What question did this study set out to answer?

This research aims to assess how HiSAN-generated embeddings can reduce algorithmic bias in cancer classification models.

March 14, 2026Open Access

Mitigating Algorithmic Bias in Cancer Site Classification Models

Key Points

This research aims to assess how HiSAN-generated embeddings can reduce algorithmic bias in cancer classification models.
Utilized SEER data for generating HiSAN embeddings.
Applied post-training pruning as an audit method.
Examined the impact of demographic factors on cancer classification outcomes.
HiSAN embeddings were shown to classify cancer sites effectively.
Significant reductions in algorithmic bias were observed after using pruning techniques.

Abstract

Our findings demonstrate that HiSAN-generated embeddings from SEER data can be used effectively in cancer site classification without significant demographic bias influencing the outcomes. Post-training pruning therefore functions as a practical audit and sensitivity check.

Read Full Paperexternally

AIに質問

Bookmark

View Full Paper

Cite This Study

Shivanna et al. (Sun,) studied this question.

synapsesocial.com/papers/69b4ba3618185d8a39802f83 https://doi.org/https://doi.org/10.1200/cci-25-00250

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

AIに質問

Bookmark

View Full Paper