Detection and classification of Plasmodium falciparum life stages from Giemsa-stained blood smear images using YOLO26m.
Detects three classes: Ring, Trophozoite, and Schizont.
| Version | mAP@0.5 | mAP@0.5:0.95 | Accuracy | Notes |
|---|---|---|---|---|
| V1 | 0.735 | 0.513 | 0.63 | Single dataset, 436 test instances |
| V2 | 0.830 | 0.638 | 0.73 | Merged dataset, 759 test instances |
The trained model, evaluation results, training curves, and full development story from V1 to V2 are available on Hugging Face:
huggingface.co/codeshujaaa/kenyanmalarai-detect
from huggingface_hub import hf_hub_download
from ultralytics import YOLO
# Load V2 model
weights_path = hf_hub_download(
repo_id = "codeshujaaa/kenyanmalarai-detect",
filename = "weights/best_v2.pt"
)
model = YOLO(weights_path)
results = model.predict("your_image.jpg", conf=0.15, imgsz=640)
results[0].show()Dataset by Roboflow community contributors.
Training on Kaggle Tesla P100 for V1, Tesla T4 x2 for V2.