UCLA

With the expansive amount of data, modern technologies including machine learning and deep learning have become extremely complex and accurate. Deep learning models started to gaining popularity in the early 2010. AlexNet is a type of Convolutional Neural Network (CNN) was first introduced and won the ImageNet Large Scale Visual Recognition Challenge (ILSVRC). This model dramatically increased accuracy compared to the previous year by nearly ten percent, resulting in many variations of CNN architectures being introduced based upon this model. In the year of 2015, deep learning reached a new milestone because of a new model architecture called Residual Network (ResNet), which is another variant of CNN architecture. ResNet was trained on ImageNet, a database that contains a wide range of various images, and the ability to accurately predict those images started to outperform human’s abilities (Figure 1). This significant improvement in Computer Vision models allows many researchers to help implement modern technologies such as self-driving cars and face recognition. After gaining its credibility and accuracy from recent state of the art models, various domains, including health care, have begun to adapt the methodology to better predict one’s medical conditions such as the criticalness of tumor and detect if the malignant or benign from medical images including X rays and CT scans. Throughout this paper, we will discuss the application of modern Computer Vision techniques in the Veterinary field with the hope of reducing the number of canine/feline patients who have serious illness and increasing their quality of lives by detecting possible illness in their early stages.