|Kim, Huisung - Purdue University|
|Awofeso, Olumide - Purdue University|
|Choi, Somi - Purdue University|
|Jung, Youngkee - Purdue University|
|Bae, Euiwon - Purdue University|
Submitted to: Applied Optics
Publication Type: Review Article
Publication Acceptance Date: 11/28/2016
Publication Date: 12/23/2016
Citation: Kim, H., Awofeso, O., Choi, S., Jung, Y., Bae, E. 2016. Colorimetric analysis of saliva–alcohol test strips by smartphone-based instruments using machine-learning algorithms. Applied Optics. doi: org/10.1364/AO.56.000084.
Technical Abstract: Strip lateral flow assays, similar to a home pregnancy test, are used widely in food safety applications to provide rapid and accurate tests for the presence of specific foodborne pathogens or other contaminants. Though these tests are very rapid, they are not very sensitive, are not quantitative, and are typically interpreted by eye creating a dependency of interpretation on the observer’s visual acuity and color perception. A smartphone-based strip analysis device was developed that can accurately detect the contaminant concentration from the color change of a lateral flow assay strip leveraging the most recently developed machine learning algorithm. Combining a choice of color-space and machine learning, minute color changes were accurately classified with an average of 80-85% correct classification. Application of these algorithms can help scientists to expand this method to any color-changing assay with high accuracy and reproducible interpretation, increasing the reliability and mobility of these rapid diagnostic tests.