We collected CT scans of 255 pancreatitis patients with IRB approval from Istanbul University Capa School of Medicine. There were two classes in the dataset: \textbf{179} pancreatitis patients \textbf{with peri-pancreatic edema} and \textbf{76} pancreatitis patients \textbf{without peri-pancreatic edema.} Each of the images had a corresponding standard pancreas mask and a label for disease class. Images with peri-pancreatic edema tissue were categorized as positive samples, denoted by label 1, other images are denoted with label 0. All standard pancreas segmentation masks were annotated by a team of expert radiologists (N=2) on consensus. After annotation, a careful evaluation was made to ensure the accuracy and reliability of the annotations. The standard pancreas segmentation masks were then used to evaluate the pancreas mask-generating model. We provide a unique dataset for the public, comprising CT imaging data from 255 patients with detailed pancreas segmentation masks. This comprehensive dataset will serves as a valuable resource for the exploration and advancement of deep learning techniques in pancreatic research in general, offering a rich source of information for future research and clinical applications. The public link is below: https://arxiv.org/abs/2404.17064 full citation for the paper will be updated soon (IEEE EMBC 2024). @misc{hong2024detection, title={Detection of Peri-Pancreatic Edema using Deep Learning and Radiomics Techniques}, author={Ziliang Hong and Debesh Jha and Koushik Biswas and Zheyuan Zhang and Yury Velichko and Cemal Yazici and Temel Tirkes and Amir Borhani and Baris Turkbey and Alpay Medetalibeyoglu and Gorkem Durak and Ulas Bagci}, year={2024}, eprint={2404.17064}, archivePrefix={arXiv}, primaryClass={eess.IV} }