An exploration of compressed sensing fMRI time series with 3 different algorithms. Typically, compressed sensing reconstructs a single volume of MRI but fMRI are composed of many volumes; sensing along the time domain could reduce the number of volumes required. Of the 3 algorithms, BSBL-BO performed the best with the error curve elbowing around 30% subsampling.
Built a JetBot + an exploration and novice implementation of motion planning on said JetBot. This computational game theory project marked my first foray into optimization and a glimpse of its power muahahaha. It ain’t exquisite but it was heading in the right direction.
This Fall 2021, I am taking a course on computational game theory, which insofar is the formulation of various games (e.g. bimatrix, Stackelberg) as mathematical programs and the algorithms that solve them, or approximate solutions. Linear complementarity problems are foundational for computing Nash equilibria of simple games.
After brief description of diffusion tensor images and what information they provide, I discuss an intuitive seed-based line propagation algorithm for computing a tractography map of a neuroimage. The open-source softwares required are
3D Slicer,
ITK for C++, and
ITK-SNAP.
StoneAnno is my first published first-authorship paper, presenting at SPIE 2022. With the long-term goal of fully-automated robotic endoscopic surgery, we built a dataset of endoscopic kidney stone removal videos and investigated U-Net, U-Net++, and DenseNet for the segmentation task. We found a U-Net++ model that consistently achieves >0.9 Dice score, with low loss, and produces realistic, convincing segmentations. Moving forward, I am implementing our model on hardware for deployment in ORs, as a part of my master’s thesis, and I helped Dr. Kavoussi submit an R21 grant in October 2021.
Although it will be published after
StoneAnno, this shape analysis is my first completed research project and technically my first first-authorship,
submitted to Brain. I wrote code in R and MATLAB to fit LMMs to the cortical data from T1w MRI of HD patients and then performed statistical analyses on the results using SurfStat and random field theory. We found that, with a novel method for measuring gyrification,
LGI uniquely detects changes in the insula.Neural network automated verification of a VAE and SegNet using NNV. Although neural networks are promising, they are easily confused, particularly if the input domain is perturbed. In this project, I demonstrate the robustness of MNIST-trained VAE and SegNet against varying brightness attacks.
I fell in love with dimensionality reduction when I was learning statistical ML. Since I also study neuroscience, I wanted to practice the art at the intersection of my interests. I compared the 3D projections of a 53-dimensional neurophysiology dataset produced by PCA and a shallow autoencoder.