Large Annotated Ultrasound Dataset of Non-Alcoholic Fatty Liver from Saudi Hospitals for Analysis and Applications

Large Annotated Ultrasound Dataset for Non-Alcoholic Fatty Liver Disease (NAFLD)

This dataset is a valuable resource for researchers and developers working on AI-assisted diagnosis and staging of Non-Alcoholic Fatty Liver Disease (NAFLD). It comprises a substantial collection of 10,352 high-resolution ultrasound images obtained from 384 patients at two major Saudi hospitals.

Key features of the dataset:

Size and diversity: A large dataset with images from a diverse patient population, capturing a wide spectrum of NAFLD severity.

Detailed annotations: Each image is meticulously annotated with NAFLD Activity Score (NAS) fibrosis staging and steatosis grading, based on corresponding liver biopsy results.

Standardization: Images undergo rigorous pre-processing, including size standardization and quality control, ensuring consistency and reliability.

Technical validation: Inter-rater reliability and image quality preservation are rigorously assessed.

Potential applications:

Development and evaluation of AI algorithms for accurate NAFLD diagnosis and staging.

Research on the correlation between ultrasound image features and NAFLD progression.

Support for the development of computer-aided diagnostic tools for clinicians.

By providing a comprehensive and well-annotated dataset, this resource aims to accelerate research and improve patient outcomes in NAFLD management.