Maria Tariq has done BS in Computer Science in 2016 from NUCES-FAST, Lahore and MS in Computer Science from NUCES-FAST, Islamabad in 2019. She became part of a biomedical research lab and started research in the field of image pre-processing and deep learning that is based on applied mathematics and data science.

Maria is currently pursuing her P.h.D with the topic of “Automatic Detection of Diabetic Retinopathy in Eye Fundus Images” from the Centre of Computational Sciences and Mathematical Modeling in Coventry University, The United Kingdom.  Her research interests include Deep Learning, Machine Learning, Image Pre-processing, Transfer Learning, Deep Reinforcement Learning .etc. 

Under the supervision of Vasile Palade and YingLiang Ma, my research aims to perform multiclass classification of diabetic retinopathy through medical images using transfer learning models. On top of that, it also deals with imbalanced classification and synthetic image generation.

I have applied different image preprocessing and data augmentation techniques with pre-trained models to do multiclass classification on the publicly available dataset, which has significant challenges of image noise, imbalanced classes, and fault annotations.

To resolve the imbalanced nature of data, I have combined reinforcement learning models and deep learning techniques. Furthermore, I am currently working on generating synthetic fundus images using Generative Adversarial Networks (GANs).