UTFacultiesTNWEventsLecture: KayalOME, a deep learning algorithm for identifying disease causing mutations in Inherited Retinal Disorders

Lecture: KayalOME, a deep learning algorithm for identifying disease causing mutations in Inherited Retinal Disorders

Join the guest lecture given by Prakash Ekambaranellore, Founder and Managing Director, WhatsNU Genomics Pte Ltd.

Abstract: Inherited retinal disorders (IRDs) are a group of clinically and genetically heterogeneous conditions with a prevalence of 1:3500 worldwide and contribute to 60% of blindness in the infants. This includes non-syndromic Mendelian diseases such as congenital stationary night blindness (CSNB), Leber Congenital Amaurosis (LCA), Retinitis pigmentosa (RP), Cone/cone rod dystrophy (CD-CORD), Stargardt's disease (STGD) and achromatopsia. Certain syndromes also include these monogenic diseases as one of the clinical entities – Bardet Biedl syndrome (BBS), Joubert syndrome, Senior- Loken syndrome, Usher syndrome and Thiamine responsive megaloblastic anaemia. In recent years, molecular diagnosis in IRDs has become more feasible, leading to the discovery of diseases causing variants in more than 260 genes. It is estimated that NGS has made it possible to identify causative genes in two-thirds of patients with IRDs and 85% of early onset IRDs. Understanding the molecular mechanism has helped in designing treatment strategies for a few IRDs. We have developed an AI based screening tool known as KayalOME to identify mutations causing in IRD patients from retinal images obtained using fundus, OCT (Optical Coherence Tomography) and ERG. Using deep convolutional neural networks (DCNN), KayalOME will facilitate segmentation, feature extraction, and classification of the deidentified retrospective retinal image data to predict the causative genetic mutations in IRDs. Application of KayalOME in risk prediction for late onset IRDs would improve the quality of life for patients through early medical intervention, prior to the onset of the disease. Predicting the likelihood of a genetic abnormality in the IRD patients would also assist the genetic testing process by a clinical laboratory by augmenting the clinical decision management process.

Prakash Ekambaranellore obtained his PhD at the Sri Venkateswara University in Tirupati, India. He has over six years of post-doctoral training from various reputed research centers in Taiwan, such as Academia Sinica, National Taiwan University Hospital, and the National Health Research Institute. Dr Ekambaranellore is pursuing research towards developing disruptive diagnostic tools using deep learning and artificial intelligence, to help clinical management of inherited retinal disorders.