Hurtful Words: Quantifying Biases in Clinical Contextual Word Embeddings

• URL: http://arxiv.org/abs/2003.11515v1
• Date: Wed, 11 Mar 2020 23:21:14 GMT
• Title: Hurtful Words: Quantifying Biases in Clinical Contextual Word Embeddings
• Authors: Haoran Zhang, Amy X. Lu, Mohamed Abdalla, Matthew McDermott, Marzyeh Ghassemi
• Abstract summary: We pretrain deep embedding models (BERT) on medical notes from the MIMIC-III hospital dataset. We identify dangerous latent relationships that are captured by the contextual word embeddings. We evaluate performance gaps across different definitions of fairness on over 50 downstream clinical prediction tasks.
• Score: 16.136832979324467
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In this work, we examine the extent to which embeddings may encode marginalized populations differently, and how this may lead to a perpetuation of biases and worsened performance on clinical tasks. We pretrain deep embedding models (BERT) on medical notes from the MIMIC-III hospital dataset, and quantify potential disparities using two approaches. First, we identify dangerous latent relationships that are captured by the contextual word embeddings using a fill-in-the-blank method with text from real clinical notes and a log probability bias score quantification. Second, we evaluate performance gaps across different definitions of fairness on over 50 downstream clinical prediction tasks that include detection of acute and chronic conditions. We find that classifiers trained from BERT representations exhibit statistically significant differences in performance, often favoring the majority group with regards to gender, language, ethnicity, and insurance status. Finally, we explore shortcomings of using adversarial debiasing to obfuscate subgroup information in contextual word embeddings, and recommend best practices for such deep embedding models in clinical settings.

Related papers

• Contrastive Learning with Counterfactual Explanations for Radiology Report Generation [83.30609465252441]
  We propose a textbfCountertextbfFactual textbfExplanations-based framework (CoFE) for radiology report generation. Counterfactual explanations serve as a potent tool for understanding how decisions made by algorithms can be changed by asking what if'' scenarios. Experiments on two benchmarks demonstrate that leveraging the counterfactual explanations enables CoFE to generate semantically coherent and factually complete reports.
  arXiv  Detail & Related papers  (2024-07-19T17:24:25Z)
• Multi-task Explainable Skin Lesion Classification [54.76511683427566]
  We propose a few-shot-based approach for skin lesions that generalizes well with few labelled data. The proposed approach comprises a fusion of a segmentation network that acts as an attention module and classification network.
  arXiv  Detail & Related papers  (2023-10-11T05:49:47Z)
• Knowledge Graph Embeddings for Multi-Lingual Structured Representations of Radiology Reports [40.606143019674654]
  We introduce a novel light-weight graph-based embedding method specifically catering radiology reports. It takes into account the structure and composition of the report, while also connecting medical terms in the report. We show the use of this embedding on two tasks namely disease classification of X-ray reports and image classification.
  arXiv  Detail & Related papers  (2023-09-02T11:46:41Z)
• Conditional Supervised Contrastive Learning for Fair Text Classification [59.813422435604025]
  We study learning fair representations that satisfy a notion of fairness known as equalized odds for text classification via contrastive learning. Specifically, we first theoretically analyze the connections between learning representations with a fairness constraint and conditional supervised contrastive objectives.
  arXiv  Detail & Related papers  (2022-05-23T17:38:30Z)
• Clinical Named Entity Recognition using Contextualized Token Representations [49.036805795072645]
  This paper introduces the technique of contextualized word embedding to better capture the semantic meaning of each word based on its context. We pre-train two deep contextualized language models, Clinical Embeddings from Language Model (C-ELMo) and Clinical Contextual String Embeddings (C-Flair) Explicit experiments show that our models gain dramatic improvements compared to both static word embeddings and domain-generic language models.
  arXiv  Detail & Related papers  (2021-06-23T18:12:58Z)
• Estimating Redundancy in Clinical Text [6.245180523143739]
  Clinicians populate new documents by duplicating existing notes, then updating accordingly. quantifying information redundancy can play an essential role in evaluating innovations that operate on clinical narratives. We present and evaluate two strategies to measure redundancy: an information-theoretic approach and a lexicosyntactic and semantic model.
  arXiv  Detail & Related papers  (2021-05-25T11:01:45Z)
• Interpretable bias mitigation for textual data: Reducing gender bias in patient notes while maintaining classification performance [0.11545092788508224]
  We identify and remove gendered language from two clinical-note datasets. We show minimal degradation in health condition classification tasks for low- to medium-levels of bias removal via data augmentation. This work outlines an interpretable approach for using data augmentation to identify and reduce the potential for bias in natural language processing pipelines.
  arXiv  Detail & Related papers  (2021-03-10T03:09:30Z)
• Unifying Relational Sentence Generation and Retrieval for Medical Image Report Composition [142.42920413017163]
  Current methods often generate the most common sentences due to dataset bias for individual case. We propose a novel framework that unifies template retrieval and sentence generation to handle both common and rare abnormality.
  arXiv  Detail & Related papers  (2021-01-09T04:33:27Z)
• Exploring Text Specific and Blackbox Fairness Algorithms in Multimodal Clinical NLP [19.936882021907266]
  We investigate a modality-agnostic fairness algorithm - equalized odds post processing - and compare it to a text-specific fairness algorithm: debiased clinical word embeddings. Despite the fact that debiased word embeddings do not explicitly address equalized odds of protected groups, we show that a text-specific approach to fairness may simultaneously achieve a good balance of performance and classical notions of fairness.
  arXiv  Detail & Related papers  (2020-11-19T03:11:24Z)
• Semi-supervised Medical Image Classification with Relation-driven Self-ensembling Model [71.80319052891817]
  We present a relation-driven semi-supervised framework for medical image classification. It exploits the unlabeled data by encouraging the prediction consistency of given input under perturbations. Our method outperforms many state-of-the-art semi-supervised learning methods on both single-label and multi-label image classification scenarios.
  arXiv  Detail & Related papers  (2020-05-15T06:57:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.