Nitin Agarwal

I am a PhD student (2014-) in the Computer Science department at UC Irvine, where I am supervised by Gopi Meenaskshisundaram in the IGRAVI Lab. Prior to joining UC Irvine, I was an algorithm developer at VisionGate, a medical device startup based in Phoenix, which uses 3D imaging of single cells for early cancer diagnosis. In summer 2016, I spent a wonderful time in the MCS Division at Argonne National Lab working on multi-view reconstruction of small insects using Lytro Camera's.

I received my Masters (2011-2013) from the University of Washington, Seattle where I was advised by Eric Seibel (Human Photonics Laboratory) and Anthony Reeves (Vision & Image Analysis group). I did my bachelors in Electrical Engineering from Birla Institute of Technology & Sciences (BITS), Pilani, India.

Email  /  CV  /  Google Scholar  /  LinkedIn  /  Github

News

Research
icvgip

Robust Registration of Mouse Brain Slice with Severe Histological Artifacts
Nitin Agarwal, Xiangmin Xu, Gopi Meenakshisundaram
Indian Conference on Computer Vision, Graphics & Image Processing (ICVGIP), 2016  
project page / poster / bibtex

We proposed an end-to-end pipeline for non-linear registration of mouse brain histology slices (with various histological artifacts) to the Allen reference atlas.

miccai

Automatic Detection of Histological Artifacts in Mouse Brain Slice Images
Nitin Agarwal, Xiangmin Xu, Gopi Meenakshisundaram
Workshop on Medical Computer Vision: Algorithms for Big Data (MICCAI), 2016  (Long Oral Presentation)
project page / slides / poster / bibtex

We proposed a method to automatically detect artifacts, which are common to slices produced from conventional histological techniques.

DNA ploidy measure of Feulgen-stained cancer cells using 3D image cytometry
Nitin Agarwal, Yiting Xie, Florence Patten, Anthony P. Reeves, Eric Seibel
IEEE Engineering in Medicine and Biology (EMBC), 2014   (Oral Presentation)
slides / bibtex

We propsed a method to accurately measure the full DNA content (in three-dimension) of single cells without compromising their 3D morphology. This has been proven to be a valuable alternative method for assessing tumorgenesis of enriched microsamples in diagnostic cells.

3D DNA image cytomtery by optical projection tomographic microscope for early cancer diagnosis
Nitin Agarwal, Alberto Biancardi, Florence Patten, Anthony P. Reeves, Eric Seibel
Journal of Medical Imaging (JMI), 2014  
bibtex

We extend the current cytopathological techniques from 2D to 3D for early cancer detection. We do this by accurately imaging individual cells along with novel 3D nuclear segmentation algorithms.

spie

Quantification of relative chromatin content in flow cytometry standards using 3D OPTM imaging technique
Nitin Agarwal, Alberto Biancardi, Florence Patten, Anthony P. Reeves, Eric Seibel
SPIE Medical Imaging (SPIE), 2013  
poster / bibtex

We accurately measure DNA content in small Flow Cytometry standards (chicken and trout nuclei).

Course Projects

Guess the Celebrity in TV Shows!
Nitin Agarwal, Jia Chen, 2015.
talk


In this project we develop a system that takes a TV show or a movie as input and not only detects but also identifies all the characters in that video sequence. We obtained intresting results when we trained our classifier on Season 9 of Friends and later tested it on the Friends movie trailer released in 2014.

ml

Predicting Keypoint location on face images
Nitin Agarwal, Jia Chen, Hao Zhang, 2015.


For this project, the problem was taken from Kaggle. To solve this multivariate (30 labels) problem we tried a gamut of approaches including linear regression, SVM, ensemble techniques, feed forward neural network, convolutional neural networks, etc.

Personal
photo


In my spare time I love to do photography. Do check out my blog.


Thanks Jon!