Algae Identification Using Machine Learning

Water brings life, and everyone deserves to know if their water is clean and safe to drink. As seen in Lake Erie from the MODIS sensor, harmful algae blooms (HABs) can grow out of control in our waters and produce lethal toxins which negatively affect our livelihood and the environment. For my PhD thesis I am exploring combining multi-spectral imaging with machine learning to automatically identify and count different types of algae in a water sample. I am constructing a custom microscope in order to collect data and then using deep learning methods to provide valuable insights into this data. Blue Lion Labs Ltd. is the spin off company that is exploring commercialization of this technology.

Relevant Publications

  1. Deglint, J. L., Jin, C., & Wong, A. (2019). Investigating the Automatic Classification of Algae via Deep Residual Learning. Springer, Lecture Notes in Computer Science.
  2. Deglint, J. L., Jin, C., & Wong, A. (2019). A Multispectral Bayesian-based Computational Microscopy Method for Enhancing Image Quality. In Imaging, Manipulation, and Analysis of Biomolecules, Cells, and Tissues XVII. International Society for Optics and Photonics.
  3. Deglint, J. L., Jin, C., Chao, A., & Wong, A. (2018). The Feasibility of Automated Identification of Six Algae Types using Feed-forward Neural Networks and Fluorescence-based Spectral-morphological Features. IEEE Access, 1–1. https://doi.org/10.1109/ACCESS.2018.2889017
  4. Deglint, J. L., Tang, L., Wang, Y., Jin, C., & Wong, A. (2018). SAMSON: Spectral Absorption-fluorescence Microscopy System for ON-site-imaging of algae. Journal of Computational Vision and Imaging Systems, 4(1), 3–3.

Awards

Scholarships

  • Davis Memorial Scholarship in Ecology provided by the University of Waterloo (September 2018 - April 2019)
  • T.E. Unny Memorial Award provided by the University of Waterloo (November 2018)
  • President’s Graduate Scholarship provided by the University of Waterloo (September 2016 – August 2019)
  • Alexander Graham Bell Canada Graduate Scholarship-Doctoral (CGS D) provided by NSERC (September 2016 – August 2019)

Artery Tracking from Ultrasound Videos

Blood is the life that runs through our veins and arteries via the heart. Determining the cardiovascular health of a patient is often assessed by imaging the arteries with a ultrasound probe. During my masters, my colleagues and I created MAUI (Measurements from Arterial Ultrasound Imaging), a software tool for cardiovascular researches to process artery data efficiently and reliably. MAUI is able to automatically detect and track artery walls captured with an ultrasound machine and is now commercially available from Hedgehog Medical Inc.

Relevant Publications

  1. Gawish, A., Deglint, J. L., Zuj, K. A., Egana, A., M., Rocha, J., Wong, A., & Hughson, R. L. (2017). Determining Arterial Blood Velocity Using MAUI Software From Recorded Doppler Ultrasound Videos. Artery Research, (17).
  2. Zuj, K., Deglint, J., Gawish, A., Wong, A., Clausi, D. A., & Hughson, R. L. (2016). A new software for determining changes in arterial diameter over time. Artery Research, (16), 97.
  3. Deglint, J., Gawish, A., Zuj, K., Wong, A., Clausi, D. A., & Hughson, R. L. (2015). Active Contours for Measuring Arterial Wall Diameter of Astronauts from Ultrasound Images. Journal of Computational Vision and Imaging Systems.

Spectral Demultiplexed Imaging

Relevant Publications

  1. Deglint, J., Kazemzadeh, F., Cho, D., Clausi, D. A., & Wong, A. (2016). Numerical demultiplexing of color image sensor measurements via non-linear random forest modeling. Scientific Reports, 6, 28665.
  2. Deglint, J. L., Schoneveld, K., Kazemzadeh, F., & Wong, A. (2016). A Compact Field-portable Computational Multispectral Microscope using Integrated Raspberry Pi. Journal of Computational Vision and Imaging Systems, 2(1).
  3. Deglint, J., Kazemzadeh, F., Wong, A., & Clausi, D. A. (2015). Numerical spectral demulitplexing microscopy of measurements from an anatomical specimen. Journal of Computational Vision and Imaging Systems.
  4. Deglint, J., Kazemzadeh, F., Wong, A., & Clausi, D. A. (2015). Inference of dense spectral reflectance images from sparse reflectance measurement using non-linear regression modeling. In Applications of Digital Image Processing XXXVIII (Vol. 9599, p. 95992G). International Society for Optics and Photonics.
  5. Deglint, J., Kazemzadeh, F., Shafiee, M. J., Li, E., Khodadad, I., Saini, S. S., … Clausi, D. A. (2015). Virtual spectral multiplexing for applications in in-situ imaging microscopy of transient phenomena. In Applications of Digital Image Processing XXXVIII (Vol. 9599, p. 95992D). International Society for Optics and Photonics.

Auto-calibration of a Camera-Projector System

Relevant Publications

  1. Scharfenberger, C., Sekkati, H., Deglint, J., & Post, M. (2017). System and method for online projector-camera calibration from one or more images.
  2. Li, F., Sekkati, H., Deglint, J., Scharfenberger, C., Lamm, M., Clausi, D., … Wong, A. (2017). Simultaneous projector-camera self-calibration for three-dimensional reconstruction and projection mapping. IEEE Transactions on Computational Imaging, 3(1), 74–83.
  3. Deglint, J., Cameron, A., Scharfenberger, C., Sekkati, H., Lamm, M., Wong, A., & Clausi, D. A. (2016). Auto-calibration of a projector–camera stereo system for projection mapping. Journal of the Society for Information Display, 24(8), 510–520.
  4. Deglint, J., Cameron, A., Scharfenberger, C., Lamm, M., Wong, A., & Clausi, D. (2015). 35.1: Distinguished Paper: Auto-Calibration for Screen Correction and Point Cloud Generation. In SID Symposium Digest of Technical Papers (Vol. 46, pp. 507–510).