Application of Convolutional Neural Networks to the Quantification of Tick Burdens on Cattle Using Infrared Thermographic Imaging

Abstract

Ticks and tick-borne diseases (TTBDs) are one of the biggest economic threats to livestock production systems in the world endangering approximately 80% of the global cattle population, especially in the sub- and tropical regions. It remains a challenge to effectively control ticks with acaricides due to the ability of ticks to develop resistance against acaricides. Algorithms for a cheap, rapid, and accurate method of quantifying tick burdens on cattle using infrared thermographic imaging technology could mitigate the danger of TTBDs in cattle. Tick counts were conducted once a month under natural challenge over a six-month period on 19 Bonsmara and 36 Nguni cattle located at ARC Roodeplaat and Loskop farms throughout both warmer climates and cooler climates. Thermographic images of both engorged & unfed females and males ticks were taken from cattle from February 2021 until July 2021. The deep learning models with architectures: “ConvNet” and “MobileNet” were trained on a dataset of 1124 “thermograms” to detect ticks on cattle. ConvNet model achieved a training and validation accuracy of $\sim 90$ and 60%, respectively. Whereas MobileNet scored a training and validation accuracy of $\sim 95$ and 75%, respectively. Finally, deep learning was successfully used to detect ticks on cattle using pretrained convolutional neural networks (CNNS).