Accurate species identification of food-contaminating beetles with quality-improved elytral images and deep learning
Food samples are routinely screened for food-contaminating beetles (i.e., pantry beetles) due to their adverse impact on the economy, environment, public health and safety. If found, their remains are subsequently analyzed to identify the species responsible for the contamination; each species poses...
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-08-01
|
| Series: | Frontiers in Artificial Intelligence |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/frai.2022.952424/full |
| _version_ | 1818084844560711680 |
|---|---|
| author | Halil Bisgin Tanmay Bera Leihong Wu Hongjian Ding Neslihan Bisgin Zhichao Liu Monica Pava-Ripoll Amy Barnes James F. Campbell Himansi Vyas Cesare Furlanello Weida Tong Joshua Xu |
| author_facet | Halil Bisgin Tanmay Bera Leihong Wu Hongjian Ding Neslihan Bisgin Zhichao Liu Monica Pava-Ripoll Amy Barnes James F. Campbell Himansi Vyas Cesare Furlanello Weida Tong Joshua Xu |
| author_sort | Halil Bisgin |
| collection | DOAJ |
| description | Food samples are routinely screened for food-contaminating beetles (i.e., pantry beetles) due to their adverse impact on the economy, environment, public health and safety. If found, their remains are subsequently analyzed to identify the species responsible for the contamination; each species poses different levels of risk, requiring different regulatory and management steps. At present, this identification is done through manual microscopic examination since each species of beetle has a unique pattern on its elytra (hardened forewing). Our study sought to automate the pattern recognition process through machine learning. Such automation will enable more efficient identification of pantry beetle species and could potentially be scaled up and implemented across various analysis centers in a consistent manner. In our earlier studies, we demonstrated that automated species identification of pantry beetles is feasible through elytral pattern recognition. Due to poor image quality, however, we failed to achieve prediction accuracies of more than 80%. Subsequently, we modified the traditional imaging technique, allowing us to acquire high-quality elytral images. In this study, we explored whether high-quality elytral images can truly achieve near-perfect prediction accuracies for 27 different species of pantry beetles. To test this hypothesis, we developed a convolutional neural network (CNN) model and compared performance between two different image sets for various pantry beetles. Our study indicates improved image quality indeed leads to better prediction accuracy; however, it was not the only requirement for achieving good accuracy. Also required are many high-quality images, especially for species with a high number of variations in their elytral patterns. The current study provided a direction toward achieving our ultimate goal of automated species identification through elytral pattern recognition. |
| first_indexed | 2024-12-10T20:00:21Z |
| format | Article |
| id | doaj.art-b2f92c3b24c34128b48cba38daa0bc75 |
| institution | Directory Open Access Journal |
| issn | 2624-8212 |
| language | English |
| last_indexed | 2024-12-10T20:00:21Z |
| publishDate | 2022-08-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Artificial Intelligence |
| spelling | doaj.art-b2f92c3b24c34128b48cba38daa0bc752022-12-22T01:35:32ZengFrontiers Media S.A.Frontiers in Artificial Intelligence2624-82122022-08-01510.3389/frai.2022.952424952424Accurate species identification of food-contaminating beetles with quality-improved elytral images and deep learningHalil Bisgin0Tanmay Bera1Leihong Wu2Hongjian Ding3Neslihan Bisgin4Zhichao Liu5Monica Pava-Ripoll6Amy Barnes7James F. Campbell8Himansi Vyas9Cesare Furlanello10Weida Tong11Joshua Xu12Department of Mathematics and Applied Sciences, University of Michigan-Flint, Flint, MI, United StatesDivision of Bioinformatics and Biostatistics, National Center for Toxicological Research, US Food and Drug Administration, Jefferson, AR, United StatesDivision of Bioinformatics and Biostatistics, National Center for Toxicological Research, US Food and Drug Administration, Jefferson, AR, United StatesFood Chemistry Lab 1, Arkansas Regional Laboratory, Office of Regulatory Affairs, US Food and Drug Administration, Jefferson, AR, United StatesDepartment of Mathematics and Applied Sciences, University of Michigan-Flint, Flint, MI, United StatesDivision of Bioinformatics and Biostatistics, National Center for Toxicological Research, US Food and Drug Administration, Jefferson, AR, United StatesOffice for Food Safety, Center for Food Safety and Applied Nutrition, US Food and Drug Administration, College Park, MD, United StatesFood Chemistry Lab 1, Arkansas Regional Laboratory, Office of Regulatory Affairs, US Food and Drug Administration, Jefferson, AR, United StatesStored Product Insect and Engineering Research Unit, US Department of Agriculture, Manhattan, KS, United StatesFood Chemistry Lab 1, Arkansas Regional