CV

Resume

Education

• B.S. in Computer Science and Engineering, B V Raju Institute of Technology, 2016-2020
• M.S. in Computer Science, University of Utah, 2022-2024
• Ph.D in Computer Science, University of Utah, 2024-2029 (expected)

Work experience

• Scientific Computing and Imaging Institute: Graduate Research Assistant - August 2022 - Current
  • Spearheading an end-to-end preprocessing pipeline for Craniosynostosis data, incorporating segmentation, registration, and bulk upload handling, improving efficiency by 20%.
  • Formulated a transformer-based architecture for unsupervised 3D medical image registration, utilizing high-resolution tokens reducing computational redundancy.
  • Achieved state-of-the-art performance on the OASIS dataset, using 16-27x fewer parameters.
  • Proposed an automated, computationally efficient, and on-the-fly adversarial data augmentation technique tailored for Statistical Shape Modeling and Regression Tasks which reduced approximately 60% training time.
  • Introduced a novel regularization technique utilizing contrastive loss. This technique aimed to generate more task and data-specific samples, leading to a substantial 10% performance enhancement across diverse modalities.
  • Designed ShapeWorks test systems for different modalities and deployed datasets in the shapeworks cloud.
  • Utilized a range of multi-task techniques, including segmentation, distance transforms, and depth estimation, to enhance shape modeling performance by 20%.
  • Conducted experiments with 3 multi-organ anatomies and pretraining strategies to optimize the performance of a single anatomy.
• Tata Consultancy Services: Systems Engineer - November 2020 - July 2022
  • Worked on an application project in AWS using Django Framework.
  • Developed application’s ORM layer and deployed an extensive suite of REST APIs.
  • Engineered web services and devised distinct lambda jobs, resulting in service optimization exceeding 25%.
  • Identified and rectified performance bottlenecks, resulting in a notable 50% enhancement in page load speed.
  • Executed over 15 stored procedures and performed operations within MySQL to improve the data retrieval and storage processes.
  • Enforced access controls for 1000-plus users, enhancing systems security and aligning data access privileges accurately.
  • Engaged in roadmap consultations, mentoring and guiding a team of 2
  • Collaborated with cross-functional teams to successfully execute a large-scale migration from Python 2.7 to Python 3.8.
• OpenText: Associate Software Engineer - September 2019 - September 2020
  • Implemented crucial features and user Access Control for the product TeamSite using AngularJS and ReactJS.
• OpenText: Engineering Intern - June 2019 - September 2020
  • Created a Report Generation Tool with Python, enabling real-time monitoring of usage and provides performance metrics for over 100 usecases.

Publications

ADASSM: Adversarial Data Augmentation in Statistical Shape Models From Images

Karanam, Mokshagna Sai Teja, Tushar Kataria, Krithika Iyer, and Shireen Y. Elhabian. "ADASSM: Adversarial Data Augmentation in Statistical Shape Models from Images." In International Workshop on Shape in Medical Imaging, pp. 90-104. Cham: Springer Nature Switzerland, 2023.

An NSF REU Site Based on Trust and Reproducibility of Intelligent Computation: Experience Report

Hall, Mary, Ganesh Gopalakrishnan, Eric Eide, Johanna Cohoon, Jeff Phillips, Mu Zhang, Shireen Elhabian et al. "An NSF REU Site Based on Trust and Reproducibility of Intelligent Computation: Experience Report." In Proceedings of the SC23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis, pp. 343-349. 2023.

EfficientMorph: Parameter-Efficient Transformer-Based Architecture for 3D Image Registration

Aziz, Abu Zahid Bin, Mokshagna Sai Teja Karanam, Tushar Kataria, and Shireen Y. Elhabian. "EfficientMorph: Parameter-Efficient Transformer-Based Architecture for 3D Image Registration." arXiv preprint arXiv:2403.11026 (2024).

Talks

Enabling data diversity: efficient automatic augmentation via regularized adversarial training

October 31, 2022

Presentation at Fall 2022, CS 7938: Image Analysis Seminar, University of Utah, Salt Lake City, Utah

ADASSM: Adversarial Data Augmentation in Statistical Shape Models From Images

September 18, 2023

Presentation at Fall 2023, CS 7938: Image Analysis Seminar, University of Utah, Salt Lake City, Utah