Philips

R&D for Endovascular Robotic Procedure Automation

Worked on building the next generation ICE Catheter guidance platform with applications to Structural Heart Diseases.

Experience researching novel AI based image guided therapy features for peripheral atherectomy devices

Scientist

Elucid

Research Scientist - Image Processing

Rochester Institute of Technology

PhD Dissertation: Segmentation of CT image of cochlea using classical computer vision and deep learning techniques to facilitate drug flow modeling within cochlear compartments
* Using Capsule Networks over regular CNNs to improve performance and reduce model size
* Network design and optimization for 3D volumetric segmentation of medical images.
* Classical computer vision approach of atlas based registration and segmentation for pre-processing the data on Insight Toolkit on C++ platform.
* Extract concentration profiles of contrast agent within the ear and model diffusion parameters using inverse optimization methods.

Graduate Research Assistant

Worked on 2 projects
1. New eye tracker method based on methods other than P-CR algorithm (Patent Pending): Worked on parallelizing computationally expensive parts of the algorithm on a GPU. Used OpenCV with CUDA module on C++ platform.

2. Effects of Vision on Biomechanics: Study of how an individual's visual system responds to obstacles in their way. Parameters varied being obstacle height, obstacle distance from start, real vs. virtual objects. Contribution involved data collection and analysis of previous collected data. Experiments were conducted in an Augmented Reality setup for displaying the virtual object and the subjects walked in a controlled AR environment.

Graduate Teaching Assistant

* Team member: Project on new two sensor scanner technology.
* Work was part of grant received by Hewlett-Packard.
* Applied and obtained patent on the new technology (Patent Pending: US20180288281A1)

Using TEE and X-ray with a combination of deep learning and computer vision to create smart solutions to aid in non-invasive cardiac procedures.

Artificial Intelligence Intern

University of Rochester

* Spearheaded the tendon 3D ultrasound image segmentation project.
* Optimized and designed networks to work with the small limited dataset.

Data Scientist Intern

Los Alamos National Laboratory

* Took responsibility of starting the cancer pathology slide classification project.
* Used core imaging concepts to pre-process the data.
* Established baselines using sparse coding techniques and deep CNN networks (VGG, ResNet).
* Work was presented at CAFCW 2017, and eventually published in BMC Bioinformatics Journal.

Applied Machine Learning Research Intern

Kodak Alaris

Part time work to analyze the need of imaging products to customers and come up with a process to introduce a new product.

Taplio

Sanketh S Moudgalya's Linkedin Analytics

Want to drive more opportunities from LinkedIn?

Taplio

.css-1jgreyo{background-image:linear-gradient(284deg, rgba(39,43,72,1) 0%, rgba(33,47,183,1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Sanketh S Moudgalya's Linkedin Analytics

Want to drive more opportunities from LinkedIn?

Sanketh S Moudgalya's Linkedin Analytics