The Journal of Machine Learning for Biomedical Imaging 2022

17 papers
The 2022 edition of The Journal of Machine Learning for Biomedical Imaging published 16 papers centered on the development of decentralized and reliable artificial intelligence. A primary research direction involved federated learning, with studies addressing data heterogeneity, non-IID distributions, and privacy-preserving frameworks for multi-view observations. Uncertainty quantification was also a central theme, particularly regarding its application in liver lesion detection and the benchmarking of ranking scores for brain tumor segmentation through the QU-BraTS challenge. Technical advancements featured the use of hypernetworks for image reconstruction and the HyperMorph framework for learning registration hyperparameters. Other research explored spatially sparse target segmentation, the impact of annotation styles on generalization, and learning medical representations from label-scarce data. Specific applications included skin lesion classification via semi-supervised peer learning and the topological quantification of pancreatic tubular networks in 3D microscopy. Prolific contributors to this volume included Raghav Mehta, Tal Arbel, Adrian V. Dalca, and Ben Glocker.

Top Research Directions

The 57 research directions with The Journal of Machine Learning for Biomedical Imaging 2022 papers, ranked by paper count, next to 2021. Share is the fraction of the edition’s papers in the direction.

#Direction20222021Share 2022Share 20212021 rank
1Parameter Synthesis for Dynamic Model Adaptation2011.8%0.0%—
2Quantifying Spatial Uncertainty via Generative Sampling2011.8%0.0%—
3Addressing Extreme Spatial Class Imbalance in Medical Segmentation105.9%0.0%—
4Aligning Image Reconstruction Metrics with Clinical Utility105.9%0.0%—
5Anisotropic Diffusion Modeling for Brain Tumor Growth105.9%0.0%—
6Anomaly Detection via Generative Manifold Projection105.9%0.0%—
7Closing the Benchmark to Bedside Gap in Medical AI105.9%0.0%—
8Counterfactual Data Augmentation via Structural Interventions105.9%0.0%—
9Counterfactual Simulation for Causal Inference105.9%0.0%—
10Decentralized Model Training via Federated Learning105.9%0.0%—
11Decoupling Neural Feature Extraction from Final Classifiers105.9%0.0%—
12Diffeomorphic Flows for Topology Preserving Deformation105.9%0.0%—
13Distance-Aware Uncertainty Estimation105.9%0.0%—
14Efficient Deep Learning for Clinical Medical Imaging105.9%0.0%—
15Efficient Hyperparameter Optimization via Multi Fidelity and Online Tuning105.9%0.0%—
16Eliminating Spurious Feature Reliance in Medical Imaging Models105.9%0.0%—
17Federated Learning Robustness to Statistical Data Heterogeneity105.9%0.0%—
18Federated Learning under Decentralized Label Scarcity105.9%0.0%—
19Interpretable Anomaly Detection105.9%0.0%—
20Latent Label Inference in Deep Generative Models105.9%0.0%—
21Learning and Inference Under Relaxed Missingness Assumptions105.9%0.0%—
22Learning from Incomplete Multi-View Data105.9%0.0%—
23Learning from Limited Labeled Data in Medical Imaging105.9%0.0%—
24Learning from Privacy Constrained Distributed Data Silos105.9%0.0%—
25Learning Medical Image Segmentation from Limited Annotations105.9%0.0%—

Download all directions as CSV

Institutions

Institutions credited by The Journal of Machine Learning for Biomedical Imaging 2022 papers: a paper credits every institution named in its affiliations, once. 78 institutions share 17 credited papers.

