Browse all
531 papers from 2021
in the searchable list.
Semantic Map
531 papers in 20 research clusters
specter2 embeddings → UMAP. Click a point to open the paper.
By cluster
Topics & Authors
Top Subject Areas
Top 20 by paper count
Top 20 Authors
By paper count. Authors are matched by name.
Review Analysis
Review Score Distribution
Raw 1-9 scale (2021 scoring)
Reviewer Disagreement
Max − min score per paper
Code & Research Trends
Code Availability by Subject Area
Areas with ≥ 8 papers
51.0% overall
Research Buzzword Frequency
Matched against title + abstract (may overlap)
Co-authorship Network
Authors with ≥ 4 papers at MICCAI 2021
Node size = paper count. Edge = shared authorship. Color = detected collaboration community; disconnected groups are laid out separately so they don't overlap the main cluster. Drag a node to reposition it, drag the background to pan, scroll to zoom. Authors are matched by name.
Paper Title Patterns
Title length and most common first words
Articles and prepositions excluded from first-word count
Top Papers
Top 20 Papers (Overall)
Ranked by mean reviewer score (1-9 scale)
| # |
Title |
Subject Area |
Mean Reviewer Score (Scores) |
Presentation |
Early Accept? |
| 1 |
Detecting when pre-trained nnU-Net models fail silently for Covid-19 lung lesion segmentation |
Lung |
8.67
(8, 9, 9)
|
Oral |
✓ |
| 2 |
Do we need complex image features to personalize treatment of patients with locally advanced rectal cancer? |
Oncology |
8.67
(8, 9, 9)
|
Poster |
✓ |
| 3 |
Hierarchical graph pathomic network for progression free survival prediction |
Computational (Integrative) Pathology |
8.67
(8, 9, 9)
|
Poster |
✓ |
| 4 |
3D-UCaps: 3D Capsules Unet for Volumetric Image Segmentation |
Image Segmentation |
8.50
(8, 9)
|
Oral |
✓ |
| 5 |
A Hierarchical Feature Constraint to CamouflageMedical Adversarial Attacks |
other |
8.33
(8, 9, 8)
|
Poster |
✓ |
| 6 |
Comprehensive Importance-based Selective Regularization for Continual Segmentation Across Multiple Sites |
Abdomen |
8.33
(8, 9, 8)
|
Poster |
✓ |
| 7 |
Continuous-Time Deep Glioma Growth Models |
Oncology |
8.33
(7, 9, 9)
|
Poster |
✓ |
| 8 |
Cross-modal Attention for MRI and Ultrasound Volume Registration |
Image Registration |
8.33
(7, 9, 9)
|
Oral |
✓ |
| 9 |
Developmental Stage Classification of Embryos Using Two-Stream Neural Network with Linear-Chain Conditional Random Field |
Microscopy |
8.33
(9, 7, 9)
|
Poster |
✓ |
| 10 |
Learned super resolution ultrasound for improved breast lesion characterization |
Breast |
8.33
(8, 8, 9)
|
Oral |
✗ |
| 11 |
Personalized Retrogress-Resilient Framework for Real-World Medical Federated Learning |
Dermatology |
8.33
(8, 7, 10)
|
Oral |
✓ |
| 12 |
Quantitative Assessments for Ultrasound Probe Calibration |
Image-Guided Interventions and Surgery |
8.33
(9, 9, 7)
|
Oral |
✗ |
| 13 |
Selective Learning from External Data for CT Image Segmentation |
Abdomen |
8.33
(8, 9, 8)
|
Poster |
✓ |
| 14 |
Variational Topic Inference for Chest X-Ray Report Generation |
Lung |
8.33
(9, 9, 7)
|
Oral |
✓ |
| 15 |
Conditional Deformable Image Registration with Convolutional Neural Network |
Image Registration |
8.00
(9, 7, 8)
|
Oral |
✓ |
| 16 |
Contrastive Learning of Relative Position Regression for One-Shot Object Localization in 3D Medical Images |
Self-supervised learning |
8.00
(8, 8, 8)
|
Poster |
✓ |
| 17 |
Controllable cardiac synthesis via disentangled anatomy arithmetic |
Cardiac |
8.00
(8, 8, 8)
|
Oral |
✓ |
| 18 |
Demystifying T1-MRI to FDG18-PET Image Translation via Representational Similarity |
Neuroimaging - Others |
8.00
(6, 10, 8)
|
Oral |
✓ |
| 19 |
Efficient and Generic Interactive Segmentation Framework to Correct Mispredictions during Clinical Evaluation of Medical Images |
Computer Aided Diagnosis |
8.00
(6, 9, 9)
|
Poster |
✓ |
| 20 |
Federated Contrastive Learning for Volumetric Medical Image Segmentation |
Image Segmentation |
8.00
(8, 8, 8)
|
Oral |
✓ |
Top 20 Papers (Rebuttal)
Not available for MICCAI 2021, because the open-access review data for this year
contains no post-rebuttal reviewer scores or verdicts, so a rebuttal ranking cannot be built.