A Multi Center Breast FNAC Whole-Slide Cytology Dataset for AI-Assisted Patch-Wise Classification Using C1 to C5 Reporting Categories

2026-06-29

AI Summary

This study addresses the absence of large-scale, multicenter whole-slide image datasets for fine-needle aspiration cytology (FNAC) of the breast annotated according to the international C1–C5 diagnostic categories. We present a high-quality, publicly available dataset comprising 470 whole-slide images from 321 patients, featuring 7,398 expert-annotated image patches. Slides were stained with both Papanicolaou and May-Grünwald Giemsa methods, digitized at 40× magnification using the Hamamatsu NanoZoomer S360, and stored in NDPI and PNG formats (approximately 950 GB total). Annotations are provided in GeoJSON format. This is the first multicenter, multistain FNAC dataset aligned with C1–C5 standards at the patch level, accompanied by an open-source toolchain and openly accessible via Zenodo, offering a critical resource for AI-assisted diagnostic research.

Abstract

We present a multi center breast fine needle aspiration cytology (FNAC) dataset designed for patch wise classification using C1 to C5 reporting labels. The prospective dataset includes 321 patients and 470 whole-slide images (WSIs) collected from participating tertiary medical centers in India between May 2023 and March 2026. Slides were stained using Papanicolaou (190 WSIs) or MayGrunwald Giemsa (280 WSIs), scanned on a Hamamatsu NanoZoomer S360 at 40X magnification and 0.25 microns per pixel, and stored directly in NDPI format. Across the 470 WSIs, 446 WSIs contain annotated patch regions, yielding 7,398 PNG image patches with expert-verified C1 to C5 labels. The release includes NDPI WSIs, WSI-level GeoJSON annotation files, extracted patch images, deidentified metadata, a data dictionary, a validation summary, a manifest linking WSIs to Zenodo records, and code for dataset inspection and reuse. The complete dataset is approximately 950 GB and is available through Zenodo.

Problem

• breast FNAC
• whole-slide cytology
• patch-wise classification
• C1-C5 reporting
• multi-center dataset

Innovation

• whole-slide imaging
• patch-wise classification
• multi-center dataset
• C1-C5 cytology reporting
• AI-assisted diagnosis