Early prediction of down syndrome using deep transfer learning-based approaches
Nirmit Patel, Tarang Ghetia, Devraj Jhala, Shubh Kapadia, Yogesh Kumar · 2024
Down Syndrome, a genetic disorder caused due to an extra copy of chromosome 21. With around 1 in 700 newborn babies being affected from it, it remains one of the most prevalent chromosomal disease. Children with down syndrome typically have physical and intellectual disabilities, but with early intervention and support, they can live long and fulfilling lives. This research investigates the use of Convolutional Neural Networks (CNNs) and transfer learning for early detection of Down Syndrome. The dataset used in this study is taken from Kaggle and consists of about 3000 facial images of children with and without Down Syndrome. The highest accuracy achieved in this study is over 95% and further shows the efficacy of transfer learning for more reliable detection of down syndrome.