Get Started
Home
Topics
Search
Library
Research questionHow can lesion-focused image classifiers preserve sparse lesion signals while retaining useful global context?Lesion evidence may occupy small, dispersed regions, allowing irrelevant background to overwhelm it during whole-image processing. Discarding the surrounding image can also remove contextual information useful for classification.
AI
Computer Vision
Health
Image & Video Processing
Machine Learning
Research Paper
Latest papersRecent research connected to this question, newest first.An Attention-Guided Global and Local Fusion Framework for Lesion-Focused Image ClassificationThe study proposes a DenseNet-121 framework with global and local branches, using Grad-CAM-masked inputs, CBAM-enhanced local features, and instance-specific fusion weights. Evidence comes from a synthetic spot-pattern dataset and skin-lesion, guava-leaf, and grape-leaf image datasets; reported accuracies include 97.75% on the skin-lesion dataset and 99.64% on the guava-leaf dataset.research paper · Sep 4, 2026
Related questions
How can medical image models segment lesions despite complex backgrounds and varied morphologies?How can whole-slide image classifiers preserve irregular tissue regions while modeling relationships across spatially separated regions?How can computational pathology distinguish disease signal from acquisition shortcuts in reactive brain tissue?How should landmark-derived features preserve task-relevant geometry under nuisance variation in structured image classification?