ghostsas001/bone-marrow-cells-yolo

YOLO — bone marrow cell classification (20 cell types)

Twenty-class classifier for single-cell images from bone marrow smears, built with Ultralytics YOLO (classification).

79.48 % balanced accuracy over 20 cell types — 8 points above the dataset paper's ResNeXt-50 (71.4 %), trained on 30 % of the images with one fixed recipe, without dataset-specific tuning.

Metric on the held-out test set (17,137 cells)	Value
Balanced accuracy (mean recall, 20 classes)	79.48 %
Accuracy	85.30 %
Macro F1	76.65 %

The companion model (SigLIP) and the full write-up: GitHub.

---

The dataset

Bone-Marrow-Cytomorphology_MLL_Helmholtz_Fraunhofer (Matek et al. 2021): 250 × 250 px single-cell images from bone marrow smears of 945 patients, May-Grünwald-Giemsa stained, expert-labelled into 21 classes. Abnormal eosinophils (8 images) were dropped, leaving 20 classes.

Split	Cells
Train (max 4,000 per class)	51,737
Validation	17,137
Test	17,137

---

The problems with this dataset

1. Extreme imbalance — from 29,424 images (segmented neutrophils) down to 8; seven classes have fewer than 500 images, and three have only 4 to 6 cells in the test split.
2. No patient identifiers — every split, ours and the paper's, is per image, so the same patient's cells can appear in training and test.
3. Neighbouring maturation stages — myelocyte → metamyelocyte → band → segmented neutrophil are judgement calls at the boundaries, and that is where the errors concentrate.

---

Compared with the dataset paper

Matek et al. (Blood 2021) report class-wise strict recall for a ResNeXt-50 trained with five-fold cross-validation on the whole dataset (classes upsampled to about 25,000 images). The mean over the same 20 classes is 71.4 %. This model is above the paper on 17 of the 20 classes; the three below it are small in the test set — basophils (44 cells), other cells (30) and immature lymphocytes (6).

Not the same protocol: the paper averages five folds; this model is tested once on a 10 % hold-out. Both are per-image random splits over the same 20 classes.

---

Results in detail

(Matrix shown for SigLIP; the YOLO errors follow the same maturation-chain pattern.)

---

Training and evaluation

Training data	at most 4,000 cells per class (51,737 of 171,366)
Training	50 epochs, 224 px, batch 64, Ultralytics default augmentation
Split	stratified 80 / 10 / 10 per image, seed 42
Model selection	best epoch on the validation split
Test	held-out 10 %, evaluated once, no tuning on the test set
Hardware	one NVIDIA T4

Usage

from huggingface_hub import hf_hub_download
from ultralytics import YOLO

model = YOLO(hf_hub_download("ghostsas001/bone-marrow-cells-yolo", "yolo.pt"))
result = model.predict("cell.jpg", imgsz=224)[0]
print(model.names[int(result.probs.top1)])

Labels: ART, BAS, BLA, EBO, EOS, FGC, HAC, KSC, LYI, LYT, MMZ, MON, MYB, NGB, NGS, NIF, OTH, PEB, PLM, PMO — abbreviations as in the dataset (for example BLA = blast, MYB = myelocyte, NGS = segmented neutrophil).

Limitations

Research model, not a medical device. Trained on one laboratory's smears and scanner; other stains, microscopes or cohorts may reduce accuracy. One training run, no confidence intervals. Classes with fewer than 50 test cells have noisy recall.

References

• Matek C., Krappe S., Münzenmayer C., Haferlach T., Marr C. Highly accurate differentiation of bone marrow cell morphologies using deep neural networks on a large image data set. Blood 138(20), 2021. doi:10.1182/blood.2020010568
• Dataset: Bone-Marrow-Cytomorphology_MLL_Helmholtz_Fraunhofer, The Cancer Imaging Archive, doi:10.7937/TCIA.AXH3-T579, CC BY 4.0
• Ultralytics YOLO — docs.ultralytics.com

Licence

Built with Ultralytics YOLO, which is licensed under AGPL-3.0; these weights are distributed under the same licence. Training images: CC BY 4.0 (Matek et al., TCIA).