Multiscale image analysis of calcium dynamics in cardiac myocytes
- Autores: Alexander Vallmitjana Lees
- Directores de la Tesis: Raúl Benítez Iglesias (dir. tes.)
- Lectura: En la Universitat Politècnica de Catalunya (UPC) ( España ) en 2017
- Idioma: español
- Tribunal Calificador de la Tesis: José Julián Rodellar Benedé (presid.), Jordi Soriano Fradera (secret.), Yohannes Shiferaw (voc.)
- Materias:
- Enlaces
- Tesis en acceso abierto en: TDX
- Resumen
Cardiac myocytes constitute a unique physiological system. They are the muscle cells that build up heart tissue and provide the force to pump blood by synchronously contracting at every beat. This contraction is regulated by calcium concentration, among other ions, which exhibits a very complex behaviour, rich in dynamical states at the molecular, cellular and tissue levels. Details of such dynamical patterns are closely related to the mechanisms responsible for cardiac function and also cardiac disease, which is the first cause of death in the modern world. The emerging field of translational cardiology focuses on the study of how such mechanisms connect and influence each other across spatial and temporal scales finally yielding to a certain clinical condition.
In order to study such patterns, we benefit from the recent and very important advances in the field of experimental cell physiology. In particular, fluorescence microscopy allows us to observe the distribution of calcium in the cell with a spatial resolution below the micron and a frame rate around the millisecond, thus providing a very accurate monitoring of calcium fluxes in the cell.
This thesis is the result of over five years¿ work on biological signal and digital image processing of cardiac cells. During this period of time the aim has been to develop computational techniques for extracting quantitative data of physiological relevance from microscopy images at different scales. The two main subjects covered in the thesis are image segmentation and classification methods applied to fluorescence microscopy imaging of cardiac myocytes. These methods are applied to a variety of problems involving different space and time scales such as the localisation of molecular receptors, the detection and characterisation of spontaneous calcium-release events and the propagation of calcium waves across a culture of cardiac cells.
The experimental images and data have been provided by four internationally renowned collaborators in the field. It is thanks to them and their teams that this thesis has been possible. They are Dr. Leif Hove-Madsen from the Institut de Ciències Cardiovasculars de Catalunya in Barcelona, Prof. S. R. Wayne Chen from the Department of Physiology and Pharmacology in the Libin Cardiovascular Institute of Alberta, University of Calgary, Dr. Peter P. Jones from the Department of Physiology in the University of Otago, and Prof. Glen Tibbits from the Department of Biomedical Physiology & Kinesiology at the Simon Fraser University in Vancouver.
The work belongs to the biomedical engineering discipline, focusing on the engineering perspective by applying physics and mathematics to solve biomedical problems. Specifically, we frame our contributions in the field of computational translational cardiology, attempting to connect molecular mechanisms in cardiac cells up to cardiac disease by developing signal and image-processing methods and machine-learning methods that are scalable through the different scales. This computational approach allows for a quantitative, robust and reproducible analysis of the experimental data and allows us to obtain results that otherwise would not be possible by means of traditional manual methods.
The results of the thesis provide specific insight into different cell mechanisms that have a non-negligible impact at the clinical level. In particular, we gain a deeper knowledge of cell mechanisms related to cardiac arrhythmia, fibrillation phenomena, the emergence of alternans and anomalies in calcium handling due to cell ageing.
