MurineCyto-Det: A High-Resolution Murine BALF Cytology Dataset for Leukocyte Segmentation and Detection

Area of Science:

• Veterinary Medicine
• Computational Biology
• Biomedical Imaging

Background:

• Automated analysis of murine bronchoalveolar lavage fluid (BALF) cytology is crucial for preclinical respiratory research.
• Progress in this area is hindered by the scarcity of publicly available, well-annotated mouse BALF image datasets.

Purpose of the Study:

• To introduce MurineCyto-Det, a high-resolution dataset for murine BALF cytology.
• To provide a standardized resource for developing and evaluating automated analysis methods.

Main Methods:

• The dataset comprises 333 high-resolution image tiles (1024x1024 pixels).
• Annotations include pixel-level segmentation masks and bounding boxes for 14,551 cell instances across five categories.
• The dataset supports cell segmentation and cell detection tasks.

Main Results:

• Evaluated representative segmentation and detection models to establish benchmark baselines.
• Demonstrated the dataset's utility while identifying challenges like class imbalance and small object sizes.
• Highlighted issues with irregular cell morphology and ambiguous debris-like structures.

Conclusions:

• MurineCyto-Det offers a valuable, publicly available resource for advancing automated murine BALF cytology analysis.
• The dataset facilitates the development, evaluation, and comparison of computational methods in respiratory research.
• Addressing identified challenges is key for improving automated cytology analysis accuracy.