{
  "abstract": "Machine learning bias is a persistent challenge because it can create unfair outcomes, limit generalization, and reduce trust in real-world applications. A key source of this problem is shortcut learning, where models exploit signals linked to sensitive attributes, such as the data source or collection site, instead of relying on task-relevant features. To address this, we propose the Deceptive Signal metric, a novel quantitative measure designed to assess the extent of a model’s reliance on hidden shortcuts during the learning process. This metric is derived via the Deceptive Bias Detection pipeline, which isolates shortcut dependence by contrasting the model’s behavior under two controlled conditions: (1) Full Exclusion, where a sensitive subgroup is completely removed from training; and (2) Partial Exclusion, where the model has limited access to specific classes within that subgroup. By calculating the behavioral shift between these settings, the Deceptive Signal metric provides a quantitative estimate of the model’s susceptibility to learning task-irrelevant patterns. In experiments with the TCGA histopathology dataset, our metric successfully quantified substantial dependencies on center-specific artifacts in models trained for cancer classification. Author summary: Deep learning models are becoming powerful tools in healthcare, but they often suffer from a critical vulnerability: they can get the right answer for the wrong reason. In medical imaging, an AI might correctly identify a tumor not by analyzing the tissue, but by recognizing irrelevant digital markers unique to the specific hospital or scanner that produced the image. This phenomenon, known as shortcut learning, makes AI systems appear accurate at first glance while potentially limiting reliability in real-world patient care settings.To solve this, our research moves beyond simple accuracy checks and introduces a specific quantitative metric for shortcut learning. We developed a testing framework that forces the model into controlled training scenarios, deliberately withholding specific shortcut information to see how the model reacts. By mathematically comparing the model’s behavior across these scenarios, we calculate a quantitative score that indicates the magnitude of the model’s dependence on irrelevant patterns. This metric allows us to put a concrete number on the model’s trustworthiness, helping assess whether predictions are more strongly driven by task-relevant biological features rather than background artifacts.",
  "abstract_shown": true,
  "affiliations": [
    "Department of Electrical, Computer and Software Engineering, Ontario Tech University, Oshawa, Ontario, Canada",
    "Department of Engineering, Brock University, St. Catharines, Ontario, Canada"
  ],
  "article_type": "Research Article",
  "arxiv": null,
  "author_total": 3,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Farnaz Kheiri"
    },
    {
      "affs": [
        1
      ],
      "name": "Shahryar Rahnamayan"
    },
    {
      "name": "Masoud Makrehchi"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/ex6whtfgjd6cr.bib",
  "cited_by": 1,
  "doi": "10.1371/journal.pdig.0001165",
  "fulltext_url": "https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001165",
  "id": "ex6whtfgjd6cr",
  "issns": [
    "2767-3170"
  ],
  "landing_url": "https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001165",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "journal",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "openalex",
      "1702|AI in cancer detection"
    ],
    [
      "plos",
      "/Biology and life sciences/Neuroscience/Cognitive science/Cognitive psychology/Learning"
    ],
    [
      "plos",
      "/Biology and life sciences/Neuroscience/Cognitive science/Cognitive psychology/Learning/Human learning"
    ],
    [
      "plos",
      "/Biology and life sciences/Neuroscience/Learning and memory/Learning"
    ],
    [
      "plos",
      "/Biology and life sciences/Neuroscience/Learning and memory/Learning/Human learning"
    ],
    [
      "plos",
      "/Biology and life sciences/Psychology/Behavior"
    ],
    [
      "plos",
      "/Biology and life sciences/Psychology/Cognitive psychology/Learning"
    ],
    [
      "plos",
      "/Biology and life sciences/Psychology/Cognitive psychology/Learning/Human learning"
    ],
    [
      "plos",
      "/Computer and information sciences/Artificial intelligence/Machine learning"
    ],
    [
      "plos",
      "/Computer and information sciences/Artificial intelligence/Machine learning/Deep learning"
    ],
    [
      "plos",
      "/Medicine and health sciences/Oncology"
    ],
    [
      "plos",
      "/Medicine and health sciences/Oncology/Cancers and neoplasms"
    ],
    [
      "plos",
      "/Medicine and health sciences/Oncology/Cancers and neoplasms/Lung and intrathoracic tumors"
    ],
    [
      "plos",
      "/Social sciences/Psychology/Behavior"
    ],
    [
      "plos",
      "/Social sciences/Psychology/Cognitive psychology/Learning"
    ],
    [
      "plos",
      "/Social sciences/Psychology/Cognitive psychology/Learning/Human learning"
    ]
  ],
  "openalex": "W4417346112",
  "pdf_url": "https://journals.plos.org/digitalhealth/article/file?id=10.1371/journal.pdig.0001165&type=printable",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-03",
  "publisher": "Public Library of Science",
  "ris": "https://research.draconitic.com/paper/ex6whtfgjd6cr.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:17:56Z",
      "native_id": "10.1371/journal.pdig.0001165",
      "source": "plos",
      "url": "https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001165"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "plos:neuroscience",
      "subfield": "neuroscience"
    },
    {
      "field": "medicine",
      "origin": "plos:oncology",
      "subfield": "oncology"
    }
  ],
  "title": "Deceptive bias measurement in deep learning: Assessing shortcut reliance in TCGA cancer models",
  "title_text": "Deceptive bias measurement in deep learning: Assessing shortcut reliance in TCGA cancer models",
  "type": "peer-reviewed",
  "updated_at": "2026-10-30T12:17:56Z",
  "url": "https://research.draconitic.com/paper/ex6whtfgjd6cr",
  "venue": "PLOS Digital Health",
  "versions": []
}