Spatial frequency domain imaging towards improved detection of gastrointestinal cancers
Early detection and treatment of gastrointestinal cancers has been shown to drastically improve patients survival rates. However, wide population based screening for gastrointestinal cancers is not feasible due to its high cost, risk of potential complications, and time consuming nature. This thesis...
| Main Author: | |
|---|---|
| Format: | Thesis (University of Nottingham only) |
| Language: | English |
| Published: |
2023
|
| Online Access: | https://eprints.nottingham.ac.uk/76713/ |
| _version_ | 1848800930341847040 |
|---|---|
| author | Crowley, Jane |
| author_facet | Crowley, Jane |
| author_sort | Crowley, Jane |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | Early detection and treatment of gastrointestinal cancers has been shown to drastically improve patients survival rates. However, wide population based screening for gastrointestinal cancers is not feasible due to its high cost, risk of potential complications, and time consuming nature. This thesis forms the proposal for the development of a cost-effective, minimally invasive device to return quantitative tissue information for gastrointestinal cancer detection in-vivo using spatial frequency domain imaging (SFDI). SFDI is a non-invasive imaging technique which can return close to real time maps of absorption and reduced scattering coefficients by projecting a 2D sinusoidal pattern onto a sample of interest. First a low-cost, conventional bench top system was constructed to characterise tissue mimicking phantoms. Phantoms were fabricated with specific absorption and reduced scattering coefficients, mimicking the variation in optical properties typically seen in healthy, cancerous, and pre-cancerous oesophageal tissue. The system shows accurate retrieval of absorption and reduced scattering coefficients of 19% and 11% error respectively. However, this bench top system consists of a bulky projector and is therefore not feasible for in-vivo imaging. For SFDI systems to be feasible for in-vivo imaging, they are required to be miniaturised. Many conditions must be considered when doing this such as various illumination conditions, lighting conditions and system geometries. Therefore to aid in the miniaturisation of the bench top system, an SFDI system was simulated in the open-source ray tracing software Blender, where the capability to simulate these conditions is possible. A material of tunable absorption and scattering properties was characterised such that the specific absorption and reduced scattering coefficients of the material were known. The simulated system shows capability in detecting optical properties of typical gastrointestinal conditions in an up-close, planar geometry, as well in a non-planar geometry of a tube simulating a lumen. Optical property imaging in the non-planar, tubular geometry was done with the use of a novel illumination pattern, developed for this work. Finally, using the knowledge gained from the simulation model, the bench top system was miniaturised to a 3 mm diameter prototype. The novel use of a fiber array producing the necessary interfering fringe patterns replaced the bulky projector. The system showed capability to image phantoms simulating typical gastrointestinal conditions at two wavelengths (515 and 660 nm), measuring absorption and reduced scattering coefficients with 15% and 6% accuracy in comparison to the bench top system for the fabricated phantoms. It is proposed that this system may be used for cost-effective, minimally invasive, quantitative imaging of the gastrointestinal tract in-vivo, providing enhanced contrast for difficult to detect cancers. |
| first_indexed | 2025-11-14T20:59:23Z |
| format | Thesis (University of Nottingham only) |
| id | nottingham-76713 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-14T20:59:23Z |
| publishDate | 2023 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-767132024-02-06T10:27:01Z https://eprints.nottingham.ac.uk/76713/ Spatial frequency domain imaging towards improved detection of gastrointestinal cancers Crowley, Jane Early detection and treatment of gastrointestinal cancers has been shown to drastically improve patients survival rates. However, wide population based screening for gastrointestinal cancers is not feasible due to its high cost, risk of potential complications, and time consuming nature. This thesis forms the proposal for the development of a cost-effective, minimally invasive device to return quantitative tissue information for gastrointestinal cancer detection in-vivo using spatial frequency domain imaging (SFDI). SFDI is a non-invasive imaging technique which can return close to real time maps of absorption and reduced scattering coefficients by projecting a 2D sinusoidal pattern onto a sample of interest. First a low-cost, conventional bench top system was constructed to characterise tissue mimicking phantoms. Phantoms were fabricated with specific absorption and reduced scattering coefficients, mimicking the variation in optical properties typically seen in healthy, cancerous, and pre-cancerous oesophageal tissue. The system shows accurate retrieval of absorption and reduced scattering coefficients of 19% and 11% error respectively. However, this bench top system consists of a bulky projector and is therefore not feasible for in-vivo imaging. For SFDI systems to be feasible for in-vivo imaging, they are required to be miniaturised. Many conditions must be considered when doing this such as various illumination conditions, lighting conditions and system geometries. Therefore to aid in the miniaturisation of the bench top system, an SFDI system was simulated in the open-source ray tracing software Blender, where the capability to simulate these conditions is possible. A material of tunable absorption and scattering properties was characterised such that the specific absorption and reduced scattering coefficients of the material were known. The simulated system shows capability in detecting optical properties of typical gastrointestinal conditions in an up-close, planar geometry, as well in a non-planar geometry of a tube simulating a lumen. Optical property imaging in the non-planar, tubular geometry was done with the use of a novel illumination pattern, developed for this work. Finally, using the knowledge gained from the simulation model, the bench top system was miniaturised to a 3 mm diameter prototype. The novel use of a fiber array producing the necessary interfering fringe patterns replaced the bulky projector. The system showed capability to image phantoms simulating typical gastrointestinal conditions at two wavelengths (515 and 660 nm), measuring absorption and reduced scattering coefficients with 15% and 6% accuracy in comparison to the bench top system for the fabricated phantoms. It is proposed that this system may be used for cost-effective, minimally invasive, quantitative imaging of the gastrointestinal tract in-vivo, providing enhanced contrast for difficult to detect cancers. 2023-12-14 Thesis (University of Nottingham only) NonPeerReviewed application/pdf en cc_by https://eprints.nottingham.ac.uk/76713/1/thesis_31102023.pdf Crowley, Jane (2023) Spatial frequency domain imaging towards improved detection of gastrointestinal cancers. PhD thesis, University of Nottingham. |
| spellingShingle | Crowley, Jane Spatial frequency domain imaging towards improved detection of gastrointestinal cancers |
| title | Spatial frequency domain imaging towards improved detection of gastrointestinal cancers |
| title_full | Spatial frequency domain imaging towards improved detection of gastrointestinal cancers |
| title_fullStr | Spatial frequency domain imaging towards improved detection of gastrointestinal cancers |
| title_full_unstemmed | Spatial frequency domain imaging towards improved detection of gastrointestinal cancers |
| title_short | Spatial frequency domain imaging towards improved detection of gastrointestinal cancers |
| title_sort | spatial frequency domain imaging towards improved detection of gastrointestinal cancers |
| url | https://eprints.nottingham.ac.uk/76713/ |