Catherine Poh

Passive endoscopic polarization sensitive optical coherence tomography with completely fiber based optical components

Polarization Sensitive Optical Coherence Tomography (PSOCT) is a functional extension of Optical Coherence Tomography (OCT) that is sensitive to well-structured, birefringent tissue such as scars, smooth muscle and cartilage. In this work, we present a novel completely fiber based swept source PSOCT system using a fiber-optic rotary pullback catheter. This PSOCT implementation uses only passive optical components and requires no calibration while adding minimal additional cost to a standard structural OCT imaging system. Due to its complete fiber construction, the system can be made compact and robust, while the fiber-optic catheter allows access to most endoscopic imaging sites. The 1.5mm diameter endoscopic probe can capture 100 frames per second at pullback speeds up to 15 mm/s allowing rapid traversal of large imaging fields. We validate the PSOCT system with known birefringent tissues and demonstrate in vivo PSOCT imaging of human oral scar tissue.

Automated segmentation of oral mucosa from wide-field OCT images (Conference Presentation)

Optical Coherence Tomography (OCT) can discriminate morphological tissue features important for oral cancer detection such as the presence or absence of basement membrane and epithelial thickness. We previously reported an OCT system employing a rotary-pullback catheter capable of in vivo, rapid, wide-field (up to 90 x 2.5mm2) imaging in the oral cavity. Due to the size and complexity of these OCT data sets, rapid automated image processing software that immediately displays important tissue features is required to facilitate prompt bed-side clinical decisions. We present an automated segmentation algorithm capable of detecting the epithelial surface and basement membrane in 3D OCT images of the oral cavity. The algorithm was trained using volumetric OCT data acquired in vivo from a variety of tissue types and histology-confirmed pathologies spanning normal through cancer (8 sites, 21 patients). The algorithm was validated using a second dataset of similar size and tissue diversity. We demonstrate application of the algorithm to an entire OCT volume to map epithelial thickness, and detection of the basement membrane, over the tissue surface. These maps may be clinically useful for delineating pre-surgical tumor margins, or for biopsy site guidance.

Novel computational image analysis to predict regional nodal disease for early-stage oral cancer

Objective: Nodal disease (N+) for early-stage oral squamous cell carcinoma (OSCC) is the most significant prognostic factor for survival. There is a lack of effective predictor to justify prophylactic neck dissection. Quantitative tissue pathology (QTP) has shown its promise in providing an objective means for diagnosis and prognosis of many cancer types. We conducted a pilot study on the utilization of QTP to evaluate risk of nodal disease. Study design: Retrospective case-control study Subjects and methods: Histological sections from 15 primary tumors of clinically node-negative (N0) patients were stained with Feulgen-Thionin followed by acquisition of digital images and image processing to measure the mean and variance of nuclear phenotype and tissue architecture features from 45,253 nuclei of 45 tumor nests. Association between features and nodal disease outcome (N0 or N+) was investigated using nested analysis of variance adjusted by patient. Ability to discriminate between N0 and N+ was analyzed using multivariate logistic regression and receiver operating characteristics (ROC) curve analysis. P-value<0.05 (2-sided) was considered significant. Results: The N+ group presented higher mean values of chromatin condensation levels and cell density compared to those of the N0 group. ROC curve showed a strong discriminative ability of chromatin condensation levels between the N+ and N0 groups with a sensitivity and specificity of 1.0 and 0.75, respectively. Conclusion: This study reports the first-ever data on QTP as a risk assessment tool for nodal disease in early-stage OSCCs. Such computational imaging analysis potentially provides a new objective approach to predict regional nodal disease. Correspondence to: Dr. Catherine F. Poh, Department of Oral Medical and Biological Sciences, Faculty of Dentistry, The University of British Columbia, 2199 Wesbrook Mall, Vancouver BC, V6T 1Z3, Canada; E-mail: cpoh@dentistry.ubc.ca

Towards biopsy guidance of oral lesions with wide-field OCT imaging

We address two issues required for OCT biopsy guidance in the oral cavity: 1) automated segmentation of wide-field OCT images, and 2) registering imaging location with fiducial markers placed on the tissue.

