About Me

I am Senior Imaging Scientist and lead Applied Imaging & Analytics team at Corteva Agriscience. My Ph.D. research involved designing image processing techniques and computational models that can facilitate automated cervical cancer diagnosis from histopathologic images of tissue biopsies. I have worked with various biomedical image databases like cervical histology (microscopy images), cervical cytology (microscopy images), chest X-ray scans, GI tract polyps from colonoscopy and wireless capsule endoscopy, foot prosthesis; and natural images datasets like human face images and scene text images.

Research Interests

  • Computer Vision
  • Digital Image Processing
  • Machine Learning
  • Deep Learning
  • Embedded Systems
  • Digital Agriculture
  • Digital Pathology

Publications

  • Sudhir Sornapudi, Rajhans Singh. "Self-Supervised Backbone Framework for Diverse Agricultural Vision Tasks." Authorea. February 07, 2023. Published in NAPPN 2023 Abstracts. DOI: 10.22541/au.167573772.22946655/v1
    [web-link] [abstract] [poster]
  • Sudhir Sornapudi. "Self-Supervised Representation Learning for Digital Agriculture." ArXiv 2024.
    [ArXiv]
  • Jesse Bier, Srinivas Sridharan, Sudhir Sornapudi, ​Qiao Hu, Siva P. Kumpatla​ (2022) "A Generative Adversarial Network-based method for High Fidelity Synthetic Data Augmentation". In the Proceedings of the 15th International Conference on Precision Agriculture (ICPA). Minneapolis, MN: International Society of Precision Agriculture.
    [web-link]
  • Brown Gregory T., Sudhir Sornapudistrong>, and Paul Fontelo. “Development of a multi-class deep-learning algorithm capable of diagnosing many histopathologic entities in digital pathology.” AMIA Annual Symposium Proceedings, 2021.
  • Sudhir Sornapudi, Ravitej Addanki, R. Joe Stanley, William V. Stoecker, Rodney Long, Sameer Antani, Rosemary Zuna, Shelliane R. Frazier. "Fully Automated End-to-end Cervical Histology Whole Slide Image Diagnosis Toolbox". J Pathol Inform. 2021 Jun 9;12:26. doi: 10.4103/jpi.jpi_52_20. PMID: 34447606; PMCID: PMC8356709.
    [web-link]
  • Sornapudi, Sudhir, "Deep learning for digitized histology image analysis" (2020). Doctoral Dissertations. 3110.
    [web-link]
  • Sudhir Sornapudi, R. Joe Stanley, William V. Stoecker, Rodney Long, Sameer Antani, Zhiyun Xue, Rosemary Zuna, Shelliane R. Frazier. "DeepCIN: Attention-based Cervical Histology Image Classification with Sequential Feature Modelling for Pathologist-Level Accuracy". J Pathol Inform 2020;11:40
    [pdf] [bibtex] [web-link]
  • Rajaraman, S., Sornapudi, S., Alderson, P. O., Folio, L. R., & Antani, S. K. (2020). “Interpreting Deep Ensemble Learning through Radiologist Annotations for COVID-19 Detection in Chest Radiographs”. MedRxiv. https://doi.org/10.1101/2020.07.15.20154385
    [pdf] [bibtex] [web-link]
  • Sornapudi, S., Stanley, R. J., Stoecker, W. V, Long, R., Xue, Z., Zuna, R., Frazier, S. R., Antani, S. (2020). “Feature based Sequential Classifier with Attention Mechanism”. ArXiv. https://arxiv.org/abs/2007.11392.
    [pdf] [bibtex] [web-link]
  • Sornapudi, S., Addanki, R., Stanley, J., Stoecker, W. V, Long, R., Zuna, R., Frazier, S. R., Antani, S. (2020). “Cervical Whole Slide Histology Image Analysis Toolbox”. MedRxiv. https://doi.org/10.1101/2020.07.22.20160366
    [pdf] [bibtex] [web-link]
  • Sornapudi S, Hagerty J, Stanley RJ, Stoecker WV, Long R, Antani S, Thoma G, Zuna R, Frazier SR. "EpithNet: Deep regression for epithelium segmentation in cervical histology images". J Pathol Inform 2020;11:10. DOI: 10.4103/jpi.jpi_53_19
    [pdf] [bibtex] [web-link]
  • Sornapudi, S., Brown, G. T., Xue, Z., Long, R., Allen, L., & Antani, S. (2020). Comparing Deep Learning Models for Multi-cell Classification in Liquid- based Cervical Cytology Image. AMIA Annual Symposium Proceedings, 2019, 820–827. PMID: 32308878; PMCID: PMC7153123.
    **Won the Distinguished Paper Award at AMIA 2019 Annual Symposium, Washington D.C. November 2019.
    [slides] [web-link] [bibtex]
  • S. Rajaraman, S. Sornapudi, M. Kohli and S. Antani, "Assessment of an ensemble of machine learning models toward abnormality detection in chest radiographs", 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Berlin, Germany, 2019, pp. 3689-3692. DOI: 10.1109/EMBC.2019.8856715
    [pdf] [bibtex] [web-link]
  • Sornapudi, Sudhir; Meng, Frank; Yi, Steven. 2019. "Region-Based Automated Localization of Colonoscopy and Wireless Capsule Endoscopy Polyps" Appl. Sci. 9, no. 12: 2404. DOI: 10.3390/app9122404.
    [pdf] [bibtex] [web-link]
  • Sornapudi, S., Stanley, R. J., Stoecker, W. V, Almubarak, H., Long, R., Antani, S., Frazier, S. R. (2018). "Deep Learning Nuclei Detection in Digitized Histology Images by Superpixel". Journal of Pathology Informatics, 9(1), 5. DOI: 10.4103/jpi.jpi_74_17.
    [pdf] [bibtex] [web-link]
  • Sornapudi, Sudhir, "Nuclei segmentation of histology images based on deep learning and color quantization and analysis of real world pill images" (2017). Masters Theses. 7710.
    [web-link]
  • Sornapudi S., Joe Stanley R., Hagerty J. and V. Stoecker W. (2017). "Real-world Pill Segmentation based on Superpixel Merge using Region Adjacency Graph". In Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, (VISIGRAPP 2017) ISBN 978-989-758-225-7, pages 182-187. DOI: 10.5220/0006135801820187.
    [pdf] [poster] [bibtex] [web-link]
  • Guo P, Stanley RJ, De S, Long LR, Antani SK, Thoma GR, Demner-Fushman D, Sornapudi S; "Features Advances to Automatically Find Images for Application to Clinical Decision Support". Medical Research Archives. 4(7) 2016. DOI: 10.18103/mra.v4i7.761.
    [pdf] [bibtex] [web-link]
  • Sudhir Sornapudi, Jason H., R. Joe Stanley, William V. Stoecker, Rodney Long, Sameer Antani, George Thoma, Rosemary Zuna, Shelliane R. Frazier. "Regression based Deep Neural Networks for Epithelium Segmentation in Histopathology Images". Poster presenation at 4th Annual Ozark Biomedical Initiative Research Symposium, Rolla, MO. September 2019. [poster]
  • William V. Stoecker, Haider A. Almubarak, Sudhir Sornapudi, Peng Guo, Jason Hagerty, R. Joe Stanley R.. "Update on Microscopic Image Processing: Detecting Successively Finer Structures – Architectures, Cells, Nuclei". Poster presenation at Ozark Biomedical Initiative Symposium, Rolla, MO. August 2018.
    [poster]
  • Sudhir Sornapudi, Joe Stanley R., Hagerty J. and V. Stoecker W. "Real-world Pill Segmentation based on Superpixel Merge using Region Adjacency Graph". Oral presenation at Ozark Biomedical Initiative Symposium, Rolla, MO. August 2017.
  • Sornapudi, S., & Jain, P. (2019). “Deep Learning-based Text Detection and Recognition in the Research Lab”. Retrieved May 5, 2020, from https://www.brightlab.com/lab-automation/deep-learning-based-text-detection-and-recognition-in-the-research-lab/ [Company Blog] [Medium Blog]
  • Education

