BiCervi: Cervical Cytology Image Dataset for Pap Smear Classification

This dataset provides curated cervical cytology (Pap-smear) microscopy images intended to support research and development of machine learning models for cervical abnormality detection and related computational pathology tasks. The release includes microscopy images labeled in a diagnostic category defined using The Bethesda System (BTS) for reporting cervical cytology, covering eight classes: Atypical Squamous Cells of Undetermined Significance (ASC-US), Atypical Squamous Cells Cannot Exclude High Grade Lesions (ASC-H), Atypical Glandular Cells (AGC), High-grade Squamous Intraepithelial Lesion (HSIL), Low-grade Squamous Intraepithelial Lesion (LSIL), Negative for Intraepithelial Lesion or Malignancy (NILM), Squamous Cell Carcinoma, and Adenocarcinoma. The dataset comprises a total of 3,000 images. All images are provided in ZIP files: ‘ASC-US.zip’ contains 518 images, ‘ASC-H.zip’ contains 183 images, ‘AGC.zip’ contains 116 images, ‘HSIL.zip’ contains 397 images, ‘LSIL.zip’ contains 319 images, ‘NILM.zip’ contains 1,447 images, ‘SC.zip’ contains 11 images, and ‘A.zip’ contains 9 images.