Yilin Liu, Ruining Deng, Juming Xiong, Regina N. Tyree, Hernan Correa, Girish Hiremath, Yaohong Wang, and Yuankai Huo. “Eosinophils Instance Object Segmentation on Whole Slide Imaging Using Multi-label Circle Representation.” Proceedings of SPIE Medical Imaging 2024: Digital and Computational Pathology, vol. 12933, 129330I, 2024.
Eosinophilic esophagitis (EoE) is a chronic condition causing inflammation in the esophagus, leading to difficulty swallowing, food getting stuck, and chest pain. These symptoms can severely affect a person’s quality of life, including their nutrition, social interactions, and mental health. Diagnosing EoE typically involves counting specific cells called eosinophils in tissue samples, a task that is labor-intensive for pathologists. To automate this process, researchers have developed an improved model called multi-label CircleSnake, which can identify and segment multiple types of cells. This new model is more precise and simpler compared to traditional methods. It outperforms existing models in accurately identifying eosinophils, making the diagnosis of EoE more efficient and accurate. The software for this model is available online for public use.