    Missouri University of Science and Technology, Rolla

    Ph.D. in Computer Engineering GPA: 3.8/4.0 July 2020

    Dissertation: Deep Learning for Digitized Histology Image Analysis

    Missouri University of Science and Technology, Rolla

    Masters in Computer Engineering GPA: 4.0/4.0 April 2017

    Thesis: Nuclei segmentation of histology images based on deep learning and color quantization and analysis of real world pill images

    JNTU Kakinada - University College of Engineering Vizianagaram

    B.Tech. in Electronics and Communications Engineering GPA: 3.8/4.0 May 2014

    UG Final project: Library Management System Using RFID

    Industry Experience

    Corteva Agriscience

    Senior Data Scientist - I 26 Dec 2022 – Present

    Leader, Applied Imaging and Analytics

  • Led a team of data scientists to deliver high-impact R&D computer vision projects within biotechnology, seeds, and crop protection sub-functions.
  • Led Corteva-ISU collaboration under the Translationa AI Center (TrAC) program. Currently serving as an Industry Board member.
  • Filed a patent for a novel way of estimating canola seed loss and another patent pending submission on insect bioassay work. Presented an abstract and poster on self-supervised representation learning at the NAPPN 2023 conference.
  • Devised new ways to interview candidates to assess their technical skills better. These methods are used by other team leads.
  • Imaging Data Scientist - II 3 Oct 2022 – 26 Dec 2022

