Cervical cancer is the second most common cause of cancer in women worldwide and the leading cause of cancer related mortality in women in developing countries. We have developed real time non-invasive point-of-care devices to detect early cancerous conditions of cervix. Our preliminary analyses on data collected from a total of 133 patients and have shown promising results. We are continuing to collect data and will include the entire data set into the analysis as part of our phase I objective. Given our limited data set, however, it is not possible to project any definite performance benchmark. We will propose this task to the training and validation phase in a Phase 2 proposal. For the present we want to analyze the data for the purpose of determining the optimum clinical device for the validation phase. To support data accrual in the validation phase we will make multiple clones of this device for deployment in a multicenter clinical trial. The objective of the data analysis is to arrive at this validation device by simplifying and consolidating the complexity and the many features built into each prototype. This will also reduce the possibility of overtraining by reducing the number of parameters we build into our algorithm.
The product of this research will be a prototype device capable to spectroscopically detecting and/or diagnosing cervical cancer. Our device has, upon commercialization, the potential to replace or aid the current Pap test as well as to replace the current methods of colposcopy, biopsy and histology. The target physician group for the cervical application is the Ob-Gyn segment. It is estimated that there are at least 30,000 physicians in this segment and upwards of $6 billion spent each year in the US on cervical disease.
