenario have been achieved for segmentation and 96.5% for abnormality detection. The performances have been compared against Bayesian set models and standalone ANN models. Keywords Retinal image segmentation  Fundus images  DRIVE  STARE  Ant colony optimization  Neural network classifier  Classification accuracy 1 Introduction Clinical imaging has witnessed a great revolution in recent times especially with the advent of state-of-the-art imaging systems and modalities. Clinical imaging has attracted a wide range of research interests due to their immense potential in early detection of medical conditions. This early detection helps in appropriate treatment through diagnosis thereby resulting in near elimination of the medical condition. An interesting feature to be observed in clinical imaging is that multiple symptoms and conditions could be decided upon from the same imaging modality. For example, retinal imaging could help in detection of cataract, blurriness of eyes, etc. Additionally, it also plays to be a vital indicator of certain blood-related conditions such as diabetes and hypertension. Hence, the role of medical imaging has been dominant in almost all health- care sectors resulting in drastic reduction in the mortality rate due to late detections. The issue of retinal imaging for detection of disorders related to retina of the eye has been taken as the primary issue of investigation in this research paper. The process of retinal image processing for detec- tion of early disorders is quite a challenging task with several constraints related to extraction of the retina from the background. It could be inferred on the whole that the Communicated by V. Loia. & D. Devarajan devarajan@egspec.org S. M. Ramesh drsmramesh@egspec.org B. Gomathy gomramesh@gmail.com 1 E.G.S Pillay College of Engineering and Technology, Nagapattinam, Tamilnadu, India 2 Bannari Amman Institute of Technology, Erode, Tamilnadu, India 123 Soft Computing https://doi.org/10.1007/s00500-020-04753-7