{"@context":"https://schema.org","@type":"CreativeWork","@id":"https://forgecascade.org/public/capsules/5db02a38-ca3a-4125-9843-f23287e1b307","identifier":"5db02a38-ca3a-4125-9843-f23287e1b307","url":"https://forgecascade.org/public/capsules/5db02a38-ca3a-4125-9843-f23287e1b307","name":"Closing the Domain Gap in Biomedical Imaging by In-Context Control Samples","text":"# Closing the Domain Gap in Biomedical Imaging by In-Context Control Samples\n\nSource-backed public reference for biomedical imaging, domain adaptation, meta-learning.\n\nSummary: The paper studies batch effects in biomedical imaging and proposes using negative control samples as in-context anchors for model adaptation. Its CS-ARM-BN method is evaluated on mechanism-of-action classification over the JUMP-CP dataset.\n\nKey points:\n- Focuses on technical batch effects that undermine reproducibility in bioimaging models.\n- Uses routinely available control samples to stabilize adaptation to new experimental batches.\n- Reports recovery of much of the accuracy lost when models move to new batches or labs.\n\nPublic review note: Source-backed technical ML reference; not diagnostic or treatment advice.\n\nSource: https://arxiv.org/abs/2604.20824\nAuthors: Ana Sanchez-Fernandez, Thomas Pinetz, Werner Zellinger, Guenter Klambauer\nPublished: 2026-04-22","keywords":["machine-learning","biomedical-imaging","domain-adaptation","reproducibility"],"about":[],"citation":[],"isPartOf":{"@type":"Dataset","name":"Forge Cascade Knowledge Graph","url":"https://forgecascade.org"},"publisher":{"@type":"Organization","name":"Forge Cascade","url":"https://forgecascade.org"},"dateCreated":"2026-05-24T14:27:14.607940Z","dateModified":"2026-06-19T01:59:49.343691Z","isBasedOn":"https://arxiv.org/abs/2604.20824","additionalProperty":[{"@type":"PropertyValue","name":"trust_level","value":100},{"@type":"PropertyValue","name":"verification_status","value":"sources_verified"},{"@type":"PropertyValue","name":"provenance_status","value":"valid"},{"@type":"PropertyValue","name":"evidence_level","value":"primary_source"},{"@type":"PropertyValue","name":"content_hash","value":"dac02cb97d8c2563a73559b191e115f0e1e23c034fe871d5b07851183357e01e"}]}