CoChat
Sign Up
Cited by 311Machine Learning in HealthcareRecommender Systems and TechniquesMedical Coding and Health Information

Pre-training of Graph Augmented Transformers for Medication Recommendation

Open access
Shang, Junyuan, Ma, Tengfei, Xiao, Cao, Sun, Jimeng·2019·Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence
DOI 10.24963/ijcai.2019/825
Open this paper in CoChat — free accountOpen PDF

Abstract

Medication recommendation is an important healthcare application. It is commonly formulated as a temporal prediction task. Hence, most existing works only utilize longitudinal electronic health records (EHRs) from a small number of patients with multiple visits ignoring a large number of patients with a single visit (selection bias). Moreover, important hierarchical knowledge such as diagnosis hierarchy is not leveraged in the representation learning process. Despite the success of deep learning techniques in computational phenotyping, most previous approaches have two limitations: task-oriented representation and ignoring hierarchies of medical codes. To address these challenges, we propose G-BERT, a new model to combine the power of Graph Neural Networks (GNNs) and BERT (Bidirectional Encoder Representations from Transformers) for medical code representation and medication recommendation. We use GNNs to represent the internal hierarchical structures of medical codes. Then we integrate the GNN representation into a transformer-based visit encoder and pre-train it on EHR data from patients only with a single visit. The pre-trained visit encoder and representation are then fine-tuned for downstream predictive tasks on longitudinal EHRs from patients with multiple visits. G-BERT is the first to bring the language model pre-training schema into the healthcare domain and it achieved state-of-the-art performance on the medication recommendation task.

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.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.24963/ijcai.2019/825

Cites

6 of 6 references verified0 citation errors · 0 not checkable

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.

A few free checks a day · papers already verified open instantly · results are public

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?

  • GBERT uses all single-visit samples as inputs, which are insufficient to capture long-term contextual information in the EHR, and uses only structured data.

    “the inputs for this model are all singlevisit samples, which are insufficient to capture longterm contextual information in the EHR.As in BEHRT, only structured data are used.”

    From Foresight-a generative pretrained transformer for modelling of patient timelines using electronic health records: a retrospective modelling study (2024) · cites this paper as [5]

    Sections 3 and 4.2 describe inputs made from diagnosis and medication codes, supporting the structured-data part. But Section 4.3 says pre-training uses visits from both single-visit and multi-visit records, and Section 4.4 says medication prediction aggregates embeddings from previous visits. The paper therefore does not support the claim that all inputs are single-visit samples or that G-BERT cannot use longitudinal context. (LLM, conf=0.85)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 18

