<?xml version="1.0" encoding="UTF-8"?><?xml-stylesheet type="text/xsl" href="static/style.xsl"?><OAI-PMH xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/ http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd"><responseDate>2026-11-21T04:39:34Z</responseDate><request verb="GetRecord" identifier="oai:macsphere.mcmaster.ca:11375/33179" metadataPrefix="dim">https://prod-ms-be.lib.mcmaster.ca/server/oai/request</request><GetRecord><record><header><identifier>oai:macsphere.mcmaster.ca:11375/33179</identifier><datestamp>2026-06-02T07:08:16Z</datestamp><setSpec>com_11375_271</setSpec><setSpec>col_11375_272</setSpec></header><metadata><dim:dim xmlns:dim="http://www.dspace.org/xmlns/dspace/dim" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:doc="http://www.lyncode.com/xoai" xsi:schemaLocation="http://www.dspace.org/xmlns/dspace/dim http://www.dspace.org/schema/dim.xsd">
   <dim:field mdschema="dc" element="contributor" qualifier="advisor">Kim, Younggy</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="author">Hemmati Sara</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="department">Civil Engineering</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="accessioned">2026-06-01T19:46:32Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="issued">2026</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">https://hdl.handle.net/11375/33179</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">https://doi.org/10.71548/42</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="abstract">Cyanobacteria in nutrient-rich environments can lead to harmful algal blooms (HABs). In HABs, toxic substances (cyanotoxins) produced by cyanobacteria can severely impact water quality and the aquatic environment. Therefore, rapid and reliable identification of cyanobacteria is important for effective water quality monitoring. In this thesis, similar-shaped cyanobacteria were classified using a deep learning model with a relatively small number of multi-wavelength
laser microscopic images. First, a group of common filamentous cyanobacteria was examined using a deep learning model. The model trained only on black-and-white microscopic images achieved an accuracy of only 58.4%. When the model was trained on images captured by a confocal laser scanning microscope using four different wavelength lasers, the classification accuracy increased to 95.5%. Moreover, species- and strain-level classification was successful only with 39 image sets per sample in machine training. Second, the deep learning model was trained to classify Microcystis, Synechococcus, and Cyanobium that co-exist and have similar sizes, shapes, and aggregating patterns. Cyanobacteria samples were classified even at the strain level using multi-wavelength laser microscopy with an accuracy of 98.4%. The effects of training image size, image resolution, and number of epochs on classification performance were also investigated, and it was found that 45 training images per sample, 256 × 256 image resolution, and 200 epochs were recommended for rapid training and reliable classification. Third, 25 cyanobacteria samples were examined using the deep learning model with scanning microscopic images acquired with four different laser wavelengths (445, 561, 594, and 639 nm). The 445-nm laser was found to be the most effective for classifying Dolichospermum lemmermanii LE011-02
with an F1-score of 95.2%. The 561-nm laser was beneficial for classifying Planktothrix rubescens samples, achieving F1-scores of 90% for both strains. However, the 561-nm laser was not useful for classifying Synechococcus samples. The images of the 639-nm laser were most beneficial for the deep learning to recognize Phormidium sp. HD798 and Pseudanabaena mucicola HH011-09A, confirming that the effective laser wavelength for classification varies across samples and even within a genus. The research outcomes of this thesis demonstrated that multi-wavelength laser microscopy can emphasize critical information for cyanobacteria classification in image processing using deep learning models, allowing reliable water quality monitoring under the influence of HABs.</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degreetype">Thesis</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degree">Doctor of Philosophy (PhD)</dim:field>
   <dim:field mdschema="dc" element="language" qualifier="iso">en</dim:field>
   <dim:field mdschema="dc" element="title">Multi-wavelength Laser Microscopy for Cyanobacteria Classification Using Artificial Intelligence</dim:field>
   <dim:field mdschema="dc" element="title" qualifier="alternative">Cyanobacteria Classification Using Artificial Intelligence</dim:field>
   <dim:field mdschema="dc" element="type" lang="en">Thesis</dim:field>
   <dim:field mdschema="others" element="access-status">embargo</dim:field>
</dim:dim></metadata></record></GetRecord></OAI-PMH>