Article

Using Resynthesized Natural Data to Probe Algorithmic Bias: A Study in De-Identification Tools

2023-09-26

Abstract excerpt

<h4>Purpose: </h4> We present a series of experiments designed to quantify the algorithmic bias inherent in six off-the-shelf de-identification systems. We frame the core measure of algorithmic bias for de-identification tools in terms of the following null hypothesis: an (unbiased) de-identification system will perform equally well regardless of the gender, race, or ethnicity of the patient. Methods. We test this...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 3b07471e-ec2e-5eda-94c5-f971145c34fc
DOI: 10.21203/rs.3.rs-3360291/v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Using Resynthesized Natural Data to Probe Algorithmic Bias: A Study in De-Identification ToolsDOI 10.21203/rs.3.rs-3360291/v1Read

Benchmarking transformer-based models for medical record deidentification: A single centre, multi-specialty evaluationSemantic proximity Read

Comparative evaluation of large-language models and purpose-built software for medical record de-identificationSemantic proximity Read

A Deep Learning Approach for Transgender and Gender Diverse Patient Identification in Electronic Health RecordsSemantic proximity Read

Building a Best-in-Class Automated De-identification Tool for Electronic Health Records Through Ensemble LearningSemantic proximity Read

De-identification of Brazilian Portuguese Clinical Records Using a Hybrid Pipeline with Local Language ModelsSemantic proximity Read

Algorithmic Penalty for Non-Native Voices: A Three-Arm Diagnostic Accuracy Audit of Five Commercial AI Text Detectors Against Pre-LLM Scientific Literature — Protocol for the AUDIT-AI StudySemantic proximity Read

Uncertainty Aware Llm Deidentification and Anonymization of Clinical NotesSemantic proximity Read

Active deep learning to detect demographic traits in free-form clinical notes.Semantic proximity Read

Drug or Pokémon? Large language model performance in identification of fabricated medicationsSemantic proximity Read

Automatic speech recognition alters diagnostic error and biomarker validity in AI-based dementia speech assessmentSemantic proximity Read

Responsible AI in NLP: GUS-Net Span-Level Bias Detection Dataset and Benchmark for Generalizations, Unfairness, and StereotypesSemantic proximity Read

Benchmark Evaluation of the WordBinary AI-Text Detection Model on 9,000 AI-Generated Texts and 2,000 Pre-2015 Human Academic-Paper SamplesSemantic proximity Read