  1. Inci M. Baytas, Cao Xiao, Xi Zhang, Fei Wang, Anil K. Jain, and Jiayu Zhou. Patient subtyp- ing via time-aware lstm networks. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowl- edge Discovery and Data Mining, 2017.
  2. Cho et al., 2014] Kyunghyun Cho, Bart Van Merriënboer, Dzmitry Bahdanau, and Yoshua Bengio. On the proper- ties of neural machine translation: Encoder-decoder ap- proaches. arXiv preprint arXiv:1409.1259, 2014. · arXiv
  3. Edward Choi, Mohammad Taha Ba- hadori, Jimeng Sun, Joshua Kulas, Andy Schuetz, and Walter Stewart. Retain: An interpretable predictive model for healthcare using reverse time attention mechanism. In Advances in Neural Information Processing Systems, pages 3504-3512, 2016.
  4. Edward Choi, Mohammad Taha Ba- hadori, Le Song, Walter F Stewart, and Jimeng Sun. Gram: Graph-based attention model for healthcare representation learning. In SIGKDD, 2017.
  5. In S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, and R. Garnett, editors, Advances in Neural Information Processing Systems 31, pages 4547- 4557.
  6. Dumitru Erhan, Yoshua Bengio, Aaron Courville, Pierre-Antoine Manzagol, Pascal Vincent, and Samy Bengio. Why does unsupervised pre-training help deep learning? Journal of Machine Learning Research, 11(Feb):625-660, 2010.
  7. J. Gilmer, S.S. Schoenholz, P.F. Riley, O. Vinyals, and G.E. Dahl. Neural message passing for quantum chemistry. In ICML, 2017.
  8. Geoffrey E Hinton, Simon Osindero, and Yee-Whye Teh. A fast learning algorithm for deep belief nets. Neural computation, 18(7):1527-1554, 2006.
  9. Thomas N Kipf and Max Welling. Semi-supervised classification with graph convolutional networks. In ICLR, 2017.
  10. Zachary C Lipton, David C Kale, Charles Elkan, and Randall Wetzel. Learning to diag- nose with lstm recurrent neural networks. arXiv preprint arXiv:1511.03677, 2015. · arXiv
  11. Luaces et al., 2012] Oscar Luaces, Jorge Díez, José Barran- quero, Juan José del Coz, and Antonio Bahamonde. Binary relevance efficacy for multilabel classification. Progress in Artificial Intelligence, 1(4):303-313, 2012.
  12. Ma et al., 2018] Tengfei Ma, Cao Xiao, and Fei Wang. Health-atm: A deep architecture for multifaceted patient health record representation and risk prediction. In Pro- ceedings of the 2018 SIAM International Conference on Data Mining, pages 261-269. SIAM, 2018.
  13. Matthew Peters, Mark Neumann, Mohit Iyyer, Matt Gardner, Christopher Clark, Kenton Lee, and Luke Zettlemoyer. Deep contextualized word representa- tions. In NAACL, volume 1, pages 2227-2237, 2018.
  14. Junyuan Shang, Cao Xiao, Tengfei Ma, Hongyan Li, and Jimeng Sun. Gamenet: Graph augmented memory networks for recommending medication combi- nation. AAAI, 2019.
  15. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. In Advances in Neural Information Processing Sys- tems, pages 5998-6008, 2017.
  16. Xiao et al., 2018a] Cao Xiao, Edward Choi, and Jimeng Sun. Opportunities and challenges in developing deep learning models using electronic health records data: a systematic review. Journal of the American Medical In- formatics Association, 2018.
  17. Xiao et al., 2018b] Cao Xiao, Tengfei Ma, Adji B. Dieng, David M. Blei, and Fei Wang. Readmission prediction via deep contextual embedding of clinical concepts. PLOS ONE, 13(4):1-15, 04 2018.
  18. Zhang et al., 2017] Yutao Zhang, Robert Chen, Jie Tang, Walter F Stewart, and Jimeng Sun. Leap: Learning to pre- scribe effective and safe treatment combinations for mul- timorbidity. In SIGKDD, pages 1315-1324, 2017.

Cited by

Sorted by citations

Authors
Shang, J., Ma, T., Xiao, C., & Sun, J.
Affiliations
  • Georgia Institute of Technology (US)
  • IBM (United States) (US)
  • IQVIA (United States) (US)
ORCID
Venue
Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence
Publisher
International Joint Conferences on Artificial Intelligence Organization
Published
2019-08
Pages
5953-5959
DOI
10.24963/ijcai.2019/825
OpenAlex
W2965570621
Cited by
311
References
6
Citations by year
2026: 21 · 2025: 50 · 2024: 64 · 2023: 54 · 2022: 33 · 2021: 72 · 2020: 16 · 2019: 1
FWCI
11.67
Type
conference paper
Language
en
Open access
Locations
  • International Joint Conference on Artificial Intelligence (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
Funders
National Science Foundation (1418511, 1838042, 1533768)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
  • Physical Sciences › Computer Science › Information Systems › Recommender Systems and Techniques
  • Health Sciences › Health Professions › Health Information Management › Medical Coding and Health Information
Keywords
drug recommendation, graph neural networks, BERT, electronic health records, selection bias, temporal prediction, pre-training

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