Abstract:
Medical imaging is a growing field that has stemmed from the need to conduct non-invasive diagnosis, monitoring and analysis of biological systems. We focused on image acquisition and image analysis using portable microscopes for point-of-care (POC) analysis and image analysis on portable computing units. This thesis focuses on Fourier ptychography, by drawing the experience from Digital In Line Holography and Bright Field Microscopy, previously concluded in other thesis of the author. FP is a computational imaging technique which makes use of a matrix illumination to acquire a set of low-resolution images, that are used to reconstruct a complex image, resulting in a high space-bandwidth product. There are 3 hypotheses in this theses that have been verified / upgraded regarding FP improvement from experimental methodology and theoretical approach. It is proved that by manually focusing the images for a 5x5 dataset to be used in FP, the result complex output has an increased resolution. Also, by changing the wavelength of the LED light source when acquiring the dataset, it produces a different output image. Since smaller wavelength is higher frequency, and high frequency is resolution in k-space, the output image should be of better quality for small wavelengths (green or blue). The third component has observed the bandwidth of the light source itself, by replacing LED light with a narrower bandwidth light source so as to focus the information in transform domain. Narrow bandwidth light source produces a better quality of the output image compared to the traditional LED array light source.
