profile picture

Hi, I’m Sadanand Singh. I work at Micron. I lead teams building AI models and DS solutions in DRAM and NAND manufacturing. I am an avid lover of Algorithms, Python, Rust, C++, Javascript. My background is in computational physics, numerical modeling and optimization and software design. I have a PhD in computational physics/chemical engineering focusing on computational models of nano-materials and bio-molecules.


My active area of research is in applications of statistics, machin learning, deep learning and computer vision in manufacturing, medical imaging and in the radiology business. I am specifically interested in designing AI models that are explainable and tractable.

Research Interests

  • Applications of deep learning algorithms to multi-modal screening mammography for classification and localization of cancer
  • Development of explainable AI models for medical applications.
  • Domain adaptation of Deep Leaning and Machine Learning models
  • Self-supervised learning to incorporate large amounts of unlabeled medical data
  • Applications of modern web technologies to make deep learning research more transparent, reproducible and easy to share.
  • Using fundamentals of stochastic learning to model consumer behavior.

CV

A copy of my CV can be seen here
This file was last updated on Sun Nov 6 2022.

AI and Computer Vision Skills

Skills I acquired during 10+ years of being in grad school and tech industry

  • Deep Learning / Machine Learning

    • PyTorch, Tensorflow and Keras
    • Self Supervised Learning
    • GANs & Autoencoders
    • Mlflow Development
    • Tree-based Models
    • Support Vector Machines
    • Clustering Algorithms
    • Mixture Models

  • Medical Imaging

    • Breast Screening
    • Tomography Synthesis Images
    • Anamoly Detection
    • Object Detection with Detectron2
    • Medical Imaging In-painting via GANs
    • Trust in Medical AI systems
    • Explanatory Models

Web Development Skills

Skills I learned during my free time and special during the quarantine

  • Front-end Development

    • React: Gatsby
    • Tailwind CSS and Component Libraries
    • 11ty
    • SEO-friendly and Accessible Designs

  • Backend Development

    • MongoDB
    • PostgreAQL
    • Flask
    • Graphql

Publications

2022

  • Deep is a Luxury We Don't Have

    Ahmed Taha, Nhi Truong Vu, Brent Mombourquette, Thomas P. Matthews, Jason Su, Sadanand Singh
    MICCAI 2022
    Arxiv

2021

  • An improved mammography malignancy model with self-supervised learning

    Nhi Truong Vu, Trevor Tsue, Jason Su, Sadanand Singh
    SPIE - Medical Imaging 2021
    PDF

  • A Multisite Study of a Breast Density Deep Learning Model for Full-Field Digital Mammography and Synthetic Mammography

    Thomas P. Matthews, Sadanand Singh, Brent Mombourquette, Jason Su, Meet P. Shah, Stefano Pedemonte, Aaron Long, David Maffit, Jenny Gurney, Rodrigo Morales Hoil, Nikita Ghare, Douglas Smith, Stephen M. Moore, Susan C. Marks, Richard L. Wahl
    Radiology: Artificial Intelligence Vol. 3(1)
    PDF Arxiv

2020

  • Adaptation of a deep learning malignancy model from full-field digital mammography to digital breast tomosynthesis

    Sadanand Singh, Thomas P. Matthews, Meet Shah, Brent Mombourquette, Trevor Tsue, Aaron Long, Ranya Almohsen, Stefano Pedemonte, Jason Su
    SPIE - Medical Imaging 2020
    PDF Arxiv

  • A hypersensitive breast cancer detector

    Stefano Pedemonte, Brent Mombourquette, Alexis Goh, Trevor Tsue, Aaron Long, Sadanand Singh, Thomas P. Matthews, Meet Shah, Jason Su
    SPIE - Medical Imaging 2020
    PDF Arxiv

2019

  • Generating a Machine Learning Model to Place SRAFs

    Sadanand Singh, Ramon Calderer, Vasudev Lal
    PATENT
    Patent

2013

2012

2011

2010

2009

2008

  • Case Study on Tubular Reactor Hot-Spot Temperature Control for Throughput Maximization

    Sadanand Singh, Shivangi Lal, Nitin Kaistha
    Ind. Eng. Chem. Vol 47(19)
    PDF

Get in touch 👋

Feel free to email me about anything. Want some advice? Give some feedback?

You can also reach me around the web: GitHub, Twitter, LinkedIn

Say hello