{"success":true,"data":{"author":"Andrea Acevedo, Anna Merino et al. (Hospital Clinic of Barcelona); repackaged by Chehab Lab","bucketAccessType":"public","bucketName":"chehablab/PBC","bucketPrefix":"","category":"Health & Biomedical","cloudProvider":"huggingface","createdAt":"2026-11-21T22:16:41.564Z","description":"The PBC dataset holds 17,092 images of individual normal peripheral blood cells acquired with a CellaVision DM96 analyzer in the core laboratory of the Hospital Clinic of Barcelona and labeled by expert clinical pathologists. Cells fall into eight groups: neutrophils (3,329), eosinophils (3,117), basophils (1,218), lymphocytes (1,214), monocytes (1,420), immature granulocytes (2,895), erythroblasts (1,551) and platelets (2,348). Images are 360x363 pixel JPEGs, kept unchanged; columns add the cell type, a finer subtype code from the filename prefix (such as SNE, BNE or MY) and an integer label. One Parquet file under data/, about 280 MB (measured 2026-11-21); read with Hugging Face datasets. Use for automated white blood cell differential counting and hematology image classification. Original data on Mendeley Data (DOI 10.17632/snkd93bnjr.1), released under CC BY 4.0. Cite Acevedo et al., Data in Brief 30:105474 (2020).","displayTitle":"PBC: 17,092 microscopy images of normal peripheral blood cells in eight classes (CellaVision, Hospital Clinic of Barcelona)","doi":"10.17632/snkd93bnjr.1","huggingFaceUrl":"https://huggingface.co/datasets/chehablab/PBC","id":"1b3234fb-11aa-5de1-b728-69efe22c2f46","license":"CC-BY-4.0","licenseEvidenceUrl":"https://data.mendeley.com/public-api/datasets/snkd93bnjr?version=1","objectCountEstimate":3,"probeStatus":"ok","referenceUrl":"https://doi.org/10.1016/j.dib.2020.105474","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"chehablab/PBC\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"chehablab/PBC\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=chehablab, license:cc-by-4.0, size_categories:10K<n<100K, format:parquet, downloads=136"],"format":"mixed","generatedAt":"2026-11-21T22:32:06.335Z","layout":{"example":"data/train-00000-of-00001.parquet","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"data/train-00000-of-00001.parquet","truncatedAt":512}],"sampleObjects":["data/train-00000-of-00001.parquet"],"sidecars":[{"bytes":"---\ndataset_info:\n  features:\n  - name: image_id\n    dtype: string\n  - name: image\n    dtype: image\n  - name: cell_type\n    dtype: string\n  - name: cell_subtype\n    dtype: string\n  - name: label\n    dtype: int32\n  splits:\n  - name: train\n    num_examples: 17092\nconfigs:\n- config_name: default\n  data_files:\n  - split: train\n    path: data/train-*\nlicense: cc-by-4.0\ntask_categories:\n- image-classification\nlanguage:\n- en\ntags:\n- microscopy\n- blood\n- hematology\n- blood-cells\n- white-blood-cells\n- medical\n- classification\npretty_name: PBC (Peripheral Blood Cells)\nsize_categories:\n- 10K<n<100K\n---\n# PBC (Peripheral Blood Cells) — Microscopic Images of Normal Peripheral Blood Cells\n\nThis dataset is derived from the **PBC dataset** (Acevedo et al., 2020), a collection of microscopic images of individual normal peripheral blood cells, originally hosted on **Mendeley Data**. The original images were acquired with the CellaVision DM96 analyzer in the core laboratory of the Hospital Clinic of Barcelona and labelled by expert clinical pathologists into eight cell groups.\nHere, we kept the original JPG images unchanged and added the cell type, the cell subtype from the image filename, and an integer label.\nThe dataset is designed for blood-cell classification.