ACL-PCL
Description
The ACL-PCL Dataset is a curated collection of knee MRI scans annotated for Anterior Cruciate Ligament (ACL) and Posterior Cruciate Ligament (PCL) conditions. It includes preprocessed, noise-reduced, and augmented images across multiple MRI views, labeled into four distinct classes for ligament assessment. Key Features: Image Types & Views: Axial view Coronal view Sagittal T2-weighted view Sagittal Proton Density Fat-Suppressed (PD FS) view Classes (Annotations): Normal ACL Normal PCL Torn ACL Torn PCL Preprocessing Steps: Resized to uniform dimensions for consistency. Noise reduction applied to enhance image quality. Manual annotation by experts (radiologists/orthopedic specialists). Data Augmentation: Applied to increase diversity and robustness (e.g., rotations, flips, brightness adjustments). Purpose: Automated ligament tear detection. Multi-class classification of knee ligament health. Comparative analysis of MRI views for ACL/PCL diagnostics. Potential Applications: Medical imaging research. AI-assisted diagnosis tools. Educational datasets for radiology/orthopedics. Dataset Statistics (Example): Total images: [970] (post-augmentation). Resolution: 640x640. Ethical Considerations: Patient identifiers removed (de-identified).
Files
Institutions
- East West University