Researchfcn 101deeplabv3high frequency ultrasoundwhole slide images
Convolutional Network Identifies Cancerous Regions in Ultrasound
7.1
A single-center prospective study (July 2022–February 2023) at Sun Yat-sen University trained CNNs (FCN-101, DeepLabV3) on 386 registered high-frequency ultrasound images and corresponding biopsy whole-slide images from 105 patients to localize breast cancer regions. FCN-101 achieved higher pixel accuracy (86.91%) and Dice (77.47%) than DeepLabV3, demonstrating pixel-level concordance with histopathology and potential for biopsy guidance.
Key Points
- 1Demonstrates CNN models (FCN-101 and DeepLabV3) localize cancerous regions on registered HFUS images using biopsy WSIs
- 2Shows FCN-101 attains higher pixel accuracy (86.91%) and Dice coefficient (77.47%) versus DeepLabV3, indicating superior segmentation
- 3Enables clinicians and researchers to visualize pixel-level tumor distribution noninvasively, aiding biopsy guidance and treatment response assessment
Scoring Rationale
Moderate novelty with peer-reviewed validation and practical methods, limited by single-center data and modest recall metrics.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
