AI Paper Atlas

Research map / Detection and segmentation / CNN features

Multi-Stage Pathological Image Classification Using Semantic Segmentation

Shusuke Takahama, Yusuke Kurose, Yusuke Mukuta, Hiroyuki Abe, Masashi Fukayama, Akihiko Yoshizawa, Masanobu Kitagawa, Tatsuya Harada

ICCV 2019 · Detection and segmentation › CNN features · approach: Domain-specific segmentation and detection · 55 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add a segmentation model over patch features (Separate Learning)Classifier Only · Stomach biopsyaccuracy: 95.92 -> 98.53Improved
add a segmentation model over patch features (Separate Learning)Classifier Only · Stomach biopsyPR-AUC: 96.08 -> 99.30Improved
add end-to-end fine-tuning of the feature extractor with segmentation loss (ablation)Separate Learning · Stomach biopsyPR-AUC: 99.30 -> 99.34Improved
add end-to-end fine-tuning of the feature extractor with segmentation loss (ablation)Separate Learning · Stomach biopsyaccuracy: 98.53 -> 98.45No clear effect
train the feature extractor on patches instead of using ImageNet features (ablation)Segmentation on ImageNet features · Stomach biopsyaccuracy: 89.35 -> 98.53Improved
use full-resolution patch features instead of downsized 1/8 whole image (ablation)Segmentation Only · Stomach biopsyPR-AUC: 87.21 -> 99.30Improved
add the segmentation model (End-to-End)Classifier Only · Camelyon16accuracy: 96.23 -> 98.14Improved
add the segmentation model (End-to-End)Classifier Only · Camelyon16PR-AUC: 96.39 -> 99.24Improved
add the segmentation model (End-to-End)Classifier Only · Camelyon17kappa score: 71.9 -> 76.4Improved
use gradient retention and partial training instead of segmenting at 1/8 resolutionSegmentation Only · Stomach biopsyGPU memory: 7,000 MiB -> 10,000 MiB at full resolutionImproved
use End-to-End instead of Separate Learning (ablation)Separate Learning · Stomach biopsytraining time: 38 hours -> 73 hoursHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy