SBIR-STTR Award

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) project is the development of a digital skeletal evaluation workflow for reproductive toxicology studies that will better ensure that potential therapeutics with teratogenic effects will not make it to the marketplace, and at reduced time and cost. The current approach for characterizing the impact of new therapeutics and chemicals on skeletal development relies upon human-based qualitative characterization, while the goal for this digital and quantitative approach will be to more accurately characterize the effect of therapeutics. An improved skeletal characterization workflow in this $2.5 billion toxicology market also will allow for the reduced usage of animals. The imaging and characterization platform developed through this project will be commercialized as a service where contract research organizations or pharmaceutical companies may send fetal skeletons for analysis. Alternatively, customers may purchase the platform. This biphasic business model will allow small contract research organizations to outsource labor intensive characterization and large contract research organizations to switch to this new digital approach. This SBIR Phase I project proposes to optimize a prototype optical coherence tomography (CT) scanner for the 3D digitization of rodent skeletons at high resolution. Using this system, rodent skeletons will be transformed into three-dimensional data matrices that can be evaluated digitally by a fetal pathologist or characterized quantitively by a computer for defects. The illumination and imaging parameters of the optical scanner will be systematically optimized to reduce noise and to enable the clear differentiation between different skeletal features. Once rodent skeletons can be imaged in a reproducible manner to create three-dimensional data sets, these data sets will be compared quantitatively side-by-side to data sets acquired from the same animals using more expensive X-ray CT imaging. Following imaging optimization and validation, a digital analysis program will be developed that will determine if bones in a rodent model deviate from control samples. This program will begin to allow for the complete digital evaluation of rodent skeletons for developmental and reproductive toxicology studies.

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project will be the development of technology to improve the accuracy of skeletal evaluation within developmental and reproductive toxicology (DART) studies. The goal is to better ensure that potential therapeutics, cosmetics and agrochemicals do not cause teratogenic effects. Today, DART studies rely upon the subjective human-based manual evaluation of animal skeletons for defects, which has a low sensitivity for defects, significant inter and intra pathologist variability, and is laborious and costly. The technology under development is based on a novel imaging and automated analysis solution for this problem that will shift the paradigm of skeletal evaluation from a qualitative to quantitative approach. Through improving DART study accuracy, the objective is to better detect teratogenic effects of compounds, reduce the overall number of animals required for these studies, and reduce the cost to develop therapeutics by improving throughput and reducing study cost. The market opportunity for this technology is expected to be significant. The intellectual merit of this SBIR Phase II project is to focus on the development and optimization of an optical CT imaging device and analysis software for use with mouse, rat and rabbit fetal samples for skeletal evaluation. The specimens will be processed such that they are optically transparent with bones that are stained red. A training library of normal and abnormal fetal samples will be generated, and from this library a machine learning-based approach will be developed to automatically identify samples that are non-normal in a statistically significant manner. To achieve this, several classification methodologies will be evaluated quantitatively for accuracy and the image acquisition parameters will be optimized for imaging quality. From this work, a 21 CFR part 11 compliant software application will be developed in accordance with the ICH analytical assay guidelines such that this software can undergo IQ/OQ/PQ, which will allow for the hardware and software system to be implemented by customers in their GLP facilities. The hardware and software product that will result from this project will be one of the first validated digital pathology platforms in the marketplace, and will ultimately allow for customers to significantly reduce their operating costs while improving accuracy. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.