Objective Image Analysis for the Interpretation of Colorimetric Tests for Drugs of Abuse

The objective of this project is to develop an image analysis method to objectively interpret the color result of three colorimetric tests for drugs of abuse.

Colorimetric tests were developed in the 1970s for presumptive drug identification by law enforcement in the field. Colorimetric tests offer several advantages over traditional analytical techniques, including low cost, portability, and ease of use. As recommended by the Scientific Working Group for the Analysis of Seized Drugs (SWGDRUG), forensic laboratories currently follow a multistep scheme for drug identification, which can include colorimetric tests [1,2]. However, a disadvantage of colorimetric tests is the subjective visual interpretation of the resulting color change. To address this limitation, a numerical representation of the resulting color change can be used to objectively interpret the color test result. Color can be represented using various color models, such as RGB (Red, Green, Blue) or HSV (Hue, Saturation, Value). Photographs of the colorimetric reaction can be acquired using readily available imaging devices, such as a mobile phone or DSLR (digital single-lens reflex) camera. The color information can then be extracted from the photos collected to obtain a numerical representation of color. The numerical color representation can then be used to establish thresholds that objectively define a positive, negative, or inconclusive colorimetric test result.

This project will focus on developing an objective image analysis scheme for interpreting the results of several colorimetric tests for drugs of abuse. The colorimetric tests will be performed on spot plates or test tubes using solution drug standards and known drug samples (if available). Photos of the resulting color reaction will be taken using an imaging device (e.g., mobile phone, DSLR camera). The photos will be processed to account for variations in the imaging device used and the lighting scenarios. A numerical representation of the resulting color will be extracted from the processed photos using a color model (e.g., RGB, HSV). Thresholds will be established for the objective determination of a positive, negative, or inconclusive colorimetric test result. Finally, the established thresholds will be evaluated using blind drug samples (if available) to estimate the Type I and Type II error rates. If authentic drug samples are unavailable, this work will focus on solution drug standards and will serve as a proof-of-concept for the objective interpretation of colorimetric tests.