Algorithms Toward a Next Generation Pacemaker
| dc.contributor.advisor | Aazhang, Behnaam | |
| dc.creator | Banta, Anton Reza | |
| dc.date.accessioned | 2021-05-03T20:59:49Z | |
| dc.date.available | 2021-05-03T20:59:49Z | |
| dc.date.created | 2021-05 | |
| dc.date.issued | 2021-04-30 | |
| dc.date.submitted | May 2021 | |
| dc.date.updated | 2021-05-03T20:59:49Z | |
| dc.description.abstract | This work describes the development and implementation of machine learning algorithms to enhance the functionality of pacemaker technology. This is done by approaching two limitations of current pacemakers: the inability to remotely monitor the $12$-lead surface electrocardiogram (ECG) and the need to hand pick the pacing parameters in the device over time. First, we propose a method to facilitate the remote follow up of patients suffering from cardiac pathologies and treated with an implantable device, by reconstructing a $12$-lead ECG from the intracardiac electrograms (EGM) recorded by the device. The method proposed to perform this reconstruction is a convolutional neural network. These methods are evaluated on a dataset retroactively collected from $14$ patients. Correlation coefficients calculated between the reconstructed and the actual ECG show that the proposed convolutional neural network method represents an efficient and accurate way to synthesize a $12$-lead ECG. Second, we propose a framework for automatically choosing the optimal parameters for a pacemaker. In a typical pacemaker implantation procedure, a cardiologist must determine optimal pacing parameters for the patient that results in healthy blood flow and heart conductance. Thirty percent of patients do not achieve a healthy vascular condition from standard pacemakers, which is believed to be due to the choice of parameters being sub optimal. Thus, the objective of this work is to develop an algorithm that finds an optimal choice of pacing parameters for a given patient. To do this, a set of $48$ different pacing parameters is used on a porcine model and the corresponding $12$-lead ECG, pressure-volume data, and intracardiac signals were measured. Metrics were calculated from this data to provide a framework for choosing the best choice of pacing parameters based on a multinomial classification model. These two bodies of work demonstrate the application and development of machine learning algorithms to pacemaker technology to improve the diagnostic and therapeutic abilities of the device. | |
| dc.format.mimetype | application/pdf | |
| dc.identifier.citation | Banta, Anton Reza. "Algorithms Toward a Next Generation Pacemaker." (2021) Master’s Thesis, Rice University. <a href="https://hdl.handle.net/1911/110401">https://hdl.handle.net/1911/110401</a>. | |
| dc.identifier.uri | https://hdl.handle.net/1911/110401 | |
| dc.language.iso | eng | |
| dc.rights | Copyright is held by the author, unless otherwise indicated. Permission to reuse, publish, or reproduce the work beyond the bounds of fair use or other exemptions to copyright law must be obtained from the copyright holder. | |
| dc.subject | Machine Learning | |
| dc.subject | Pacemakers | |
| dc.subject | Neural Networks | |
| dc.title | Algorithms Toward a Next Generation Pacemaker | |
| dc.type | Thesis | |
| dc.type.material | Text | |
| thesis.degree.department | Electrical and Computer Engineering | |
| thesis.degree.discipline | Engineering | |
| thesis.degree.grantor | Rice University | |
| thesis.degree.level | Masters | |
| thesis.degree.name | Master of Science |
Files
Original bundle
1 - 1 of 1