Papers
Topics
Authors
Recent
Search
2000 character limit reached

MiranDa: Mimicking the Learning Processes of Human Doctors to Achieve Causal Inference for Medication Recommendation

Published 23 Jul 2024 in cs.CY, cs.AI, and cs.LG | (2408.01445v1)

Abstract: To enhance therapeutic outcomes from a pharmacological perspective, we propose MiranDa, designed for medication recommendation, which is the first actionable model capable of providing the estimated length of stay in hospitals (ELOS) as counterfactual outcomes that guide clinical practice and model training. In detail, MiranDa emulates the educational trajectory of doctors through two gradient-scaling phases shifted by ELOS: an Evidence-based Training Phase that utilizes supervised learning and a Therapeutic Optimization Phase grounds in reinforcement learning within the gradient space, explores optimal medications by perturbations from ELOS. Evaluation of the Medical Information Mart for Intensive Care III dataset and IV dataset, showcased the superior results of our model across five metrics, particularly in reducing the ELOS. Surprisingly, our model provides structural attributes of medication combinations proved in hyperbolic space and advocated "procedure-specific" medication combinations. These findings posit that MiranDa enhanced medication efficacy. Notably, our paradigm can be applied to nearly all medical tasks and those with information to evaluate predicted outcomes. The source code of the MiranDa model is available at https://github.com/azusakou/MiranDa.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (95)
  1. Nick Barber. What constitutes good prescribing? Bmj, 310(6984):923–925, 1995.
  2. Methodological variations and their effects on reported medication administration error rates. BMJ quality & safety, 22(4):278–289, 2013.
  3. Systems analysis of adverse drug events. Jama, 274(1):35–43, 1995.
  4. Five years after to err is human: what have we learned? Jama, 293(19):2384–2390, 2005.
  5. Drug related problems in clinical practice: a cross-sectional study on their prevalence, risk factors and associated pharmaceutical interventions. Scientific reports, 11(1):1–11, 2021.
  6. Medication errors and adverse drug events in a uk hospital during the optimisation of electronic prescriptions: a prospective observational study. The Lancet Digital Health, 1(8):e403–e412, 2019.
  7. Medical error—the third leading cause of death in the us. Bmj, 353, 2016.
  8. The economic burden of preventable adverse drug reactions: a systematic review of observational studies. Expert opinion on drug safety, 17(7):681–695, 2018.
  9. Incidence, causes, and consequences of preventable adverse drug reactions occurring in inpatients: a systematic review of systematic reviews. PloS one, 13(10):e0205426, 2018.
  10. Adverse drug reaction deaths reported in united states vital statistics, 1999-2006. Annals of Pharmacotherapy, 46(2):169–175, 2012.
  11. Recent changes in drug poisoning mortality in the united states by urban–rural status and by drug type. Pharmacoepidemiology and drug safety, 17(10):997–1005, 2008.
  12. Trends in poisoning hospitalization and mortality in taiwan, 1999-2008: a retrospective analysis. BMC Public Health, 11:1–8, 2011.
  13. Peter A Chyka. How many deaths occur annually from adverse drug reactions in the united states? The American journal of medicine, 109(2):122–130, 2000.
  14. Association of hospitalist years of experience with mortality in the hospitalized medicare population. JAMA internal medicine, 178(2):196–203, 2018.
  15. Seventy-five trials and eleven systematic reviews a day: how will we ever keep up? PLoS medicine, 7(9):e1000326, 2010.
  16. Electronic prescribing improves medication safety in community-based office practices. Journal of general internal medicine, 25:530–536, 2010.
  17. Doctor ai: Predicting clinical events via recurrent neural networks. In Machine learning for healthcare conference, pages 301–318. PMLR, 2016.
  18. Change matters: Medication change prediction with recurrent residual networks. arXiv preprint arXiv:2105.01876, 2021.
  19. Medi-care ai: Predicting medications from billing codes via robust recurrent neural networks. Neural Networks, 124:109–116, 2020.
  20. Retain: An interpretable predictive model for healthcare using reverse time attention mechanism. Advances in neural information processing systems, 29, 2016.
  21. Leap: learning to prescribe effective and safe treatment combinations for multimorbidity. In proceedings of the 23rd ACM SIGKDD international conference on knowledge Discovery and data Mining, pages 1315–1324, 2017.
  22. Gamenet: Graph augmented memory networks for recommending medication combination. In proceedings of the AAAI Conference on Artificial Intelligence, volume 33, pages 1126–1133, 2019.
  23. Safedrug: Dual molecular graph encoders for recommending effective and safe drug combinations. arXiv preprint arXiv:2105.02711, 2021.
