Selective volume illumination microscopy is a technique that achieves volumetric imaging of samples on a mesoscopic scale with enough resolution that it is possible to distinguish details up to a cellular level of the biologic samples. This approach is based on the idea of acquiring multiple images of the sample illuminated with several light sheets that together compose a set of axially modulated patterns, and then to computationally reconstruct the whole volume. The possibility of spatially modulating the illumination paves the way to the exploitation of a variety sampling and reconstruction strategies, such as compressive sensing. Thanks to these approaches, it is possible to enhance the performances of the microscopy technique in terms measurement time, the signal-to-noise ratio and spatial resolution of the acquired images. The goal of this thesis is to optimize the implementation of this technique by working both on the optical setup and on the computational methods. To these purposes, different alternatives were tested with the aim of enhancing the performances of the illumination branch of the microscope; in particular the focus was to improve significatively the amount of light illuminating the sample, thus achieving a better signal to noise ratio of the measurements. Finally, the data acquired with this optimized setup were analysed exploiting different algorithms based on deconvolution and compressive sensing for the volume reconstruction.
Selective volume illumination microscopy è una tecnica in grado di fare imaging volumetrico di campioni su una scala mesoscopica con abbastanza risoluzione da distinguere dettagli sino a livello cellulare di campioni biologici. Questo approccio è basato sull’idea di acquisire più immagini del campione illuminato con diversi foglietti di luce che insieme compongono un set di pattern modulati spazialmente, per poi ricostruire digitalmente l’intero volume. La possibilità di modulare spazialmente l’illuminazione consente di sfruttare una grande varietà di strategie per sampling e ricostruzione, come per esempio il compressive sensing. Tramite questi algoritmi è possibile migliore le perfomance della tecnica di microscopia in termini di tempo totale di misura, di rapporto segnale-rumore e di risoluzione spaziale delle immagini acquisite. Lo scopo di questa tesi è di ottimizzare l’implementazione di questa tecnica lavorando sia sul setup ottico sia su metodi computazionali. Per questi scopi, sono state testate diverse alternative con lo scopo di migliorare le perfomance del ramo di illuminazione del sistema ottico; in particolare il focus è stato di incrementare significativamente la quantità di luce che incide sul campione, in modo di ottenere un miglior rapporto segnale-rumore della misura. Infine, i dati acquisite con questo setup sono stati poi analizzati usando diversi algoritmi basati su deconvoluzione e compressive sensing per la ricostruzione del volume.
Optimization and computational methods for selective volume illumination microscopy
CAMBRIA, FRANCESCO LUCIANO
2019/2020
Abstract
Selective volume illumination microscopy is a technique that achieves volumetric imaging of samples on a mesoscopic scale with enough resolution that it is possible to distinguish details up to a cellular level of the biologic samples. This approach is based on the idea of acquiring multiple images of the sample illuminated with several light sheets that together compose a set of axially modulated patterns, and then to computationally reconstruct the whole volume. The possibility of spatially modulating the illumination paves the way to the exploitation of a variety sampling and reconstruction strategies, such as compressive sensing. Thanks to these approaches, it is possible to enhance the performances of the microscopy technique in terms measurement time, the signal-to-noise ratio and spatial resolution of the acquired images. The goal of this thesis is to optimize the implementation of this technique by working both on the optical setup and on the computational methods. To these purposes, different alternatives were tested with the aim of enhancing the performances of the illumination branch of the microscope; in particular the focus was to improve significatively the amount of light illuminating the sample, thus achieving a better signal to noise ratio of the measurements. Finally, the data acquired with this optimized setup were analysed exploiting different algorithms based on deconvolution and compressive sensing for the volume reconstruction.| File | Dimensione | Formato | |
|---|---|---|---|
|
Tesi _Francesco_Cambria.pdf
accessibile in internet per tutti
Dimensione
2.15 MB
Formato
Adobe PDF
|
2.15 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/170905