Name: Rohit Malyala
Type: User
Company: Faculty of Medicine, University of British Columbia
Bio: Medical Student at the University of British Columbia. 3D modeler/animator, Unity, Blender, Python, and R user.
Location: Vancouver, BC
Rohit Malyala's Projects
A procedural Blender pipeline for photorealistic training image generation
Non-invasive real-time spectrophotometric quantification of ex-vivo hemoglobinuria with one-step calibration. Accompanying Arduino code for use with the AMS AS7262 spectral sensing unit, as part of the open hematuria monitor project.
Intersection over union for multi-bounding-box annotations between workers and valid experts. Works even if number of boxes between worker and expert annotations is not equal. Made to work with batch_results.csv files from Amazon Mechanical Turk.
This is a blog with .Rmd posts knitted into .md files and rendered with Jekyll.
Functions to draw graphs on LCD
A simplified, modifiable R function to conduct the Jonckheere-Terpstra trend test, a method of comparing trends in distributions corresponding to ordinal indepedent factors.
Personal website for Rohit Malyala
A collection of files and resources to test the following hypothesis: increased stereoscopic depth as produced by higher inter-camera distance in side-by-side stereoscopic videography improves learning of neuroanatomy video modules
Official repository for Citation Style Language (CSL) citation styles.
Published manuscript attached. A collection of custom R commands for typical epidemiological analyses and publication-ready output. Includes functions for looping univariate regressions, latex table output, and publication-appropriate ggthemes. A series of Excel functions to simplify retrospective survival analysis are also provided within.
A collection of notes and Python/R code enabling the analysis of a high-saliency dataset of anesthesia time courses from transplant surgeries performed at VGH.
A collection of combined Python/R scripts to reproduce analysis for a study on the use of the Synthetic Minority Oversampling Technique (SMOTE) to mitigate protected class-imbalances in transplantation datasets and improve prediction in minority classes.
ChatGPT web interface using the OpenAI API
Introducing Virtual Vesalius. A medical education venture aiming to bring true-3D VR to the smartphones of anatomy students everywhere.
A Blender addon for generating synthetic ground truth data for Computer Vision applications
ViVRe is a Unity editor/game application designed for the creation of volume-rendered virtual reality learning objects, primarily sourced from medical imaging data. This volume rendering implementation allows users to "slice" learning objects to visualize deep anatomy, all with relatively inexpensive hardware (i.e. smartphones and Cardboard headsets).
Wendlerized application