CoChat

Bias and Class Imbalance in Oncologic Data—Towards Inclusive and Transferrable AI in Large Scale Oncology Data Sets

Open access
Tasci, Erdal, Zhuge, Ying, Camphausen, Kevin, Krauze, Andra V. · 2022 · Cancers

The DOI matches its Crossref record; 65 of the 66 references it cites were found in a registry or an index.

Abstract

Recent technological developments have led to an increase in the size and types of data in the medical field derived from multiple platforms such as proteomic, genomic, imaging, and clinical data. Many machine learning models have been developed to support precision/personalized medicine initiatives such as computer-aided detection, diagnosis, prognosis, and treatment planning by using large-scale medical data. Bias and class imbalance represent two of the most pressing challenges for machine learning-based problems, particularly in medical (e.g., oncologic) data sets, due to the limitations in patient numbers, cost, privacy, and security of data sharing, and the complexity of generated data. Depending on the data set and the research question, the methods applied to address class imbalance problems can provide more effective, successful, and meaningful results. This review discusses the essential strategies for addressing and mitigating the class imbalance problems for different medical data types in the oncologic domain.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/cancers14122897

Cites

65 of 65 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 The number of publications on class imbalance listed in Pubmed by year for years 2012 to 2021 (blue = total publications per year on class imbalance, orange = total publications per year on class imbalance in oncology, grey = the number of publications on class imbalance pertaining to oncology as a percent of the total number of publications that year) [8].
Figure 1 The number of publications on class imbalance listed in Pubmed by year for years 2012 to 2021 (blue = total publications per year on class imbalance, orange = total publications per year on class imbalance in oncology, grey = the number of publications on class imbalance pertaining to oncology as a percent of the total number of publications that year) [8].CC BY · PubMed Central
Figure 2 Clinical, imaging, omic and outcome based algorithms (left panel) are generated based on data sets that harbor underrepresentation or overrepresentation of one or several categories as related to demographics (gender, race, social determinants of health–(right top panel), management (extent of surgical resection, type of upfront management, management upon recurrence (right middle panel) and disease characteristics (molecular classification, response to systemic management and radiation therapy (right lower panel). Class imbalance affects ancillary data sets such as imaging, omics based on biospecimens and outcomes (middle panel). The resulting algorithms result in limited reproducibility in other cohorts (Study population, B, C, and a theoretical infinite other population that does not share identical class imbalance). As both the sources and extent of over and underrepresentation of certain classes and class imbalance are altered in other populations/data sets (lower panel), defining features reflective of class imbalance and mitigating these via compensatory methods in all data subtypes, can address the lack of reproducibility and help identify additional features that can be used to further optimize algorithms allowing for transferrable results (lower panel).
Figure 2 Clinical, imaging, omic and outcome based algorithms (left panel) are generated based on data sets that harbor underrepresentation or overrepresentation of one or several categories as related to demographics (gender, race, social determinants of health–(right top panel), management (extent of surgical resection, type of upfront management, management upon recurrence (right middle panel) and disease characteristics (molecular classification, response to systemic management and radiation therapy (right lower panel). Class imbalance affects ancillary data sets such as imaging, omics based on biospecimens and outcomes (middle panel). The resulting algorithms result in limited reproducibility in other cohorts (Study population, B, C, and a theoretical infinite other population that does not share identical class imbalance). As both the sources and extent of over and underrepresentation of certain classes and class imbalance are altered in other populations/data sets (lower panel), defining features reflective of class imbalance and mitigating these via compensatory methods in all data subtypes, can address the lack of reproducibility and help identify additional features that can be used to further optimize algorithms allowing for transferrable results (lower panel).CC BY · PubMed Central
Figure 3 An overview of the existing methods for the class imbalance problem.
Figure 3 An overview of the existing methods for the class imbalance problem.CC BY · PubMed Central
Figure 4 The overview of the feature selection methods.
Figure 4 The overview of the feature selection methods.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 65 references

Research works printed but not linked: 4

  1. (2021, December 01). National Library of Medicine, Available online: https://pubmed.ncbi.nlm.nih.gov.
  2. Abraham, 2013, A Review of Class Imbalance Problem, J. Netw. Innov. Comput., 1, 332
  3. Jaeger, P.F., Kohl, S.A.A., Bickelhaupt, S., Isensee, F., Kuder, T.A., Schlemmer, H.-P., and Maier-Hein, K.H. Retina U-Net: Embarrassingly Simple Exploitation of Segmentation Supervision for Medical Object Detection. Proceedings of the Machine Learning for Health NeurIPS Workshop, Proceedings of Machine Learning Research, Available online: http://proceedings.mlr.press/v116/jaeger20a/jaeger20a.pdf.
  4. Le, D.N.T., Le, H.X., Ngo, L., and Ngo, H.T. (2020). Transfer learning with class-weighted and focal loss function for automatic skin cancer classification. arXiv.

Metadata

Authors
Tasci, E., Zhuge, Y., Camphausen, K., & Krauze, A. V.
Affiliations
  • Center for Cancer Research, National Cancer Institute, NIH, Building 10, Bethesda, MD 20892, USA
  • National Institutes of Health (US)
Corresponding author
Krauze, Andra V.
Venue
Cancers
Publisher
MDPI AG
Published
2022-06-12
Volume
14
Issue
12
Pages
2897
ISSN
2072-6694 (ISSN-L)
OpenAlex
W4282548202
Cited by
117
References
66
Citations by year
2026: 38 · 2025: 21 · 2024: 28 · 2023: 23 · 2022: 7
FWCI
1.38
Type
journal article
Language
en
Open access
Locations
  • Cancers (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
Funders
NCI NIH intramural program (ZID BC 010990)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Imbalanced Data Classification Techniques
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
Keywords
class imbalance, algorithmic bias, medical datasets, precision medicine, computer-aided diagnosis, transfer learning, data privacy, data sharing

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Bias and Class Imbalance in Oncologic Data—Towards Inclusive and Transferrable AI in Large Scale Oncology Data Sets. https://app.cochat.ai/papers/10.3390/cancers14122897
The paper
Tasci, E., Zhuge, Y., Camphausen, K., & Krauze, A. V. (2022). Bias and Class Imbalance in Oncologic Data—Towards Inclusive and Transferrable AI in Large Scale Oncology Data Sets. Cancers. https://doi.org/10.3390/cancers14122897

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.