  24. Leveraging multiple types of domain knowledge for safe and effective drug recommendation. In Proceedings of the 31st ACM international conference on information & knowledge management, pages 2169–2178, 2022.
  25. Patient condition change network for safe medication recommendation. In 2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), pages 1046–1051. IEEE, 2022.
  26. Self-supervised adversarial distribution regularization for medication recommendation. In IJCAI, pages 3134–3140, 2021.
  27. Grasp: generic framework for health status representation learning based on incorporating knowledge from similar patients. In Proceedings of the AAAI conference on artificial intelligence, volume 35, pages 715–723, 2021.
  28. Deep reinforcement learning for sepsis treatment. arXiv preprint arXiv:1711.09602, 2017.
  29. Representation and reinforcement learning for personalized glycemic control in septic patients. arXiv preprint arXiv:1712.00654, 2017.
  30. Supervised reinforcement learning with recurrent neural network for dynamic treatment recommendation. In Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining, pages 2447–2456, 2018.
  31. Pervasive label errors in test sets destabilize machine learning benchmarks. arXiv preprint arXiv:2103.14749, 2021.
  32. Medication errors: an overview for clinicians. In Mayo Clinic Proceedings, volume 89, pages 1116–1125. Elsevier, 2014.
  33. MR Chen and HF Wang. The reason and prevention of hospital medication errors. Practical Journal of Clinical Medicine, 4, 2013.
  34. The incident reporting system does not detect adverse drug events: a problem for quality improvement. The Joint Commission journal on quality improvement, 21(10):541–548, 1995.
  35. A controlled trial of arthroscopic surgery for osteoarthritis of the knee. New England Journal of Medicine, 347(2):81–88, 2002.
  36. Relating human genetic variation to variation in drug responses. Trends in genetics, 28(10):487–495, 2012.
  37. Pharmacogenomics: the inherited basis for interindividual differences in drug response. Annual review of genomics and human genetics, 2(1):9–39, 2001.
  38. A second chance to get causal inference right: a classification of data science tasks. Chance, 32(1):42–49, 2019.
  39. Judea Pearl. Causal inference in statistics: An overview. Statistics Surveys, 3(none):96 – 146, 2009.
  40. Learning representations for counterfactual inference. In International conference on machine learning, pages 3020–3029. PMLR, 2016.
  41. Causal inference using deep-learning variable selection identifies and incorporates direct and indirect causalities in complex biological systems. bioRxiv, pages 2021–07, 2021.
  42. Causal effect inference with deep latent-variable models. Advances in neural information processing systems, 30, 2017.
  43. Ganite: Estimation of individualized treatment effects using generative adversarial nets. In International conference on learning representations, 2018.
  44. Amortized inference for causal structure learning. Advances in Neural Information Processing Systems, 35:13104–13118, 2022.
  45. Perfect match: A simple method for learning representations for counterfactual inference with neural networks. arXiv preprint arXiv:1810.00656, 2018.
  46. Bryan Lim. Forecasting treatment responses over time using recurrent marginal structural networks. Advances in neural information processing systems, 31, 2018.
  47. Causal machine learning for healthcare and precision medicine. Royal Society Open Science, 9(8):220638, 2022.
  48. Global comparators project: international comparison of hospital outcomes using administrative data. Health services research, 48(6pt1):2081–2100, 2013.
  49. Drug repositioning: identifying and developing new uses for existing drugs. Nature reviews Drug discovery, 3(8):673–683, 2004.
  50. Patient similarity analysis with longitudinal health data. arXiv preprint arXiv:2005.06630, 2020.
  51. Introduction to reinforcement learning, volume 135. MIT press Cambridge, 1998.
  52. Mimic-iii, a freely accessible critical care database. Scientific data, 3(1):1–9, 2016.
  53. Mimic-iii clinical database (version 1.4). https://doi.org/10.13026/C2XW26, 2016.
  54. Physiobank, physiotoolkit, and physionet: components of a new research resource for complex physiologic signals. circulation, 101(23):e215–e220, 2000.
  55. Mimic-iv (version 0.4). https://doi.org/10.13026/a3wn-hq05, 2020.
  56. Attention is all you need. Advances in neural information processing systems, 30, 2017.
  57. The microbiome as a human organ. Clinical Microbiology and Infection, 18:2–4, 2012.
  58. Icu procedures of the critically ill. Respirology, 14(8):1092–1097, 2009.
  59. Template matching for auditing hospital cost and quality. Health services research, 49(5):1446–1474, 2014.
  60. Enhanced recovery after surgery: a consensus review of clinical care for patients undergoing colonic resection. Clinical nutrition, 24(3):466–477, 2005.
  61. Prospect: evidence-based, procedure-specific postoperative pain management. Best practice & research Clinical anaesthesiology, 21(1):149–159, 2007.
