{"doi":"10.59275/j.melba.2022-5aa9","title":"Compound Figure Separation of Biomedical Images: Mining Large Datasets for Self-supervised Learning","abstract":"With the rapid development of self-supervised learning (e.g., contrastive learning), the importance of having large-scale images (even without annotations) for training a more generalizable AI model has been widely recognized in medical image analysis. However, collecting large-scale task-specific unannotated data at scale can be challenging for individual labs. Existing online resources, such as digital books, publications, and search engines, provide a new resource for obtaining large-scale images. However, published images in healthcare (e.g., radiology and pathology) consist of a considerable amount of compound figures with subplots. In order to extract and separate compound figures into usable individual images for downstream learning, we propose a simple compound figure separation (SimCFS) framework without using the traditionally required detection bounding box annotations, with a new loss function and a hard case simulation. Our technical contribution is four-fold: (1) we introduce a simulation-based training framework that minimizes the need for resource extensive bounding box annotations; (2) we propose a new side loss that is optimized for compound figure separation; (3) we propose an intra-class image augmentation method to simulate hard cases; and (4) to the best of our knowledge, this is the first study that evaluates the efficacy of leveraging self-supervised learning with compound image separation. From the results, the proposed SimCFS achieved state-of-the-art performance on the ImageCLEF 2016 Compound Figure Separation Database. The pretrained self-supervised learning model using large-scale mined figures improved the accuracy of downstream image classification tasks with a contrastive learning algorithm. The source code of SimCFS is made publicly available at https://github.com/hrlblab/ImageSeperation.","journal":"The Journal of Machine Learning for Biomedical Imaging","year":2022,"id":293920,"datarank":0.10397207708399181,"base_score":0.6931471805599453,"endowment":0.6931471805599453,"self_citation_contribution":0.10397207708399181,"citation_network_contribution":0.0,"self_endowment_contribution":0.10397207708399181,"citer_contribution":0.0,"corpus_percentile":22.178386323199504,"corpus_rank":9377,"citation_count":1,"citer_count":1,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":true,"is_dataset_confidence":0.804,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2022-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":978912,"name":"Chang Qu","orcid":"0009-0006-7654-4321","position":1,"is_corresponding":false},{"id":978913,"name":"Jun Long","orcid":"0000-0002-6755-1312","position":2,"is_corresponding":false},{"id":978914,"name":"Quan Liu","orcid":"0000-0002-6730-539X","position":3,"is_corresponding":false},{"id":643160,"name":"Ruining Deng","orcid":"0000-0001-6300-8518","position":4,"is_corresponding":false},{"id":979346,"name":"Yuanhan Tian","orcid":null,"position":5,"is_corresponding":false},{"id":700453,"name":"Jiachen Xu","orcid":"0009-0003-4854-885X","position":6,"is_corresponding":false},{"id":750320,"name":"Aadarsh Jha","orcid":null,"position":7,"is_corresponding":false},{"id":978915,"name":"Zuhayr Asad","orcid":"0000-0002-2401-5026","position":8,"is_corresponding":false},{"id":425888,"name":"Shunxing Bao","orcid":"0000-0001-6376-4292","position":9,"is_corresponding":false},{"id":749470,"name":"Mengyang Zhao","orcid":"0000-0001-5952-1263","position":10,"is_corresponding":false},{"id":271631,"name":"Agnes B. Fogo","orcid":"0000-0003-3698-8527","position":11,"is_corresponding":false},{"id":104551,"name":"Bennett A. Landman","orcid":"0000-0001-5733-2127","position":12,"is_corresponding":false},{"id":340022,"name":"Haichun Yang","orcid":"0000-0003-4265-7492","position":13,"is_corresponding":false},{"id":251969,"name":"Catie Chang","orcid":"0000-0003-1541-9579","position":14,"is_corresponding":false},{"id":291228,"name":"Yuankai Huo","orcid":"0000-0002-2096-8065","position":15,"is_corresponding":false},{"id":978911,"name":"Tianyuan Yao","orcid":"0000-0002-1848-079X","position":0,"is_corresponding":true}],"reference_count":37,"raw_metadata":null,"created_at":"2026-07-19T00:30:57.469517Z","pmid":"37077404","pmcid":"PMC10112832","fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}