Assessment of Sexual Assault Kit (SAK) Evidence Selection Leading to Development of SAK Evidence Machine-Learning Model (SAK-ML Model)

The purpose of this study is two-fold. The first purpose is to evaluate crime laboratories’ decision-making protocols on the selection of sexual assault kit DNA evidence to develop research-based guidelines on how many and which swabs should have completed DNA analysis. To achieve the first purpose of the study, data will be collected from three publicly funded forensic laboratories on DNA analysis findings, and from corresponding forensic nursing teams on variables found to be predictive of developing CODIS eligible profiles.

The second purpose of the study is to create and implement a software program, Sexual Assault Kit evidence Machine Learning Model (SAK-ML Model), to generate probabilities of developing CODIS-eligible DNA profiles from sexual assault kit swabs. The SAK-ML Model will guide forensic scientists in the selection of the most probative swabs to analyze.

Development of the SAK-ML Model is possible, as the Utah Bureau of Forensic Services and the partnering Brigham Young University research team (Julie Valentine, PI) already have compiled a large dataset on DNA findings from sexual assault kits. Utah will enact statewide forensic electronic medical records beginning in July 2019, which will allow for automatic generation of the SAK-ML Model. Implementation of the SAK-ML Model will enhance crime laboratory efficiency and improve DNA analysis outcomes from sexual assault kits.

Note: This project contains a research and/or development component, as defined in applicable law, and complies with Part 200 Uniform Requirements - 2 CFR 200.210(a)(14).

CA/NCF