  62. Reinforcement learning in healthcare: A survey. ACM Computing Surveys (CSUR), 55(1):1–36, 2021.
  63. Bayesian learning via stochastic gradient langevin dynamics. In Proceedings of the 28th international conference on machine learning (ICML-11), pages 681–688, 2011.
  64. Using auc and accuracy in evaluating learning algorithms. IEEE Transactions on knowledge and Data Engineering, 17(3):299–310, 2005.
  65. The area under the precision-recall curve as a performance metric for rare binary events. Methods in Ecology and Evolution, 10(4):565–577, 2019.
  66. Confidence interval for micro-averaged f 1 and macro-averaged f 1 scores. Applied Intelligence, 52(5):4961–4972, 2022.
  67. A probabilistic interpretation of precision, recall and f-score, with implication for evaluation. In European conference on information retrieval, pages 345–359. Springer, 2005.
  68. Luciano da F Costa. Further generalizations of the jaccard index. arXiv preprint arXiv:2110.09619, 2021.
  69. Ssi–ddi: substructure–substructure interactions for drug–drug interaction prediction. Briefings in Bioinformatics, 22(6):bbab133, 2021.
  70. Hyperbolic geometry of complex networks. Physical Review E, 82(3):036106, 2010.
  71. Sustaining the internet with hyperbolic mapping. Nature communications, 1(1):62, 2010.
  72. Random projection in dimensionality reduction: applications to image and text data. In Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining, pages 245–250, 2001.
  73. Nonlinear projection with curvilinear distances: Isomap versus curvilinear distance analysis. Neurocomputing, 57:49–76, 2004.
  74. Machine learning meets complex networks via coalescent embedding in the hyperbolic space. Nature communications, 8(1):1615, 2017.
  75. Umap: Uniform manifold approximation and projection for dimension reduction. arXiv preprint arXiv:1802.03426, 2018.
  76. Learning continuous hierarchies in the lorentz model of hyperbolic geometry. In International conference on machine learning, pages 3779–3788. PMLR, 2018.
  77. Poincaré maps for analyzing complex hierarchies in single-cell data. Nature communications, 11(1):2966, 2020.
  78. Stuart Lloyd. Least squares quantization in pcm. IEEE transactions on information theory, 28(2):129–137, 1982.
  79. Claude Sammut and Geoffrey I. Webb, editors. TF–IDF, pages 986–987. Springer US, Boston, MA, 2010.
  80. PyHealth: A deep learning toolkit for healthcare predictive modeling. In Proceedings of the 27th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD) 2023, 2023.
  81. C Zöllner and C Stein. Opioids. Analgesia, pages 31–63, 2007.
  82. Interaction between drugs and the gut microbiome. Gut, 69(8):1510–1519, 2020.
  83. Hypertension pharmacological treatment in adults: a world health organization guideline executive summary. Hypertension, 79(1):293–301, 2022.
  84. A comprehensive review on machine learning in healthcare industry: Classification, restrictions, opportunities and challenges. Sensors, 23(9):4178, 2023.
  85. On classifying continuous constraint satisfaction problems. In 2021 IEEE 62nd Annual Symposium on Foundations of Computer Science (FOCS), pages 781–791. IEEE, 2022.
  86. An overview of the action space for deep reinforcement learning. In Proceedings of the 2021 4th International Conference on Algorithms, Computing and Artificial Intelligence, pages 1–10, 2021.
  87. Program evaluation of population-and system-level policies: evidence for decision making. Medical Decision Making, 42(1):17–27, 2022.
  88. Unifying cardiovascular modelling with deep reinforcement learning for uncertainty aware control of sepsis treatment. PLOS Digital Health, 1(2):e0000012, 2022.
  89. Continuous state-space models for optimal sepsis treatment: a deep reinforcement learning approach. In Machine Learning for Healthcare Conference, pages 147–163. PMLR, 2017.
  90. Primary retroperitoneal masses: what is the differential diagnosis? Abdominal imaging, 40:1887–1903, 2015.
  91. Deep learning. nature, 521(7553):436–444, 2015.
  92. Adaptive perturbation-based gradient estimation for discrete latent variable models. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, pages 9200–9208, 2023.
  93. Gradient perturbation is underrated for differentially private convex optimization. arXiv preprint arXiv:1911.11363, 2019.
  94. Towards the unification and robustness of perturbation and gradient based explanations. In International Conference on Machine Learning, pages 110–119. PMLR, 2021.
  95. Perturbations, optimization, and statistics. MIT Press, 2016.

Summary

No one has generated a summary of this paper yet.

Sign Up to Summarize

Paper to Video (Beta)

No one has generated a video about this paper yet.

Sign Up to Generate All Videos Subscribe on YouTube

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Sign Up to Generate

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Generate Now

Continue Learning

We haven't generated follow-up questions for this paper yet.

Generate Now

Collections

Sign up for free to add this paper to one or more collections.

Sign Up