#InstitutionPapersShare20212021 rank
1Imperial College London 529.4%0—
2Technical University of Munich 423.5%0—
3Massachusetts General Hospital 317.6%0—
4Massachusetts Institute of Technology 317.6%114
5Friedrich-Alexander-Universität Erlangen-Nürnberg 211.8%0—
6Institut national de recherche en sciences et technologies du numérique 211.8%0—
7McGill University 211.8%0—
8Mila - Quebec Artificial Intelligence Institute 211.8%0—
9University College London 211.8%17
10University of Bern 211.8%0—
11University of Copenhagen 211.8%21
12University of Oxford 211.8%0—
13University of Zurich 211.8%0—
1400bardy64 15.9%0—
15010842375 15.9%0—
1601q9sj412 15.9%0—
170280k9j44 15.9%0—
1802bv3zr67 15.9%0—
1904dm1cm79 15.9%0—
2005e8jge82 15.9%0—
2105hsqsk33 15.9%0—
22Athinoula A. Martinos Center for Biomedical Imaging 15.9%0—
23Centre Hospitalier Universitaire Vaudois 15.9%0—
24Centre Inria d'Université Côte d'Azur 15.9%0—
25Centre National de la Recherche Scientifique 15.9%0—
26Commissariat à l'Énergie Atomique et aux Énergies Alternatives 15.9%0—
27Cornell Tech 15.9%0—
28Cornell University 15.9%117
29Eindhoven University of Technology 15.9%0—
30Frederick National Laboratory for Cancer Research 15.9%0—

Download all institutions as CSV

About this data

Papers. The Journal of Machine Learning for Biomedical Imaging 2022 lists 17 papers on Lacuna, duplicate versions counted once; rejected and withdrawn submissions are left out.

Directions. Directions are Lacuna’s clustering of the concept elements extracted from each paper, so a paper can belong to several and direction counts add up to more than the paper count. A direction’s share is its papers divided by the edition’s papers. Comparisons with 2021 use shares, since the edition went from 6 to 17 papers.

Institutions. A paper credits each institution named in its affiliations, resolved to a ROR record, once. 17 of 17 papers (100.0%) credit at least one institution and are the base for institution shares; the rest have no resolved affiliation yet.

Data. Counts come from Lacuna’s data build of 2026-11-19 and change as papers are added. Tables as CSV: directions · institutions. Cite as: Lacuna, “The Journal of Machine Learning for Biomedical Imaging 2022”, https://lacuna.tiptreesystems.com/venue/the-journal-of-machine-learning-for-biomedical-imaging/2022.

Top Authors

AuthorPapers
Adrian V Dalca2
Ben Glocker2
Bernhard Kainz2
Tal Arbel2
Aasa Feragen1
Alexandre Gramfort1
Andrew Hoopes1
Baba C. Vemuri1
bjoern menze1
Bruce Fischl1
Chun-Hao Yang1
Daniel Rueckert1
Dou Quan1
Hugo Kuijf1
Jean-Philippe Thiran1
Jean-Pierre R. Falet1
John Guttag1
Jun Ma1
Konstantinos Kamnitsas1
Malte Hoffmann1

Impactful Papers

Papers from The Journal of Machine Learning for Biomedical Imaging 2022 that shaped research clusters.

Foundational

Influential

Papers

Title
A Differentially Private Probabilistic Framework for Modeling the Variability Across Federated Datasets of Heterogeneous Multi-View Observations
A Review of Causality for Learning Algorithms in Medical Image Analysis
Computing Multiple Image Reconstructions with a Single Hypernetwork
Detecting Outliers with Foreign Patch Interpolation
Distributional Gaussian Processes Layers for Out-of-Distribution Detection
Evaluation of 3D GANs for Lung Tissue Modelling in Pulmonary CT
Focused Decoding Enables 3D Anatomical Detection by Transformers
Influence of uncertainty estimation techniques on false-positive reduction in liver lesion detection
Inverting brain grey matter models with likelihood-free inference: a tool for trustable cytoarchitecture measurements
Learning the Effect of Registration Hyperparameters with HyperMorph
Nested Grassmannians for Dimensionality Reduction with Applications
Patch-based Medical Image Segmentation using Matrix Product State Tensor Networks
Quantifying Topology In Pancreatic Tubular Networks From Live Imaging 3D Microscopy
QU-BraTS: MICCAI BraTS 2020 Challenge on Quantifying Uncertainty in Brain Tumor Segmentation – Analysis of Ranking Scores and Benchmarking Results
Rethinking Generalization: The Impact of Annotation Style on Medical Image Segmentation
Semi-Supervised Federated Peer Learning for Skin Lesion Classification
Validation and Generalizability of Self-Supervised Image Reconstruction Methods for Undersampled MRI