Neary-Zajiczek, Lydia;
(2022)
Improving the feasibility of computer-assisted pathology through optimized tissue imaging.
Doctoral thesis (Ph.D), UCL (University College London).
Preview |
Text
Lydia Neary-Zajiczek PhD Thesis_corrected_Redacted.pdf - Accepted Version Download (243MB) | Preview |
Abstract
Diagnostic services are crucial to modern healthcare systems. Rapid processing of pathology results ensure patients enter the appropriate treatment pathway in a timely manner, and intraoperative pathology can maximize the effectiveness of interventions such as resection surgery. Despite a projected increase in demand, almost all pathology departments are severely understaffed; many pathologists are close to retirement and uptake of training places is low, adding further pressure. A proposed mitigation strategy is the digitization of pathology whereby samples are scanned and inspected as digital images, with the development of computer-aided diagnostic tools being the ultimate goal to reduce caseloads. This thesis focuses on two areas of computer-assisted pathology research: challenges preventing wider digitization of pathology, and the limited applicability of intraoperative pathology techniques. Two closely-related issues are high costs and pathologist perception of inadequate digital image quality. These issues have prevented routine use of digital slides, which has in turn hindered the development of automated tools through a lack of annotated data. In this thesis I quantify the minimum image quality necessary for accurate high-level diagnostics, showing that lower-resolution imaging is feasible to reduce costs while maintaining acceptable diagnostic accuracy. Intraoperative pathology currently relies on the time-consuming process of staining and assessing excised tissue while the patient remains under anaesthetic, limiting its applicability due to increased patient risk. I also describe a computer-assisted intraoperative diagnostic tool that combines nanomechanical measurements of tissue properties with a computer vision algorithm to infer the presence of cancerous tissue from an low-resolution image of the sample being assessed without the need for staining.
| Type: | Thesis (Doctoral) |
|---|---|
| Qualification: | Ph.D |
| Title: | Improving the feasibility of computer-assisted pathology through optimized tissue imaging |
| Open access status: | An open access version is available from UCL Discovery |
| Language: | English |
| Additional information: | Copyright © The Author 2022. Original content in this thesis is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Licence (https://creativecommons.org/licenses/by-nc/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request. |
| UCL classification: | UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science UCL > Provost and Vice Provost Offices > UCL BEAMS UCL |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10156239 |
Archive Staff Only
 |
View Item |
