Exploring Text Specific and Blackbox Fairness Algorithms in Multimodal Clinical NLP

John Chen, Ian Berlot-Attwell, Xindi Wang, Safwan Hossain, Frank Rudzicz


Abstract
Clinical machine learning is increasingly multimodal, collected in both structured tabular formats and unstructured forms such as free text. We propose a novel task of exploring fairness on a multimodal clinical dataset, adopting equalized odds for the downstream medical prediction tasks. To this end, 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 classical notions of fairness. Our work opens the door for future work at the critical intersection of clinical NLP and fairness.
Anthology ID:
2020.clinicalnlp-1.33
Volume:
Proceedings of the 3rd Clinical Natural Language Processing Workshop
Month:
November
Year:
2020
Address:
Online
Venue:
ClinicalNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
301–312
Language:
URL:
https://aclanthology.org/2020.clinicalnlp-1.33
DOI:
10.18653/v1/2020.clinicalnlp-1.33
Bibkey:
Cite (ACL):
John Chen, Ian Berlot-Attwell, Xindi Wang, Safwan Hossain, and Frank Rudzicz. 2020. Exploring Text Specific and Blackbox Fairness Algorithms in Multimodal Clinical NLP. In Proceedings of the 3rd Clinical Natural Language Processing Workshop, pages 301–312, Online. Association for Computational Linguistics.
Cite (Informal):
Exploring Text Specific and Blackbox Fairness Algorithms in Multimodal Clinical NLP (Chen et al., ClinicalNLP 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2020.clinicalnlp-1.33.pdf
Video:
 https://slideslive.com/38939838
Code
 johntiger1/multimodal_fairness
Data
MIMIC-III