  • Achieved ambitious goals by delivering results in the areas of insect bioassays, plant phenotyping, and biotechnology.
  • Strong technical knowledge and upheld the values of Be Curious and Build Together.
  • Contributed to the Hackathon, LeScientific, and ICPA paper on GANs, and brought self-supervised learning to the company.
  • Associate Imaging Data Scientist May 2020 - Present

  • Proposed various imaging and computer vision-based research solutions by understanding the challenges faced in predictive analytics in agriculture to improve, enhance or modify the existing processes to generate accurate and reliable information for the end-users.
  • Conducted meetings and communicated with clients and project stakeholders to update them with findings and progress on proposed image analytic solutions.
  • Developed and deployed containerized solutions using Docker, Kubernetes, and MaaS.
  • Designed scalable systems with Argo workflows to process large image and video datasets for deep neural network architectures on AWS and GCP instances using multi-core, GPU/TPU instances.
  • Presented research findings and proposed solutions to share and disseminate knowledge at internal and external venues.
  • Led an innovation and technology team to apply self-supervised learning to solve digital agriculture problems
  • MilliporeSigma, Merck KGaA

    Computer Vision Intern May 2019 - August 2019

  • Designed a prototype to automatically update digital inventory by scanning images of chemical reagents.
  • Implemented deep learning based text detection and recognition modules to extract information from the labels attached to the chemical reagents.
  • The model helped scientists to onboard and update inventory within seconds.
  • Contributed to Brightlab team at hackathon event by detecting the weights from the images of a digital balance.
  • U.S. National Library of Medicine, NIH, HHS

    Research Scientist Co-op August 2018 – December 2018

  • Designed a pipeline to read raw cervical cytology raw slide images and produce clean high-resolution annotated patch data.
  • Implemented a novel graph-based approach to detect nuclei and cell boundaries from complex overlapping cell images.
  • Evaluated and analyzed various CNN models to classify and differentiate abnormal cell images from collection of normal cell images.
  • Brown Bag Lecture at US National Library of Medicine (Abstract)
  • Won the Distinguished Paper Award for presenting this work at AMIA 2019 Annual Symposium, Washington D.C. November 2019.
  • Xyken LLC

    Software (Image Processing) Intern May 2018 – August 2018

  • Implemented a customized region-based CNN model for performing instance segmentation and detection on colonoscopy and wireless capsule endoscopy (WCE) polyp frames.
  • Designed a tool to annotate capsule endoscopy videos.
  • Worked on skin segmentation of foot image frames, extracted from a user video, using a deep neural network and Gaussian mixture model to ultimately recreate a 3D model of a foot for foot prosthesis.
  • The skin segmentation approach is directly incorporated in the Xyken’s iDr 3D android mobile application.
  • Naval Science & Technological Laboratory, DRDO

    Research Intern November 2012 – December 2012

  • Modeled Pitch Controller of Unmanned Underwater Vehicle (UUV).
  • Implemented the transfer function and carried out the design of linear controller using the standard control system toolbox from MATLAB software.
  • Academic Experience

    Missouri University of Science and Technology

    Research Assistant, Image Processing Laboratory November 2015 - July 2020

  • Image Analysis and Object Recognition. Perform research into software design and development involving image processing and feature extraction for automated nuclei detection in histology images using novel algorithms involving computer vision and computational intelligence imaging techniques.
  • Deep Learning and Pattern Recognition. Investigate computational intelligence paradigms for automatic Cervical Cancer image recognition. Develop novel and hybrid artificial intelligence algorithms to improve nuclei identification and epithelium segmentation. Superpixels extraction and classification using Neural Network based on Backpropagation, SVM, Clustering, Deep Convolutional Neural Networks and Recurrent Neural Networks.
  • Hands on experience with data collection, feature selection, data training and evaluation.
  • Missouri University of Science and Technology

    Graduate Teaching Assistant, Digital Electronics Laboratory August 2016 - May 2020

  • Taught Digital Logic (CpE 2211) and Microcontrollers (CpE 3151) laboratory courses.
  • Delivered a range of teaching and assessment activities including tutorials for students.
  • Provided effective timely and appropriate feedback to students to support their learning.
  • Teaching topics include AVR microcontroller, assembly language programming and embedded C programming; and digital logic design and analysis using Altera Quartus II, Model Sim (firmware) and FPGA (hardware).
  • JNTU Kakinada - University College of Engineering Vizianagaram

    Co-Founder and Chief Designer, ESPECTRO (Electronics and Communications Engineering Monthly Wall Magazine) April 2013 - April 2014

  • Started ECE department’s monthly magazine ‘ESPECTRO’ to encourage creative skills of students and to create awareness of recent trends in technology.
  • Formed and worked with various groups to gather valuable content required to publish in the magazine.
  • Circulated the magazine copies to nearby colleges.
  • Get In Touch.

    || Contact me ||