JDS Journal of Data Science 1683-86021680-743X1680-743X School of Statistics, Renmin University of China JDS1103 10.6339/23-JDS1103 Data Science in Action A Video Segmentation Pipeline for Assessing Changes in Pupil Response to Light After Cannabis Consumption SteinhartBenjamin1 Brooks-RussellAshleyPhD2 KosnettMichael J.MD, MPH34 SubramanianPrem S.MD, PhD56 WrobelJuliaPhDjulia.wrobel@cuanschutz.edu1∗ Department of Biostatistics and Informatics, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, Colorado, USA Injury and Violence Prevention Center, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, Colorado, USA Department of Medicine, CU School of Medicine, University of Colorado Anschutz Medical Campus, Aurora, Colorado, USA Department of Environmental and Occupational Health, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, Colorado, USA Departments of Ophthalmology, Neurology, and Neurosurgery, Sue Anschutz-Rodgers Eye Center, CU School of Medicine, University of Colorado Anschutz Medical Campus, Aurora, Colorado, USA Department of Surgery, Uniformed Services University of the Health Sciences, Bethesda, Maryland, USA Corresponding author. Email: julia.wrobel@cuanschutz.edu. 20242062023221138151 Supplementary Material

The supplementary materials contain a zipped folder with that is also available on GitHub at https://github.com/benjamin643/pupil_segmentation_and_analysis. This folder contains code to reproduce all analyses in this paper, one example video, and data for reproducing all results post pupil segmentation.

1022023252023 2024 The Author(s). Published by the School of Statistics and the Center for Applied Statistics, Renmin University of China.2024 Open access article under the CC BY license.

Due to long-standing federal restrictions on cannabis-related research, the implications of cannabis legalization on traffic and occupational safety are understudied. Accordingly, there is a need for objective and validated measures of acute cannabis impairment that may be applied in public safety and occupational settings. Pupillary response to light may offer an avenue for detection that outperforms typical sobriety tests and tetrahydrocannabinol concentrations. We developed a video processing and analysis pipeline that extracts pupil sizes during a light stimulus test administered with goggles utilizing infrared videography. The analysis compared pupil size trajectories in response to a light for those with occasional, daily, and no cannabis use before and after smoking. Pupils were segmented using a combination of image pre-processing techniques and segmentation algorithms which were validated using manually segmented data and found to achieve 99% precision and 94% F-score. Features extracted from the pupil size trajectories captured pupil constriction and rebound dilation and were analyzed using generalized estimating equations. We find that acute cannabis use results in less pupil constriction and slower pupil rebound dilation in the light stimulus test.

