Mahir Patel

665 Commonwealth Ave, Boston, MA 02215
mahirp@bu.edu

First-Year Computer Vision Ph.D. student supervised by Prof. Margrit Betke, Prof. Bryan Plummer, and Prof. Michael Hasselmo. I am interested in interpreting the environment from different visual modalities, especially for applications to clinical and behavioral neuroscience research.

Research Projects

OptiPose

A deep-learned pose optimizer framework that uses spatio-temporal architecture to learn postural information.

Rodent3D

Dataset

A multi-view multi-modal Rodent dataset

D.I.S.H.A

An automated tool to analyze One-Hybrid yeast assays to accelerate research.

D.I.S.H.A Viewer

A visualization software counter-part for the D.I.S.H.A package.

Publications

Journals, Conference, and Thesis

Paper: Paired yeast one-hybrid assays to detect DNA-binding cooperativity and antagonism across transcription factors - Nature Communications

Berenson, Lane, Soto-Ugaldi, Patel, Ciausu, Li, Chen, Shah, Santoso, Liu, Spirohn, Hao, Hill, Vidal, Fuxman Bass

October 2023

Paper: Animal Pose Tracking: 3D Multimodal Dataset and Token-based Pose Optimization - IJCV

Patel, Gu, Carstensen, Hasselmo, Betke

November 2022

Thesis: Learned Structural And Temporal Context For Dynamic 3d Pose Optimization And Tracking - BU

Patel

May 2022

Abstract & Posters

Abstract: BU-CVKit: Extendable Computer Vision Framework for Behavioral Neuroscience - Society For Neuroscience

Patel, Carstensen, Gu, Malmberg, Betke, Hasselmo

November 2023

Abstract: Longitudinal tracking of spatial responses to environmental manipulations in the retro-splenial cortex - Society for Neuroscience

Malmberg, Mattessich, Carstensen, LaChance, Patel, Everett, Alexander, Goddard, Hasselmo

November 2023

Poster: BU-CVKit: Extendable Computer Vision Framework for Species Independent Tracking and Analysis - CV4Animals@CVPR

Patel, Carstensen, Gu, Hasselmo, Betke

June 2023

Poster: Animal Pose Tracking: 3D Multimodal Dataset and Token-based Pose Optimization - CV4Animals@CVPR

Patel, Gu, Carstensen, Hasselmo, Betke

June 2022

Poster: A Rate Code of Environmental Alteration and High-Resolution 3D Pose Tracking During Behavior - BU

Carstensen, Patel, Gu, Alexander, Chapman, Lee, Betke, Hasselmo

May 2022

Abstract: Neural correlates of environmental alterations in retro-splenial cortex and high-resolution infrared tracking of behavior

Carstensen, Alexander, Chapman, Lee, Patel, Gu, Jalal, Betke, Hasselmo

2021

Pet Projects

Snake-AI

CUDA

An Intelligent agent for playing my port of the classic videogame Snake

SnakeClassic

Port of the classic video game Snake

Snake 2

New version with custom sprites and level designing capabilities

Level Editor

JavaFx

Level editor and sprite selector tool for my game

Professional Experience

Research Technician

Psychology and Brain science @ BU

Setting up multi-view tracking cameras and hardware synchronization.
Writing recording software and neural data analysis/visualizing tools.
Designing new architecture for end-to-end 3D pose estimation and optimization.
Designing new architecture for solving the abstraction and reasoning challenge.

June 2022 - June 2023

Software Engineer Trainee

Jekson Vision

Created multithreaded efficient communication API for two industrial inkjet printers and one laser printer.
Maintained and resolved bugs for the software that manages high-speed serialization for pharmaceutical applications with team of 13 people in 3-month sprints.
Led team of trainees to implement distributed packing system server to parallelize the standard pharmaceutical serialization process and reduce costs.
Presented demo of the distributed packing system to international customers.

August 2019 - September 2020

Software Development Intern

Silicon Valley Infomedia

Created communication protocol, Server and APIs for collecting data from different GPS tracking devices. Which were then used to analyze patterns of the path of the school vehicles and detect deviations from the path.

June 2018 - December 2018

Academic Experience

Graduate Research Assistant

Neuro-Autonomous Robotics @ BU

Implemented 3D reconstruction from multi-view thermal cameras and designed configurable software to generate analytical data such as displacement vectors and velocity vectors.
Customized the 3D calibration cube with Peltier devices for thermal cameras.
Annotated 100s of frames for each view and trained deep neural network to generate landmarks on rodents.
Ported old proprietary FLIR SFMOV video editing code to Linux Systems.

September 2020 - May 2022

Research Affiliate

Strategic Analytics Lab, Harvard Medical School

Scraped more than 15,000 research articles from PubMed, from different reputed journals.
Created a software to help annotate abstracts and build the dataset.
Fine tuned PubMedBERT by Microsoft to classify research articles.

June 2021 - December 2021

Research Engineer Volunteer

Disha Y1H Analyzer @ Fuxman Bass Lab

Designed a computer vision pipeline to localize and segment quad-colonies from Yeast one-hybrid assay images, assigning scores to represent the interactions of different permutation of Transcription Factors, to accelerate the research in fields like Cancer research and immune responses.

February 2021 - Present