Comparative visual analytics for assessing medical records with sequence embedding
Machine learning for data-driven diagnosis has been actively studied in medicine to provide better healthcare. Supporting analysis of a patient cohort similar to a patient under treatment is a key task for clinicians to make decisions with high confidence. However, such analysis is not straightforwa...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2020-06-01
|
| Series: | Visual Informatics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2468502X20300139 |
| _version_ | 1819041271583342592 |
|---|---|
| author | Rongchen Guo Takanori Fujiwara Yiran Li Kelly M. Lima Soman Sen Nam K. Tran Kwan-Liu Ma |
| author_facet | Rongchen Guo Takanori Fujiwara Yiran Li Kelly M. Lima Soman Sen Nam K. Tran Kwan-Liu Ma |
| author_sort | Rongchen Guo |
| collection | DOAJ |
| description | Machine learning for data-driven diagnosis has been actively studied in medicine to provide better healthcare. Supporting analysis of a patient cohort similar to a patient under treatment is a key task for clinicians to make decisions with high confidence. However, such analysis is not straightforward due to the characteristics of medical records: high dimensionality, irregularity in time, and sparsity. To address this challenge, we introduce a method for similarity calculation of medical records. Our method employs event and sequence embeddings. While we use an autoencoder for the event embedding, we apply its variant with the self-attention mechanism for the sequence embedding. Moreover, in order to better handle the irregularity of data, we enhance the self-attention mechanism with consideration of different time intervals. We have developed a visual analytics system to support comparative studies of patient records. To make a comparison of sequences with different lengths easier, our system incorporates a sequence alignment method. Through its interactive interface, the user can quickly identify patients of interest and conveniently review both the temporal and multivariate aspects of the patient records. We demonstrate the effectiveness of our design and system with case studies using a real-world dataset from the neonatal intensive care unit of UC Davis. |
| first_indexed | 2024-12-21T09:22:21Z |
| format | Article |
| id | doaj.art-d1a6a204968d4e85b31e2655bdb6de30 |
| institution | Directory Open Access Journal |
| issn | 2468-502X |
| language | English |
| last_indexed | 2024-12-21T09:22:21Z |
| publishDate | 2020-06-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Visual Informatics |
| spelling | doaj.art-d1a6a204968d4e85b31e2655bdb6de302022-12-21T19:08:59ZengElsevierVisual Informatics2468-502X2020-06-01427285Comparative visual analytics for assessing medical records with sequence embeddingRongchen Guo0Takanori Fujiwara1Yiran Li2Kelly M. Lima3Soman Sen4Nam K. Tran5Kwan-Liu Ma6Department of Computer Science, Beihang University, Beijing, China; Corresponding author.Department of Computer Science, University of California, Davis, United StatesDepartment of Computer Science, University of California, Davis, United StatesDepartment of Pathology and Laboratory Medicine, University of California, Davis, United StatesDepartment of Surgery, University of California, Davis, United StatesDepartment of Pathology and Laboratory Medicine, University of California, Davis, United StatesDepartment of Computer Science, University of California, Davis, United StatesMachine learning for data-driven diagnosis has been actively studied in medicine to provide better healthcare. Supporting analysis of a patient cohort similar to a patient under treatment is a key task for clinicians to make decisions with high confidence. However, such analysis is not straightforward due to the characteristics of medical records: high dimensionality, irregularity in time, and sparsity. To address this challenge, we introduce a method for similarity calculation of medical records. Our method employs event and sequence embeddings. While we use an autoencoder for the event embedding, we apply its variant with the self-attention mechanism for the sequence embedding. Moreover, in order to better handle the irregularity of data, we enhance the self-attention mechanism with consideration of different time intervals. We have developed a visual analytics system to support comparative studies of patient records. To make a comparison of sequences with different lengths easier, our system incorporates a sequence alignment method. Through its interactive interface, the user can quickly identify patients of interest and conveniently review both the temporal and multivariate aspects of the patient records. We demonstrate the effectiveness of our design and system with case studies using a real-world dataset from the neonatal intensive care unit of UC Davis.http://www.sciencedirect.com/science/article/pii/S2468502X20300139Electronic medical recordsEvent sequence dataAutoencoderSelf-attentionSequence similarityVisual analytics |
| spellingShingle | Rongchen Guo Takanori Fujiwara Yiran Li Kelly M. Lima Soman Sen Nam K. Tran Kwan-Liu Ma Comparative visual analytics for assessing medical records with sequence embedding Visual Informatics Electronic medical records Event sequence data Autoencoder Self-attention Sequence similarity Visual analytics |
| title | Comparative visual analytics for assessing medical records with sequence embedding |
| title_full | Comparative visual analytics for assessing medical records with sequence embedding |
| title_fullStr | Comparative visual analytics for assessing medical records with sequence embedding |
| title_full_unstemmed | Comparative visual analytics for assessing medical records with sequence embedding |
| title_short | Comparative visual analytics for assessing medical records with sequence embedding |
| title_sort | comparative visual analytics for assessing medical records with sequence embedding |
| topic | Electronic medical records Event sequence data Autoencoder Self-attention Sequence similarity Visual analytics |
| url | http://www.sciencedirect.com/science/article/pii/S2468502X20300139 |
| work_keys_str_mv | AT rongchenguo comparativevisualanalyticsforassessingmedicalrecordswithsequenceembedding AT takanorifujiwara comparativevisualanalyticsforassessingmedicalrecordswithsequenceembedding AT yiranli comparativevisualanalyticsforassessingmedicalrecordswithsequenceembedding AT kellymlima comparativevisualanalyticsforassessingmedicalrecordswithsequenceembedding AT somansen comparativevisualanalyticsforassessingmedicalrecordswithsequenceembedding AT namktran comparativevisualanalyticsforassessingmedicalrecordswithsequenceembedding AT kwanliuma comparativevisualanalyticsforassessingmedicalrecordswithsequenceembedding |