GRAM: Graph-based Attention Model for Healthcare Representation Learning
Open accessAbstract
accuracy, data needs, interpretability) of GRAM to various methods including the recurrent neural network (RNN) in two sequential diagnoses prediction tasks and one heart failure prediction task. Compared to the basic RNN, GRAM achieved 10% higher accuracy for predicting diseases rarely observed in the training data and 3% improved area under the ROC curve for predicting heart failure using an order of magnitude less training data. Additionally, unlike other methods, the medical concept representations learned by GRAM are well aligned with the medical ontology. Finally, GRAM exhibits intuitive attention behaviors by adaptively generalizing to higher level concepts when facing data insufficiency at the lower level concepts.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Two separate GNNs may be used to learn representations of protein and small molecule graphs for predicting drug-target affinity.
“Also, two separate GNNs may be used to learn representations of protein and small molecule graphs for predicting drug-target affinity [173].”
From Graph representation learning in biomedicine and healthcare (2022) · cites this paper as [173]
Fusing compound sequence, structure, and clinical implications has significantly improved drug-drug and drug-target interaction predictions.
“Fusing compound sequence, structure, and clinical implications has significantly improved drug-drug and drug-target interaction predictions.”
From Graph representation learning in biomedicine and healthcare (2022) · cites this paper as [173]
An attention mechanism was adopted in 13 of the 49 included studies to improve performance and provide a degree of interpretability.
“To improve performance and provide a degree of interpretability to the model, an attention mechanism is adopted in 13 studies [6,14,27,31,43,57,64,72,76–78,80,88].”
From Deep representation learning of patient data from Electronic Health Records (EHR): A systematic review (2021) · cites this paper as [27]
A hierarchical knowledge graph was incorporated into learning representations in 6 of the 49 included studies.
“A hierarchical knowledge graph has been incorporated into learning representations in 6 studies [27,31–33,81,91].”
From Deep representation learning of patient data from Electronic Health Records (EHR): A systematic review (2021) · cites this paper as [27]
8 papers visualized with t-SNE to interpret the representations, and one study visualized with MDS.
“We observed 8 papers visualized with t-SNE to interpret the representations [14,27,31,60,66,73,75,79], and one study visualized with MDS [70].”
From Deep representation learning of patient data from Electronic Health Records (EHR): A systematic review (2021) · cites this paper as [27]
RETAIN was commonly used as a baseline in subsequent advanced models (GRAM, Dipole, Health-ATM).
“RETAIN was also commonly used as baselines in the subsequent series of advanced models (i.e., GRAM [27], Dipole [76], Health-ATM [72]).”
From Deep representation learning of patient data from Electronic Health Records (EHR): A systematic review (2021) · cites this paper as [27]
These methods include developing additional components to models for capturing a hierarchical representation from event-level, to visit-level, to patient-level; for encoding the longitudinal factors between clinical events; for incorporating domain knowledge into the representation; and for adopting multiple modalities of EHR data.
“These methods include developing additional components to models for capturing a hierarchical representation from event-level, to visit-level, to patient-level [81,112]; for encoding the longitudinal factors between clinical events [65,77]; for incorporating domain knowledge into the representation [27,31–33,91]; and for adopting multiple modalities of EHR data [11,14,57].”
From Deep representation learning of patient data from Electronic Health Records (EHR): A systematic review (2021) · cites this paper as [27]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 46 · Cited by 637
Research works printed but not linked: 2
- [10]Choi Edward, Bahadori Mohammad Taha, Schuetz Andy, Stewart Walter F., and Sun Jimeng. 2016. RETAIN: Interpretable Predictive Model in Healthcare using Reverse Time Attention Mechanism. In NIPS.
- [38]The Theano Development Team. 2016. Theano: A Python framework for fast computation of mathematical expressions. arXiv:1605.02688 (2016). · arXiv
- Authors
- Choi, E., Bahadori, M. T., Song, L., Stewart, W. F., & Sun, J.
- Affiliations
- Georgia Institute of Technology (US)
- Sutter Health (US)
- Venue
- Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
- Publisher
- ACM
- Published
- 2017-08-04
- Volume
- 2017
- Pages
- 787-795
- PMC
- PMC7954122
- OpenAlex
- W2557074642
- Cited by
- 637
- References
- 46
- Citations by year
- 2026: 38 · 2025: 64 · 2024: 70 · 2023: 79 · 2022: 77 · 2021: 102 · 2020: 86 · 2019: 76 · 2018: 42 · 2017: 3
- FWCI
- 25.82
- Type
- conference paper
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Locations
- ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) (publishedVersion)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- Funders
- National Institutes of Health
- Nvidia
- UCB
- Intel
- Office of Naval Research (N00014-15-1-2340)
- Children's Healthcare of Atlanta
- Google Faculty Award
- National Science Foundation
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Biomedical Text Mining and Ontologies
- Keywords
- gram, diagnostic prediction, heart failure prediction, recurrent neural networks, attention mechanism, insufficient data, interpretability
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
