05/12/2025
🔬 AI is transforming urinary particle analysis! A groundbreaking study leverages deep learning for multi-class urine cell detection-covering 15 crucial cellular components (erythrocytes, leukocytes, crystals, casts, fungi & more)!
🚀 Key wins:
82.86% mean average precision (mAP) on test sets
Blazing-fast 192ms per image processing
Outperforms Faster R-CNN & YOLO in accuracy/speed
💡 Built on RetinaNet (ResNet50 + FPN), the model tackles clinical pain points: manual microscopy’s inefficiency, limited cell type detection, and class imbalance.
Notably, this insightful argument is from our company’s founder—and it has been successfully applied in practice, powering our RT-U600 Urine Hybrid Analysis System!
✅RT-U600 automates urine formed elements analysis, delivering disease insights for kidney & urinary tract disorder screening. Based on AI deep learning, it achieves over 90% accuracy for formed elements recognition.
Reetoo, unlocks new potential for automated urine testing!
Read the article: https://doi.org/10.22541/au.157878399.96593318
Abstract Aim: Traditional artificial microscopic technologies cannot meet the current demands of automated urine detection. Furthermore, the number of cell types detected in previous studies was relatively limited; therefore, previous studies are cons