\n\n\n## 📦 Dataset Structure\n\nThe dataset has a single `train` split, with each row corresponding to a single blood-cell image:\n\n| Field | Description |\n|---|---|\n| `image_id` | Original image filename without extension (e.g., `BA_100102`) |\n| `image` | RGB microscopy image of a single cell (360 × 363 pixels, JPG) |\n| `cell_type` | Cell group: `basophil`, `eosinophil`, `erythroblast`, `ig`, `lymphocyte`, `monocyte`, `neutrophil`, `platelet` |\n| `cell_subtype` | Filename prefix of the original image (e.g., `SNE`, `BNE`, `MY`), which encodes finer subtypes for some groups |\n| `label` | Integer cell group: `0`–`7` (see Feature Descriptions) |\n\n**Class Distribution:**\n- Basophil: 1,218 images\n- Eosinophil: 3,117 images\n- Erythroblast: 1,551 images\n- Immature granulocyte (IG): 2,895 images\n- Lymphocyte: 1,214 images\n- Monocyte: 1,420 images\n- Neutrophil: 3,329 images\n- Platelet: 2,348 images\n- Total: 17,092 images\n\n\n## ⚙️ Preprocessing\n\n- `PBC_dataset_normal_DIB.zip` was downloaded from Mendeley Data (`snkd93bnjr`, version 1)\n- Images were kept as the original JPG files (no re-encoding, resizing or color normalization)\n- Cell type was taken from each image's parent folder name\n- Cell subtype was taken from the image filename prefix (the part before the last `_`)\n- All images were preserved\n\n\n## 🚀 Usage\n\n```python\nfrom datasets import load_dataset\nimport matplotlib.pyplot as plt\n\nds = load_dataset(\"chehablab/PBC\", split=\"train\")\n\n# One example image per cell group\nclasses = [\"basophil\", \"eosinophil\", \"erythroblast\", \"ig\",\n           \"lymphocyte\", \"monocyte\", \"neutrophil\", \"platelet\"]\nlabels = ds[\"label\"]\nfig, axes = plt.subplots(1, len(classes), figsize=(16, 2.6))\nfor ax, label in zip(axes, range(len(classes))):\n    sample = ds[labels.index(label)]\n    ax.imshow(sample[\"image\"])\n    ax.set_title(f\"{classes[label]}\\n{sample['cell_subtype']}\")\n    ax.axis(\"off\")\nplt.tight_layout()\nplt.show()\n\n# Filter by cell group\nneutrophils = ds.filter(lambda x: x[\"cell_type\"] == \"neutrophil\")\nprint(f\"Neutrophils: {len(neutrophils)}\")\n\n# No official split: make a stratified train/test split\nsplit = ds.class_encode_column(\"label\").train_test_split(test_size=0.2, stratify_by_column=\"label\", seed=0)\n```\n\n---\n\n## 🏷️ Feature Descriptions\n\n### Cell Groups\nEight groups of normal peripheral blood cells:\n\n- **Basophil** (`label=0`): 1,218 images — Granulocyte with large dark-purple granules\n- **Eosinophil** (`label=1`): 3,117 images — Granulocyte with orange-red granules\n- **Erythroblast** (`label=2`): 1,551 images — Nucleated red blood cell precursor\n- **Immature granulocyte, IG** (`label=3`): 2,895 images — Promyelocytes, myelocytes and metamyelocytes\n- **Lymphocyte** (`label=4`): 1,214 images — Mononuclear cell with a round nucleus and scant cytoplasm\n- **Monocyte** (`label=5`): 1","encoding":"utf8","key":"README.md","name":"README.md","size":8256,"truncated":true}],"sourceObjectKey":"data/train-00000-of-00001.parquet","starterCell":"from datasets import load_dataset\nds = load_dataset(\"chehablab/PBC\")\nprint(ds)","starterSamplePath":"data/train-00000-of-00001.parquet","totalObjects":3},"summary":"The PBC dataset holds 17,092 images of individual normal peripheral blood cells acquired with a CellaVision DM96 analyzer in the core laboratory of the Hospital Clinic of…","title":"PBC: 17,092 microscopy images of normal peripheral blood cells in eight classes (CellaVision, Hospital Clinic of Barcelona)","updatedAt":"2026-11-22T01:09:40.429Z","visibility":"public","websiteUrl":"https://data.mendeley.com/datasets/snkd93bnjr/1"}}