mlsanityData to 14 Oct 2026

Benchmarking Embedding Aggregation Methods in Computational Pathology: A Clinical Data Perspective

arXiv 2407.07841 ↗First posted 10 Jul 2024cs.CVcited by 10

Not on the map. Why?

Abstract (arXiv)

Recent advances in artificial intelligence (AI), in particular self-supervised learning of foundation models (FMs), are revolutionizing medical imaging and computational pathology (CPath). A constant challenge in the analysis of digital Whole Slide Images (WSIs) is the problem of aggregating tens of thousands of tile-level image embeddings to a slide-level representation. Due to the prevalent use of datasets created for genomic research, such as TCGA, for method development, the performance of these techniques on diagnostic slides from clinical practice has been inadequately explored. This study conducts a thorough benchmarking analysis of ten slide-level aggregation techniques across nine clinically relevant tasks, including diagnostic assessment, biomarker classification, and outcome prediction. The results yield following key insights: (1) Embeddings derived from domain-specific (histological images) FMs outperform those from generic ImageNet-based models across aggregation methods. (2) Spatial-aware aggregators enhance the performance significantly when using ImageNet pre-trained models but not when using FMs. (3) No single model excels in all tasks and spatially-aware models do not show general superiority as it would be expected. These findings underscore the need for more adaptable and universally applicable aggregation techniques, guiding future research towards tools that better meet the evolving needs of clinical-AI in pathology. The code used in this work is available at urlhttps://github.com/fuchs-lab-public/CPathSABenchmark.

Key facts. Cited by 10 papers in the map (data through 14 Oct 2026). Builds on 23 of its 34 references that are in the map. First posted on arXiv 10 Jul 2024.

Builds on

23 of 34 references.

References · 23

  1. 2017Graph Attention Networkscited by 8,923
  2. 2018Attention-based Deep Multiple Instance Learningcited by 1,095
  3. 2020Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Imagescited by 745
  4. 2021TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classificationcited by 555
  5. 2020Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learningcited by 478
  6. 2022DTFD-MIL: Double-Tier Feature Distillation Multiple Instance Learning for Histopathology Whole Slide Image Classificationcited by 256
  7. 2020Whole Slide Images based Cancer Survival Prediction using Attention Guided Deep Multiple Instance Learning Networkscited by 178
  8. 2021Whole Slide Images are 2D Point Clouds: Context-Aware Survival Prediction using Patch-based Graph Convolutional Networkscited by 124
  9. 2023Artificial Intelligence for Digital and Computational Pathologycited by 108
  10. 2022A graph-transformer for whole slide image classificationcited by 84
  11. 2023Multiple Instance Learning Framework with Masked Hard Instance Mining for Whole Slide Image Classificationcited by 60
  12. 2022Bi-directional Weakly Supervised Knowledge Distillation for Whole Slide Image Classificationcited by 43
11 more
  1. 2023A General-Purpose Self-Supervised Model for Computational Pathologycited by 42
  2. 2023Computational Pathology at Health System Scale -- Self-Supervised Foundation Models from Three Billion Imagescited by 31
  3. 2022ReMix: A General and Efficient Framework for Multiple Instance Learning based Whole Slide Image Classificationcited by 30
  4. 2021DeepSMILE: Contrastive self-supervised pre-training benefits MSI and HRD classification directly from H&E whole-slide images in colorectal and breast cancercited by 27
  5. 2022An Aggregation of Aggregation Methods in Computational Pathologycited by 20
  6. 2018Terabyte-scale Deep Multiple Instance Learning for Classification and Localization in Pathologycited by 16
  7. 2023Prompt-MIL: Boosting Multi-Instance Learning Schemes via Task-specific Prompt Tuningcited by 15
  8. 2022Incorporating intratumoral heterogeneity into weakly-supervised deep learning models via variance poolingcited by 11
  9. 2022H&E-based Computational Biomarker Enables Universal EGFR Screening for Lung Adenocarcinomacited by 8
  10. 2022Kernel Attention Transformer (KAT) for Histopathology Whole Slide Image Classificationcited by 7
  11. 2023Multi-scale Efficient Graph-Transformer for Whole Slide Image Classificationcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 7; 2026: 2.

10 papers.

  1. 2025Revisiting Automatic Data Curation for Vision Foundation Models in Digital Pathologycited by 4
  2. 2024Pediatric brain tumor classification using digital histopathology and deep learning: evaluation of SOTA methods on a multi-center Swedish cohortcited by 1
  3. 2025Extreme Learning Machines for Attention-based Multiple Instance Learning in Whole-Slide Image Classificationcited by 0
  4. 2025Single GPU Task Adaptation of Pathology Foundation Models for Whole Slide Image Analysiscited by 0
  5. 2025Predict Patient Self-reported Race from Skin Histological Imagescited by 0
  6. 2025Identifying actionable driver mutations in lung cancer using an efficient Asymmetric Transformer Decodercited by 0
  7. 2025Revisiting Data Challenges of Computational Pathology: A Pack-based Multiple Instance Learning Training Frameworkcited by 0
  8. 2025PathBench-MIL: A Comprehensive AutoML and Benchmarking Framework for Multiple Instance Learning in Histopathologycited by 0
  9. 2026Reducing Redundancy in Whole-Slide Image Patching for Scalable Indexing and Retrievalcited by 0
  10. 2026Seeing Beyond the Lesion: Disease Recognition from Reactive CNS Tissuecited by 0