Abstract A13: Quantitative pathology toolbox: Improvement in prediction of progression risk for oral premalignant lesions using both interactive and automated image analysis

For the promise of Quantitative Pathology (QP) to be clinically accepted it must provide new information that is clinically usable (and hence actionable). This must be accomplished in a fashion that is not onerous (expensive/time-consuming/require specialized training) to the pathologist relative to the value of the information provided. As an example a major barrier to oral cancer prevention is the inability to predict progression risk for oral premalignant lesions by conventional pathology alone that can be addressed by QP. We present two approaches for the quantitative analysis of FFPE sectioned oral tissue. In one approach the user selects the area of epithelium to be analyzed and visually filters the cells to be analyzed, the other is an automated approach in which the user only circles the epithelium to be analyzed. In both approaches the cell nuclei are automatically segmented, 110 features per nuclei calculated and used to determine how normal or cancer like the nucleus is (and in the fully automated approach if the objects are single intact nucleus or not), then the distribution of nuclei values within the area of interest is used to generate a Quantitative Pathology Scores (QPS) for the tissue. These tissue measures were used alone or in combination with other markers to perform risk assessment in patients from a very large oral cancer prediction longitudinal study. Also the scores can be combined with other risk markers such as Loss of Heterozygosity (LOH) analysis to improve risk stratification. A combination of LOH based predictors and QPS thresholds were trained to refine three previously validated LOH defined- risk groups. The combined model defined a low, a medium and a high risk of progression to cancer categories. For the 104 low risk cases so classified, 98.1% do not progress to cancer (used to define a relative risk [RR] of 1). In contrast, 15% of the 106 classified medium risk cases (RR= 7.85) and 65% of the 26 high-risk cases (RR = 34) progress. This is a substantial improvement over just the LOH based classification and significantly better than dysplasia grade for risk prediction. In a validation set of 43 mild to moderate dysplasia cases with long term follow-up, 100% of the 23 cases classified as low risk by the combined algorithm did not progress, 43% of the 7 cases classified as medium risk by the combined algorithm progressed and 92.3% (12 out 13) of the cases classified as high risk by the combined algorithm progressed. These validation results strongly support the combination of these approaches for facilitating risk prediction and improving patient management. This combined risk model is also a suitable intermediate endpoint biomarker of transformation risk for oral tissue and is being used in multicenter (8) Canadian Optically guides Oral Cancer Surgical Trial as part of the quantitative evaluation of surgical margin tissue. This work was funded by the NIDCR, NIH and by the TFRI. Note: This abstract was withdrawn after the Proceedings were printed and, therefore, was not presented at the conference. Citation Format: Calum MacAulay, Miriam Rosin, Lewei Zhang, Catherine Poh, Michele Williams, Martial Guillaud. Quantitative pathology toolbox: Improvement in prediction of progression risk for oral premalignant lesions using both interactive and automated image analysis. [abstract]. In: Proceedings of the Thirteenth Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2014 Sep 27-Oct 1; New Orleans, LA. Philadelphia (PA): AACR; Can Prev Res 2015;8(10 Suppl): Abstract nr A13.

Oblique polarized reflectance spectroscopy for depth sensitive measurements in the epithelial tissue

Optical spectroscopy has shown potential as a tool for precancer detection by discriminating alterations in the optical properties within epithelial tissues. Identifying depth-dependent alterations associated with the progression of epithelial cancerous lesions can be especially challenging in the oral cavity due to the variable thickness of the epithelium and the presence of keratinization. Optical spectroscopy of epithelial tissue with improved depth resolution would greatly assist in the isolation of optical properties associated with cancer progression. Here, we report a fiber optic probe for oblique polarized reflectance spectroscopy (OPRS) that is capable of depth sensitive detection by combining the following three approaches: multiple beveled fibers, oblique collection geometry, and polarization gating. We analyze how probe design parameters are related to improvements in collection efficiency of scattered photons from superficial tissue layers and to increased depth discrimination within epithelium. We have demonstrated that obliquely-oriented collection fibers increase both depth selectivity and collection efficiency of scattering signal. Currently, we evaluate this technology in a clinical trial of patients presenting lesions suspicious for dysplasia or carcinoma in the oral cavity. We use depth sensitive spectroscopic data to develop automated algorithms for analysis of morphological and architectural changes in the context of the multilayer oral epithelial tissue. Our initial results show that OPRS has the potential to improve the detection and monitoring of epithelial precancers in the oral cavity.