 image processing marijuana impairment quantile GAMS time series data cleaning Colorado Department of Public Health and EnvironmentNational Institutes of HealthR01 DA049800The authors gratefully acknowledge the Colorado Department of Public Health and Environment (Co-PI: Brooks-Russell, Kosnett) and the National Institutes of Health (R01 DA049800). References Beucher S, Meyer F (2018). The morphological approach to segmentation: The watershed transformation. In: Mathematical Morphology in Image Processing, 433–481. CRC Press. Bradski G (2000). The OpenCV Library. Dr. Dobb’s Journal of Software Tools. Brooks-Russell A, Brown T, Friedman K, Wrobel J, Schwarz J, Dooley G, et al. (2021). Simulated driving performance among daily and occasional cannabis users. Accident Analysis and Prevention, 160: 106326. https://doi.org/10.1016/j.aap.2021.106326 Brown B, Adams AJ, Haegerstrom-Portnoy G, Jones RT, Flom MC (1977). Pupil size after use of marijuana and alcohol. American Journal of Ophthalmology, 83(3): 350–354. https://doi.org/10.1016/0002-9394(77)90732-2 Brubacher JR, Chan H, Erdelyi S, Macdonald S, Asbridge M, Mann RE, et al. (2019). Cannabis use as a risk factor for causing motor vehicle crashes: A prospective study. Addiction, 114(9): 1616–1626. https://doi.org/10.1111/add.14663 Campobasso CP, De Micco F, Corbi G, Keller T, Hartung B, Daldrup T, et al. (2020). Pupillary effects in habitual cannabis consumers quantified with pupillography. Forensic Science International, 317: 110559. https://doi.org/10.1016/j.forsciint.2020.110559 Colizzi M, Bhattacharyya S (2018). Cannabis use and the development of tolerance: A systematic review of human evidence. Neuroscience and Biobehavioral Reviews, 93: 1–25. https://doi.org/10.1016/j.neubiorev.2018.07.014 Ding L, Goshtasby A (2001). On the canny edge detector. Pattern Recognition, 34(3): 721–725. https://doi.org/10.1016/S0031-3203(00)00023-6 Downey LA, King R, Papafotiou K, Swann P, Ogden E, Boorman M, et al. (2012). Detecting impairment associated with cannabis with and without alcohol on the standardized field sobriety tests. Psychopharmacology, 224(4): 581–589. https://doi.org/10.1007/s00213-012-2787-9 Fant RV, Heishman SJ, Bunker EB, Pickworth WB (1998). Acute and residual effects of marijuana in humans. Pharmacology, Biochemistry and Behavior, 60(4): 777–784. https://doi.org/10.1016/S0091-3057(97)00386-9 Fasiolo M, Wood SN, Zaffran M, Nedellec R, Goude Y (2021a). Fast calibrated additive quantile regression. Journal of the American Statistical Association, 116(535): 1402–1412. https://doi.org/10.1080/01621459.2020.1725521 Fasiolo M, Wood SN, Zaffran M, Nedellec R, Goude Y (2021b). qgam: Bayesian nonparametric quantile regression modeling in R. Journal of Statistical Software, 100(9): 1–31. https://doi.org/10.18637/jss.v100.i09 Fitzgibbon AW, Fisher RB, et al. (1996). A buyer’s guide to conic fitting. Citeseer. Halekoh U, Højsgaard S, Yan J (2006). The r package geepack for generalized estimating equations. Journal of Statistical Software, 15(2): 1–11. https://doi.org/10.18637/jss.v015.i02 Haralick RM, Sternberg SR, Zhuang X (1987). Image analysis using mathematical morphology. IEEE Transactions on Pattern Analysis and Machine Intelligence, 4: 532–550. https://doi.org/10.1109/TPAMI.1987.4767941 Hartman RL, Richman JE, Hayes CE, Huestis MA (2016). Drug recognition expert (dre) examination characteristics of cannabis impairment. Accident Analysis and Prevention, 92: 219–229. https://doi.org/10.1016/j.aap.2016.04.012 IACP (2022). The 12-step dre protocol. https://www.theiacp.org/12-step-process. Accessed: 2022-05-30. Lamprecht MR, Sabatini DM, Carpenter AE (2007). Cellprofiler: Free, versatile software for automated biological image analysis. BioTechniques, 42(1): 71–75. https://doi.org/10.2144/000112257 Larson MD, Behrends M (2015). Portable infrared pupillometry: A review. Anesthesia and Analgesia, 120(6): 1242–1253. https://doi.org/10.1213/ANE.0000000000000314 Lira MC, Heeren TC, Buczek M, Blanchette JG, Smart R, Pacula RL, et al. (2021). Trends in cannabis involvement and risk of alcohol involvement in motor vehicle crash fatalities in the united states, 2000–2018. American Journal of Public Health, 111(11): 1976–1985. https://doi.org/10.2105/AJPH.2021.306466 Logan B, Kacinko SL, Beirness DJ (2016). An evaluation of data from drivers arrested for driving under the influence in relation to per se limits for cannabis. McKay RE, Neice AE, Larson MD (2018). Pupillary unrest in ambient light and prediction of opioid responsiveness: Case report on its utility in the management of 2 patients with challenging acute pain conditions. A&A Practice, 10(10): 279–282. https://doi.org/10.1213/XAA.0000000000000710 Merzouki A, Mesa JM, Louktibi A, Kadiri M, Urbano G (2008). Assessing changes in pupillary size in rifian smokers of kif (cannabis sativa l.). Journal of forensic and Legal Medicine, 15(5): 335–338. https://doi.org/10.1016/j.jflm.2007.08.001 Newmeyer MN, Swortwood MJ, Taylor ME, Abulseoud OA, Woodward TH, Huestis MA (2017). Evaluation of divided attention psychophysical task performance and effects on pupil sizes following smoked, vaporized and oral cannabis administration. Journal of Applied Toxicology, 37(8): 922–932. https://doi.org/10.1002/jat.3440 Ortiz-Peregrina S, Ortiz C, Castro-Torres JJ, Jiménez JR, Anera RG (2020). Effects of smoking cannabis on visual function and driving performance. a driving-simulator based study. International Journal of Environmental Research and Public Health, 17(23): 9033. https://doi.org/10.3390/ijerph17239033 Otsu N (1979). A threshold selection method from gray-level histograms. IEEE Transactions on Systems, Man and Cybernetics, 9(1): 62–66. https://doi.org/10.1109/TSMC.1979.4310076 Papafotiou K, Carter JD, Stough C (2005). An evaluation of the sensitivity of the standardised field sobriety tests (sfsts) to detect impairment due to marijuana intoxication. Psychopharmacology, 180(1): 107–114. https://doi.org/10.1007/s00213-004-2119-9 Shahidi Zandi A, Comeau FJ, Mann RE, Di Ciano P, Arslan EP, Murphy T, et al. (2021). Preliminary eye-tracking data as a nonintrusive marker for blood δ-9-tetrahydrocannabinol concentration and drugged driving. Cannabis and Cannabinoid Research, 6(6): 537–547. https://doi.org/10.1089/can.2020.0141 Stark M, Englehart K, Sexton B, Tunbridge R, Jackson P (2003). Use of a pupillometer to assess change in pupillary size post-cannabis. Journal of Clinical Forensic Medicine, 10(1): 9–11. https://doi.org/10.1016/S1353-1131(02)00162-1 van der Walt S, Schönberger JL, Nunez-Iglesias J, Boulogne F, Warner JD, Yager N, et al. (2014). scikit-image: Image processing in Python. PeerJ, 2: e453. https://doi.org/10.7717/peerj.453