Spectroscopy and Artificial Intelligence to disrupt the status quo in Cervical Cancer Screening , Canada
Objective: Cervical cancer, a preventable disease, kills one woman every two minutes. Current screening programs, where available, involve methods that are invasive, time consuming and require skilled clinicians and costly infrastructure. Without advancing our current screening methods, it is estimated that the incidence of cervical cancer will continue to increase. This research team is accepting a challenge set forth by WHO to propose ways to help eliminate this preventable disease with the aim of reducing the incidence of cervical cancer to a point where it would no longer be considered a public health problem. The current strategy for screening involves conventional cervical cytology, which relies heavily on repeated testing at frequent intervals. Urine HPV testing has been evaluated in limited studies and has similar diagnostic accuracy to cervical samples and even better, urine may be more acceptable and easier to obtain. Although this testing is still expensive, sometimes too expensive and technologically challenging. Spectroscopy of biofluids is a relatively new technology being evaluated for clinical applications. This technology facilitates assessment of biofluid to determine disease biomarkers and studies have shown high sensitivity and specificity in endometrial cancers. Early proof-of-concept studies have demonstrated similar potential in urine testing for gynaecological cancers. Spectroscopy is a robust, portable, relatively simple and inexpensive technique. The results are instant, reproducible and reliable. Data from spectroscopy could then be used to train artificial neural networks (ANNs) to accurately identify clinically relevant disease precursors. These platforms have the capacity to be miniaturized and connect to the cloud for data acquisition and storage. To date there is no published experience using spectroscopy to identify HPV in urine, however members of this group will be assessing genotyped urine samples with cervical matching stored at IARC biobank in the near future at Prof. Rehman’s lab (personal Communication P Basu, IARC). To date there is no published experience using spectroscopy to identify HPV in urine, however members of this group will be assessing genotyped urine samples with cervical matching stored at IARC biobank in the near future at Prof. Rehman’s lab (personal Communication P Basu, IARC). We propose to investigate the ability of spectroscopy to recognise HPV in urine and to compare performance metrics of the test (sensitivity, specificity) to the current validated HPV test. Once the spectroscopy urine test is shown to be non-inferior to a validated gold standard HPV test, we will use the spectroscopy signals to develop artificial neural networks (ANNs) trained to recognize the presence of HPV in urine and identify cervical intra-epithelial neoplasia 2+, the clinically recognized precursor of cervical cancer.
|Registered Biobank Name||Spectroscopy and Artificial Intelligence to disrupt the status quo in Cervical Cancer Screening|
|Biobank Leader||Janet Slaunwhite|
|Email for biobank firstname.lastname@example.org|
|Principal Investigator||James Bentley|
||Mono - Collection aimed at supporting a specific, single research project|