Abstract PL05-04: From basic science to public health: The Canadian approach to oral leukoplakia

Oral cancer is a major health problem worldwide (over 300,000 new cases identified each year) with numbers projected to climb steadily. Although the site is easily accessed, it is most often detected at late-stage, leading to high mortality and morbidity rates that have shown little change for decades. Much of the projected increase in this disease will occur in low- and middle-income countries with very limited resources. Oral cancer is a good target for employing systems models for prevention and intervention. It shares etiology with many cancers (tobacco usage, alcohol consumption, diet, and infection) so prevention approaches have a universal application. Common challenges exist in both developing and wealthier countries: low awareness of health care professionals, low public awareness, low priority within public health, inequities in care, infrastructure and human resources deficits, system fragmentation and lack of coordinated advocacy and leadership. Approaches tend to be treatment rather than prevention focused. There is a need for development of new strategies for management of the disease. The objective of this presentation is to describe a population health-focused approach taken in British Columbia to improve oral cancer control that has drawn together provincial resources, integrating them into a comprehensive system of structures and players. The effort spans the continuum of care (prevention, early detection, diagnosis and treatment of the disease) working through multiple approaches, at multiple levels and across multiple sectors. Its growth is guided through partnerships and networks with health professionals and societies, by social marketing and epidemiological surveillance, and by technology development for early detection, risk assessment and cancer treatment within an ongoing NIH-funded longitudinal study. The approach has grown around an opportunistic screening network of dentist and ENT surgeons that has existed in British Columbia for ∼27 years. This network is linked through a centralized oral biopsy service to referral resources to facilitate patient management. We have engaged and empowered this network through a combination of approaches (focus groups of community dental health professionals, continuing education forums, news releases, journal publications, and social marketing strategies) with an aim to increase awareness of the importance of screening activity and to build social norms among dental professionals to enhance their ownership of this domain. In collaboration with the College of Dental Surgeons of British Columbia, a Guideline was released in 2008 to move activities from an ad hoc to a standardized, quality controlled approach, with formal integration of screening into the regular oral exam for each patient. This has resulted in increased screening among dental professionals who have seen the guideline, among whom 84% screen as part of routine exams, compared to 65% of those not aware of the guideline. A second offshoot of activity within the program is the development within the Oral Biopsy Service of an Oral Leukoplakia and Dysplasia Registry that will link to the British Columbia Cancer Registry to track the impact of changes in practice on outcomes. This project is supported by the Canadian Partnership Against Cancer (CPAC) with input from other provincial oral biopsy services across Canada. This will allow for development of parallel dental network and registry structures to better define the natural history of the disease and improve its management at a national level. We expect to reach 2/3 of the BC population through a strengthened dental office network and referral pathway. However, access to this network remains a challenge for the remaining population. Tailored strategies are being developed in partnership with the South Asian community, the Vancouver First Nations community, the Downtown Eastside in Vancouver, and rural BC. Since developing countries share many of the same barriers to health care utilization with these underserved communities, we have opened dialogues with international partners to gain input and provide knowledge exchange around facilitators and barriers to screening activities in low-resource settings. This increased screening activity has been coupled to the development of a streamlined referral pathway for risk assessment in dysplasia clinics for standardized risk assessment and follow-up of high-risk patients (an NIH-NIDC-funded initiative). These clinics use technology developed within an ongoing longitudinal study to identify and assess high-risk lesions: optical devices to improve clinical visualization; high-throughput computer microscopy systems to predict future behavior of early lesions in a cost-effective fashion, using subtle histological and cytological indicators; and molecular markers of risk. The impact of this technology on driving change in clinical practice is already apparent. Optical fluorescence technology has been used to detect occult disease in patients going to surgery. Its use to guide surgical margin delineation has resulted in a marked reduction in recurrence rates ∼10 fold or more (from ∼25% recurrence to close to zero) that is now being explored in multicenter randomized control trial for device efficacy supported by the Terry Fox Research Institute. This initiative promises to facilitate the better management of early disease detected within the system, and provides a structure for rigorous assessment of these innovations as well as a foundation for knowledge translation and scale-up. Conclusion: We are engaged in trials of innovative technology, social mobilization, and knowledge translation to support system-wide change for equitable population-based control of oral cancer in British Columbia. This effort is a model that demonstrates how resources can be pulled together and optimized to provide a centralized structure for shaping new approaches to detection, management, and prevention of disease. Citation Information: Cancer Prev Res 2010;3(12 Suppl):PL05-04.