Attention-Based Imputation of Missing Values in Electronic Health Records Tabular Data

• Published in: 2024 IEEE 12th International Conference on Healthcare Informatics (ICHI) Vol. 2024; pp. 177 - 182
• Main Authors: Kowsar, Ibna, Rabbani, Shourav B., Samad, Manar D.
• Format: Conference Proceeding Journal Article
• Language: English
• Published: United States IEEE 01.06.2024
• Subjects: between-sample attention; Biostatistics; Contrastive learning; Data models; electronic health records; Imputation; Machine learning; Medical services; missing value imputation; Predictive models; self-attention; tabular data; self-attention; between-sample attention; electronic health records; tabular data; missing value imputation
• ISSN: 2575-2626; 2575-2634
• DOI: 10.1109/ICHI61247.2024.00030

The imputation of missing values (IMV) in electronic health records tabular data is crucial to enable machine learning for patient-specific predictive modeling. While IMV methods are developed in biostatistics and recently in machine learning, deep learning-based solutions have shown limited success in learning tabular data. This paper proposes a novel attention-based missing value imputation framework that learns to reconstruct data with missing values leveraging between-feature (self-attention) or between-sample attentions. We adopt data manipulation methods used in contrastive learning to improve the generalization of the trained imputation model. The proposed self-attention imputation method outperforms state-of-the-art statistical and machine learning-based (decision-tree) imputation methods, reducing the normalized root mean squared error by 18.4% to 74.7% on five tabular data sets and 52.6% to 82.6 % on two electronic health records data sets. The proposed attention-based missing value imputation method shows superior performance across a wide range of missingness (10 % to 50 %) when the values are missing completely at random.