Computer Vision and Deep Learning Research Intern at Adventum

Adventum specializes in imaging as well as medical Imaging processing using cutting-edge technologies such as Advanced AI and Quantum computing. At Adventum, we were focused on developing disruptive technology to analyze and detect abnormalities of the retina at a very early stage, through the images provided by the Fundus and Optical Coherence Tomography (OCT) scan. • Implementing and training deep learning architectures for Semantic segmentation, Anomaly detection, Classification, and Auto-Encoders for Medical Images.

• Developed an Al-powered end-to-end pipeline to predict Glaucoma using fundus imaging.

• Took part in data collection, annotation, pre-processing, ML research, model development, and deployment.

• Dealt with classical Computer Vision techniques like Denoising, Segmenting, Registering, and Restoration.

• Worked on extracting 3D optical coherence tomography OCT and fundus scans from DICOM files for training models.

• Building and deploying trained machine learning models using Flask.

• ML tasks: Transfer Learning, Ensemble Learning, and Multi-modal learning.

Skills: Python (Programming Language) · Machine Learning · Flask · Deep Learning · Image Processing · TensorFlow · Pandas · Git