Show simple item record

dc.contributor.advisorKaya, Mehmet
dc.contributor.authorPetersen, Cassondra Michelle
dc.date.accessioned2021-06-17T15:28:03Z
dc.date.available2021-06-17T15:28:03Z
dc.date.created2021-05
dc.date.issued2021-05
dc.date.submittedMay 2021
dc.identifier.urihttp://hdl.handle.net/11141/3360
dc.descriptionThesis (M.S.) - Florida Institute of Technology, 2021.en_US
dc.description.abstractCardiovascular diseases continue to plague the world as the number one killer of adults each year. As cardiovascular diseases can start with seemingly no symptoms, heart disease has been coined as the silent killer. With the ongoing COVID-19 global pandemic, the need for advances in cardiovascular advances has become even more prevalent as this virus is known to cause more severe damage to those who have underlying cardiovascular problems. These two reasons show the dire clinical need for researchers to continue to use novel techniques such as neural networks to find new approaches for helping physicians non-invasively and quickly add predictive technologies into their routine practices to evaluate a patient’s risk of developing any cardiovascular diseases. The cardiovascular system has historically been hard to replicate using computational flow fluid dynamic software as it is one of the most complex flow systems. Over the years, changes in aortic diameter, blood pressure (BP), and arterial compliance have all been identified as indicators of cardiovascular diseases in humans. Thus, being able to routinely monitor aortic diameter size can help detect a cardiovascular disease sooner and potentially help physicians find meaningful solutions to prevent the progression of these diseases. Currently, there is no approved or widely accepted non-invasive neural network (NN)technology for determining the risk of developing cardiovascular disease in the medical field as there has not been sufficient validation studies of the proposed methods presented by researchers. The goal of this study is to validate the neural network that is trained using simulated blood pressure and flow data to predict aortic systolic and diastolic diameters with clinical data and blood pressure waveforms through the use of MATLAB software. Numerical models are performed on the clinical blood pressure waveforms to estimate the flow and arterial compliance corresponding to the given systolic and diastolic blood pressures. Arterial compliance values are then inputted into the neural network, and the targets provide the systolic and diastolic aorta diameter predictions. The preliminary validation testing results based on human clinical data are discussed and reviewed against published clinical data on aortic diameters.en_US
dc.format.mimetypeapplication/pdf
dc.language.isoen_USen_US
dc.rightsCopyright held by author.en_US
dc.titleValidation of Arterial Distensibility Neural Network Model with Clinical Dataen_US
dc.typeThesisen_US
dc.date.updated2021-06-15T12:47:12Z
thesis.degree.nameMaster of Science in Biomedical Enginneringen_US
thesis.degree.levelMastersen_US
thesis.degree.disciplineBiomedical Engineeringen_US
thesis.degree.departmentBiomedical and Chemical Engineering and Sciencesen_US
thesis.degree.grantorFlorida Institute of Technologyen_US
dc.type.materialtext


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record