Abstract

Forensic palynology applies pollen grains and spores as trace evidence in scenes of crime. Identification by hand is labor intensive and subject to observer bias and error. Major objective is to highlight a conceptual model for automating the analysis of pollen grains with imaging and artificial intelligence (AI). A lab-based simulation study created from online available datasets and microscopic images of pollen grains. This study presents a conceptual model for automating pollen analysis through imaging and artificial intelligence (AI). A lab-based simulation utilized publicly available microscopy datasets of 25 pollen species to develop a theoretical framework. The morphology of pollen grains and spores were assessed. The workflow that was proposed consisted of: data acquisition, image pre-processing, image analysis, and dataset generation; and classified pollen grains and spores using artificial intelligence models (Artificial Neural Networks, Convolutional Neural Networks (CNN) and support vector machine (SVM). AI and automation present a significant number of possibilities to enhance the effectiveness and speed of forensic palynology. There is a requirement for a properly curated dataset in the development for effective automated tools.

• 1.   Crispino F., Ribaux O., Houck M., Margot P. Forensic science–A true science?. Australian Journal of Forensic Sciences. 2011 Jun 1; 43(2-3): 157-76.
• 2.   Tilstone W.J. Forensic science: An encyclopedia of history, methods, and techniques. Bloomsbury Publishing USA; 2006 Mar 24.
• 3.   Thornton J.I. Uses and abuses of forensic science. In Science and Law 2019 Jun 21 (pp. 79-90). Routledge.
• 4.   Houck M.M., Siegel J.A. Fundamentals of forensic science. Academic Press; 2009 Nov 30.
• 5.   Mildenhall D.C., Wiltshire P.E., Bryant V.M. Forensic palynology: why do it and how it works. Forensic science international. 2006 Nov 22; 163(3): 163-72.
• 6.   Kumari M., Sankhla M.S., Nandan M., Sharma K., Kumar R. Role of forensic palynology in crime investigation. IJournals: International Journal of Social Relevance & Concern. 2017; 5(3): 1-3.
• 7.   Rakshanda A., Reddy J. Pollen in Forensic Palynology: An Exploration into a Crime Solving Tool. Int J Environ Agric Biotechnol. 2022; 7: 048-53.
• 8.   Walsh K.A., Horrocks M. Palynology: its position in the field of forensic science. Journal of forensic sciences. 2008 Sep; 53(5): 1053-60.
• 9.   Bryant V.M., Jones G.D. Forensic palynology: Current status of a rarely used technique in the United States of America. Forensic Science International. 2006 Nov 22; 163(3): 183-97.
• 10.   Alotaibi S.S., Sayed S.M., Alosaimi M., Alharthi R., Banjar A., Abdulqader N., Alhamed R. Pollen molecular biology: Applications in the forensic palynology and future prospects: A review. Saudi journal of biological sciences. 2020 May 1; 27(5): 1185-90.
• 11.   Mahmood T., Wahid A., Hong J.S., Kim S.G., Park K.R. A novel convolution transformerbased network for histopathology-image classification using adaptive convolution and dynamic attention. Engineering Applications of Artificial Intelligence. 2024 Sep 1; 135: 108824.
• 12.   Chiou E.K., Lee J.D. Trusting automation: Designing for responsivity and resilience.Human factors. 2023 Feb; 65(1): 137-65.
• 13.   Gallardo-Caballero R., García-Orellana C.J., García-Manso A., González-Velasco HM, Tormo-Molina R., Macías-Macías M. Precise pollen grain detection in bright field microscopy using deep learning techniques. Sensors. 2019 Aug 17; 19(16): 3583.
• 14.   Sevillano V., Aznarte J.L. Improving classification of pollen grain images of the POLEN23E dataset through three different applications of deep learning convolutional neural networks. PloS one. 2018 Sep 14; 13(9): e0201807.
• 15.   Sobol M.K., Finkelstein S.A. Predictive pollenbased biome modeling using machine learning. PloS one. 2018 Aug 23; 13(8): e0202214.16. Viertel P., König M. Pattern recognition methodologies for pollen grain image classification: a survey. Machine Vision and Applications. 2022 Jan; 33(1): 18.
• 17.   Bourel B., Marchant R., de Garidel-Thoron T., Tetard M., Barboni D., Gally Y., Beaufort L. Automated recognition by multiple convolutional neural networks of modern, fossil, intact and damaged pollen grains. Computers & Geosciences. 2020 Jul 1; 140: 104498.
• 18.   Maione C., Barbosa Jr F., Barbosa R.M. Predicting the botanical and geographical origin of honey with multivariate data analysis and machine learning techniques: A review. Computers and Electronics in Agriculture. 2019 Feb 1; 157: 436-46.
• 19.   Rodriguez-Damian M., Cernadas E., Formella A., Fernandez-Delgado M., De Sa-Otero P. Automatic detection and classification of grains of pollen based on shape and texture. IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews). 2006 Jun 19; 36(4): 531-42.
• 20.   Walker J.W. Evolution of exine structure in the pollen of primitive angiosperms. American Journal of Botany. 1974 Sep; 61(8): 891-902.
• 21.   Dunker S., Motivans E., Rakosy D., Boho D., Mader P., Hornick T., Knight T.M. Pollenanalysis using multispectral imaging flow cytometry and deep learning. New Phytologist. 2021 Jan; 229(1): 593-606.
• 22.   Rokach L. Pattern classification using ensemble methods. World Scientific; 2010.
• 23.   h t t p s : / / w w w . r e s e a r c h g a t e . n e t / publication/365145644/figure/fig1/AS:1143 1281388514671@1745175031098/Pipeline-foralgorithm-development-1-The-pollen-wasplaced-in-TTC-solution-for-1-h_Q320.jpg

Data Sharing Statement

Funding