Article

Predicting Generalization of AI Colonoscopy Models to Unseen Data

2024-08-07

Abstract excerpt

<title>Abstract</title> <p>Generalizability of AI colonoscopy algorithms is important for wider adoption in clinical practice. However, current techniques for evaluating performance on unseen data require expensive and time-intensive labels. We show that a "Masked Siamese Network" (MSN), trained to predict masked out regions of polyp images without labels, can predict the performance of Computer Aided Detection (...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 770342fe-8946-51b2-b295-ccf3cc4061e6
DOI: 10.21203/rs.3.rs-4307921/v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Predicting Generalization of AI Colonoscopy Models to Unseen DataDOI 10.21203/rs.3.rs-4307921/v1Read

Using Synthetic and Pseudosynthetic Data to Enhance Polyp Detection in Future AI-Assisted Endoscopy Frameworks. Is it the Right Time?Semantic proximity Read

Real-World Colonoscopy Video Integration to Improve Artificial Intelligence Polyp Detection Performance and Reduce Manual Annotation LaborSemantic proximity Read

AI based pre-screening of large bowel cancer via weakly supervised learning of colorectal biopsy histology imagesSemantic proximity Read

Biomimetic Transfer Learning-based Complex Gastrointestinal Polyp ClassificationSemantic proximity Read

Accurate recognition of colorectal cancer with semi-supervised deep learning on pathological imagesSemantic proximity Read

Performance of a deep learning system for automatic diagnosis of protruding lesions in colon capsule endoscopy: a multicentric studySemantic proximity Read

Text-Guided Synthesis in Medical Multimedia Retrieval: A Framework for Enhanced Colonoscopy Image Classification and SegmentationSemantic proximity Read

WITHDRAWN: Hybrid Attention U-Net++ with EfficientNet-B3 Encoder for Automatic Polyp Segmentation in Colonoscopy ImagesSemantic proximity Read

Computer-Aided Diagnosis of Serrated Colorectal Lesions Using Non-Magnified White Light Endoscopic ImagesSemantic proximity Read

Interpretable Self-Supervised Contrastive Learning for Colorectal Cancer Histopathology: GradCAM VisualizationSemantic proximity Read

Beyond Binary: Four-Class Risk Stratification from Gastrointestinal Endoscopy Using Asymmetric-Cost Lightweight CNN–Transformer LearningSemantic proximity Read

PolySAM-Lite: Parameter-efficient adaptation of the Segment Anything Model for colorectal polyp segmentationSemantic proximity Read