Laboratory, Office of Regulatory Affairs, US Food and Drug Administration, Jefferson, AR, United StatesHK3 Lab, Milan, ItalyDivision of Bioinformatics and Biostatistics, National Center for Toxicological Research, US Food and Drug Administration, Jefferson, AR, United StatesDivision of Bioinformatics and Biostatistics, National Center for Toxicological Research, US Food and Drug Administration, Jefferson, AR, United StatesFood samples are routinely screened for food-contaminating beetles (i.e., pantry beetles) due to their adverse impact on the economy, environment, public health and safety. If found, their remains are subsequently analyzed to identify the species responsible for the contamination; each species poses different levels of risk, requiring different regulatory and management steps. At present, this identification is done through manual microscopic examination since each species of beetle has a unique pattern on its elytra (hardened forewing). Our study sought to automate the pattern recognition process through machine learning. Such automation will enable more efficient identification of pantry beetle species and could potentially be scaled up and implemented across various analysis centers in a consistent manner. In our earlier studies, we demonstrated that automated species identification of pantry beetles is feasible through elytral pattern recognition. Due to poor image quality, however, we failed to achieve prediction accuracies of more than 80%. Subsequently, we modified the traditional imaging technique, allowing us to acquire high-quality elytral images. In this study, we explored whether high-quality elytral images can truly achieve near-perfect prediction accuracies for 27 different species of pantry beetles. To test this hypothesis, we developed a convolutional neural network (CNN) model and compared performance between two different image sets for various pantry beetles. Our study indicates improved image quality indeed leads to better prediction accuracy; however, it was not the only requirement for achieving good accuracy. Also required are many high-quality images, especially for species with a high number of variations in their elytral patterns. The current study provided a direction toward achieving our ultimate goal of automated species identification through elytral pattern recognition.https://www.frontiersin.org/articles/10.3389/frai.2022.952424/fullfood-contaminating beetlespecies identificationdeep learningconvolutional neural networksmachine learningfood safety |
| spellingShingle | Halil Bisgin Tanmay Bera Leihong Wu Hongjian Ding Neslihan Bisgin Zhichao Liu Monica Pava-Ripoll Amy Barnes James F. Campbell Himansi Vyas Cesare Furlanello Weida Tong Joshua Xu Accurate species identification of food-contaminating beetles with quality-improved elytral images and deep learning Frontiers in Artificial Intelligence food-contaminating beetle species identification deep learning convolutional neural networks machine learning food safety |
| title | Accurate species identification of food-contaminating beetles with quality-improved elytral images and deep learning |
| title_full | Accurate species identification of food-contaminating beetles with quality-improved elytral images and deep learning |
| title_fullStr | Accurate species identification of food-contaminating beetles with quality-improved elytral images and deep learning |
| title_full_unstemmed | Accurate species identification of food-contaminating beetles with quality-improved elytral images and deep learning |
| title_short | Accurate species identification of food-contaminating beetles with quality-improved elytral images and deep learning |
| title_sort | accurate species identification of food contaminating beetles with quality improved elytral images and deep learning |
| topic | food-contaminating beetle species identification deep learning convolutional neural networks machine learning food safety |
| url | https://www.frontiersin.org/articles/10.3389/frai.2022.952424/full |
| work_keys_str_mv | AT halilbisgin accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning AT tanmaybera accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning AT leihongwu accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning AT hongjianding accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning AT neslihanbisgin accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning AT zhichaoliu accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning AT monicapavaripoll accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning AT amybarnes accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning AT jamesfcampbell accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning AT himansivyas accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning AT cesarefurlanello accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning AT weidatong accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning AT joshuaxu accuratespeciesidentificationoffoodcontaminatingbeetleswithqualityimprovedelytralimagesanddeeplearning |