AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Interactive and Explainable Region-Guided Radiology Report Generation

Tim Tanida, Philip Müller, Georgios Kaissis, Daniel Rueckert

CVPR 2023 · Vision-language and multimodal › Human and embodied modeling · approach: Neural image captioning · 201 citations (OpenAlex)

Read the paper

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
Use region-guided generation with anatomy detection, selection, and region-specific sentencesRGRG · MIMIC-CXRMETEOR: 0.158 -> 0.168Improved
Use region-guided generation with anatomy detection, selection, and region-specific sentencesRGRG · MIMIC-CXRROUGE-L: 0.285 -> 0.264Hurt
Use region-guided generation with anatomy detection, selection, and region-specific sentencesRGRG · MIMIC-CXRF1, ex-14: 0.384 -> 0.447Improved
Add region selection to the detector-plus-language-model base (ablation)RGRG · MIMIC-CXRBLEU-4: 0.104 -> 0.114Improved
Add region selection to the detector-plus-language-model base (ablation)RGRG · MIMIC-CXRF1, mic-5: 0.443 -> 0.523Improved
Add abnormality classification to the detector-plus-language-model base (ablation)RGRG · MIMIC-CXRF1, mic-5: 0.443 -> 0.501Improved
Add abnormality classification to the detector-plus-language-model base (ablation)RGRG · MIMIC-CXRBLEU-4: 0.104 -> 0.107No clear effect
Add abnormality classification to the detector-plus-language-model base (ablation)RGRG · MIMIC-CXRR, mic-5: 0.359 -> 0.461Improved
Add both region selection and abnormality classification (ablation)RGRG · MIMIC-CXRF1, mic-5: 0.443 -> 0.547Improved
Use RoI feature vectors instead of pooled RoI feature maps and a linear transformation (ablation)RGRG object detector · MIMIC-CXRobject detector performanceHurt
Vary the manually drawn box position away from the target region (ablation)RGRG selection-based sentence generation · Chest ImaGenomeMETEORHurt
Vary manually drawn box aspect ratio (ablation)RGRG selection-based sentence generation · Chest ImaGenomeMETEORNo clear effect

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