About me

I am Mokshagna Sai Teja Karanam studying PhD in computer science at University of Utah, Utah in School of Computing. Currently, I’m in Scientific Computing and Imaging Institute under the guidance of Prof. Shireen Elhabian.

Research Interests

With a focus on innovative research, Currently, I’m working on advanced Machine Learning/deep learning methods to analyze and quantify medical images. My research interests also include Computer Vision, Image Processing and Probabilistic Machine Learning. I’m interested in developing advanced algorithms and techniques in Medical Imaging through the utilization of deep learning. My focus also lies in Statistical Shape Modeling. Specifically I’m interested in the leveraging concepts of Self-Supervised Learning, MultiModal Learning, Adversarial Training, Domain Adaptation, Diffusion Models and the challenges posed by limited data and the intricacies of solving Regression tasks. The ultimate objective is to develop methodologies and algorithms capable of comprehending semantic information across various modalities and tasks. The idea of tackling intricate data challenges in the industry is exciting to me and inspires me to collaborate with organizations that can help me enhance my skills and knowledge.

Advisor: Prof. Shireen Elhabian

Collaborators: Tushar Kataria, Krithika Iyer

Recent News

-[Nov 2023 EduHPC-23: Workshop on Education for High Performance Computing] An NSF REU Site Based on Trust and Reproducibility of Intelligent Computation: Experience Report.

Author List: Mary Hall, Ganesh Gopalakrishnan, Harvey Dam, Artem Yadrov, Amirmohammed Tavakkoli, Johanna Cohoon, Sameeran Joshi, Aditya Bhaskara, Eric Eide, Jeff Phillips, Mu Zhang, Shireen Elhabian, Tushar Kataria, and Mokshagna Sai Teja Karanam.

-[July 2023] ADASSM: Adversarial Data Augmentation in Statistical Shape Models From Images., Accepted in ShapeMI 2023, MICCAI Workshops 2023.

Author List: Mokshagna Sai Teja Karanam, Tushar Kataria, Krithika Iyer, Shireen Elhabian.

-[Sep 2024] EfficientMorph: Parameter-Efficient Transformer-Based Architecture for 3D Image Registration, Accepted in WACV 2025, WACV 2025.

Author List: Abu Zahid Bin Aziz*, Mokshagna Sai Teja Karanam*, Tushar Kataria, Shireen Elhabian. (*Equal Contribution)

-[Feb 2025] Mesh2SSM++: A Probabilistic Framework for Unsupervised Learning of Statistical Shape Model of Anatomies from Surface Meshes, Submitted in Arxiv.

Author List: Krithika Iyer, Mokshagna Sai Teja Karanam, Shireen Elhabian

-[Feb 2025] MORPH-LER: Log-Euclidean Regularization for Population-Aware Image Registration, Submitted in MIDL 2025, MIDL 2025.

Author List: Mokshagna Sai Teja Karanam*, Krithika Iyer*, Sarang Joshi, Shireen Elhabian. (*Equal Contribution)