Show simple item record

dc.contributor.advisorLong, David S.
dc.contributor.authorContreras, Miguel
dc.date.accessioned2022-06-20T16:31:27Z
dc.date.available2022-06-20T16:31:27Z
dc.date.issued2022-05
dc.identifier.othert22006
dc.identifier.urihttps://soar.wichita.edu/handle/10057/23447
dc.descriptionThesis (M.S.)-- Wichita State University, College of Engineering, Dept. of Biomedical Engineering
dc.description.abstractGaining insight into different cell behaviors is key to better understanding different pathologies. These behaviors may be explained in part through close observation of 3D cell morphology. Therefore, the objective of this research was to develop a machine learning (ML) framework that can predict 3D subcellular morphological variation of endothelial cells (ECs) to generate digital twins. ECs were cultured and their membrane, nucleus, and focal adhesion (FA) sites were stained and imaged with confocal microscopy. An open-source pre-trained ML algorithm was first tested to evaluate the feasibility of predicting EC subcellular morphology based on membrane morphology. Transfer learning was used to train this algorithm to predict nuclear morphology. A second open-source ML algorithm based on a conditional generative adversarial network (cGAN) was then tested to predict 2D morphology of FA sites. A new image similarity metric named Discrete Protein Metric (DPM) was developed to evaluate FA sites prediction accuracy. Finally, the cGAN algorithm was adapted to make 3D predictions of nucleus and FA sites based on membrane morphology. This adaptation of the cGAN algorithm was used to build the ML framework, which was then trained and tested. After training the framework, the results on an independent test showed an average prediction accuracy of ~87% for nucleus and ~70% for FA sites. The predictions were used to build a digital twin of each EC and compared to their respective ground truth, showing an average ~79% global accuracy and ~84% accuracy in FA-Nucleus distribution. The results presented show the effectiveness of the developed ML framework to generate digital twins of ECs using limited amount of data. These digital twins can be used to couple EC morphology with different behaviors. The ML framework can be potentially expanded to predict morphology of other subcellular structures as well as to study other types of cells.
dc.format.extentxii, 103 pages
dc.language.isoen_US
dc.publisherWichita State University
dc.rights© Copyright 2022 by Miguel Contreras All Rights Reserved
dc.subject.lcshElectronic dissertations
dc.titleA machine learning framework to predict subcellular morphology of endothelial cells for digital twin generation
dc.typeThesis


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record