Indonesian Journal of Electrical Engineering and Computer Science Vol. 31, No. 1, July 2023, pp. 359~368 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v31.i1.pp359-368  359 Journal homepage: http://ijeecs.iaescore.com Age and gender classification with bone images using deep learning algorithms Sathyavathi Sundarasamy 1 , Baskaran Kuttuva Rajendran 2 1 Department of Information Technology, Kumaraguru College of Technology, Coimbatore, India 2 Department of Computer Science and Engineering, Kumaraguru College of Technology, Coimbatore, India Article Info ABSTRACT Article history: Received Oct 7, 2022 Revised Mar 8, 2023 Accepted Mar 12, 2023 In paediatrics, bone age is a crucial indicator of how a child's skeleton is developing. They have had great success ever since the creation of deep learning (DL)-based bone age prediction tools. Deep features learning, however, has a significant computing overhead problem. Deep convolution layers are used in this technique to learn representative features in the small yet useful regions that are extracted for feature learning. This work suggests using an extreme learning machine algorithm as the fundamental architecture in the final bone age assessment study to realise the rapid computation speed and feature interaction. The viability and efficacy of the suggested strategy have been verified by experiments using data that is openly accessible. The suggested model is explicitly trained using a cutting-edge end-to-end learning architecture employing bone scans to extract the most discriminative patches from the original high-resolution image. The bone picture is the foundation of the procedure. Our main objective is to categorize individuals by age using convolution neural network (CNN) classification models, such as the Xception and Mobile Net models. As a result, we have achieved results that are 90% and 94% accurate in classifying people by age using CNN models. Keywords: Bone age assessment Chronological age Classification Convolutional network Deep learning This is an open access article under the CC BY-SA license. Corresponding Author: Sathyavathi Sundarasamy Department of Information Technology, Kumaraguru College of Technology Coimbatore 641049, India Email: sathyavathi.s.it@kct.ac.in 1. INTRODUCTION A person's skeletal and biological maturity can be determined by their bone age. Unlike chronological age, which is established by a person's birth date, this age is different. Paediatricians used to compare a child's bone age to their chronological age to diagnose conditions that make kids to shape their stature. These measures are helpful to evaluate how well these diseases' treatments are working. Formula have also been developed for calculating a child's final adult height from figures for the child's bone age in healthy, normal children. The estimation of bone age is used to determine chronological age for the kids whose birth records are not available. A major issue in our region of the world is the lack of birth records. The hand and wrist bones' ossification pattern is frequently predictable and age-specific. By comparing the maturity of the hand and wrist bones, the standard age associated with normal ageing has been determined. The bone age study can assess how quickly or slowly a child's skeleton is developing, which can assist doctors in identifying diseases that either slow down or accelerate up physical development. Typically, doctors or paediatric endocrinologists will request this test. Identifying the age of death, birth date, year of death, and gender of unidentified human remains in the context of a criminal investigation might help detectives make the right identification out of a possible match.