An intelligent diagnostic method for porcine gastrointestinal infectious diseases based on multimodal AI and large language model

The swine farming industry, a key pillar of Chinese animal husbandry, faces significant challenges due to frequent outbreaks of porcine gastrointestinal infectious diseases (PGID). Traditional diagnostic methods reliant on human expertise suffer from low efficiency, high subjectivity, and poor accuracy. To address these issues, this paper proposes a multimodal diagnostic method based on artificial intelligence (AI) and large language model (LLM) for six common types of PGID. In this method, ChatGPT and image augmentation techniques were first used to expand the dataset. Next, the Multi-scale TextCNN (MS-TextCNN) model was employed to capture multi-granularity semantic features from text. Subsequently, an improved Mask R-CNN model was applied to segment small intestine lesion regions, after which seven convolutional neural network (CNN) models were used to classify the segmented images. Finally, five machine learning models were utilized for multimodal classification diagnosis. Experimental results demonstrate that the multimodal diagnostic model can accurately identify six common types of PGID. This study provides an efficient and accurate intelligent solution for diagnosing PGID and demonstrates superior performance compared with single-modality methods.