LungVision-16: Deep Learning-Powered Lung Cancer Detection in Animals via CT Imaging

Lung cancer, often characterized by the abnormal and uncontrolled growth of cells in lung tissues, poses a significant health threat to animals, leading to respiratory distress and systemic complications if left undiagnosed. As a leading cause of mortality in animals, early and accurate diagnosis is critical to improving treatment outcomes. This research focuses on developing a deep learning model to detect lung cancer in animals, specifically targeting cancer subtypes such as adenocarcinoma, cancer, nodule, and nodule multiple. Three convolutional neural networks were evaluated, such as VGG16, EfficientNetB0, and DenseNet201. The base model yielded results that were mediocre. Among them the LungVision-16 demonstrated the highest accuracy of 97%, marginally outperforming the other employed models. The LungVision-16 also showed exceptional performance in terms of precision and recall across different cancer classes. As a result of this study, veterinary oncologists now have a robust tool for early diagnosis and effective treatment of lung cancer in animals.