Health Informatics via Machine Learning for the Clinical Management of Patients

Summary

Objectives: To review how health informatics systems based on machine learning methods have impacted the clinical management of patients, by affecting clinical practice.

Methods: We reviewed literature from 2010-2015 from databases such as Pubmed, IEEE xplore, and INSPEC, in which methods based on machine learning are likely to be reported. We bring together a broad body of literature, aiming to identify those leading examples of health informatics that have advanced the methodology of machine learning. While individual methods may have further examples that might be added, we have chosen some of the most representative, informative exemplars in each case.

Results: Our survey highlights that, while much research is taking place in this high-profile field, examples of those that affect the clinical management of patients are seldom found. We show that substantial progress is being made in terms of methodology, often by data scientists working in close collaboration with clinical groups.

Conclusions: Health informatics systems based on machine learning are in their infancy and the translation of such systems into clinical management has yet to be performed at scale.

Keywords

Health informatics - data mining - information systems - electronic health records

• References

• 1 Vincent JL. Critical care: where have we been and where are we going?. Crit Care 2013; 17 (Suppl. 01) S2.
• 2 Knaus WA, Zimmerman JE, Wagner DP, Draper EA, Lawrence DE. APACHE, acute physiology and chronic health evaluation: a physiologically based classification system. Crit Care Med 1981; 9 (Suppl. 08) 591-7.
• 3 Johnson AEW, Kramer AE, Clifford GD. A new severity of illness scale using a subset of acute physiology and chronic health evaluation data elements shows comparable predictive accuracy. Crit Care Med 2013; 41 (Suppl. 07) 1711-8.
• 4 Siontis GCM, Tzoulaki I, Ioannidis JPA. Predicting death: an empirical evaluation of predictive tools for mortality. Arch Intern Med 2011; 171 (Suppl. 19) 1721-6.
• 5 Saria S, Rajani AK, Gould J, Koller D, Penn AA. Integration of early physiological responses predicts later illness severity in preterm infants. Sci Transl Med 2010; 2 (Suppl. 48) 48-65.
• 6 Lehman LW, Adams RP, Mayaud L, Moody GB, Malhotra A, Mark RG. A physiological time series dynamics-based approach to patient monitoring and outcome prediction. IEEE J Biomed Health Inform 2015; 19 (Suppl. 03) 1068-76.
• 7 Moorman JR, Carlo WA, Kattwinkel J, Schelonka RL, Porcelli PJ, Navarrete CT. et al. Mortality reduction by heart rate characteristic monitoring in very low birth weight neonates: a randomized trial. J Pediatr 2011; 159 (Suppl. 06) 900-6.
• 8 Ghassemi M, Naumann T, Joshi R, Rumshisky A. Topic models for mortality modelling in intensive care units. Proceedings of International Conference on Machine Learning; 2012 p. 1-4
• 9 Ghassemi M, Naumann T, Doshi-Velez F, Brim-mer N, Joshi R, Rumshisky A. et al. Unfolding physiological state: mortality modelling in intensive care units. Proceedings of Knowledge Discovery and Data Mining; 2014 p. 75-84
• 10 Cleverley K, Mousavi N, Stronger L, Ann-Bordun K, Hall L, Tam JW. et al. The impact of telemetry on survival of in-hospital cardiac arrests in non-critical care patients. Resuscitation 2013; 84 (Suppl. 07) 878-82.
• 11 Chen EH. Appropriate use of telemetry monitoring in hospitalized patients. Current Emergency Hospital Medical Reports 2013; 2 (Suppl. 01) 52-6.
• 12 Benjamin E, Klugman RA, Luckmann R, Fairchild DG, Abookire SA. Impact of cardiac telemetry on patient safety and cost. Am J Manag Care 2013; 19 (Suppl. 06) e225-232.
• 13 Bulger J, Nickel W, Messler J, Goldstein J, O‘ Callaghan J, Auron M. et al. Choosing wisely in adult hospital medicine: five opportunities for improved healthcare value. J Hosp Med 2013; 8 (Suppl. 09) 486-92.
• 14 Falun N, Nordrehaug JE, Hoff PI, Langorgen J, Moons P, Norekval TM. Evaluation of the appropriateness and outcome of in-hospital telemetry monitoring. Am J Cardiol 2013; 112 (Suppl. 08) 1219-23.
• 15 Walsh JA, Topol EJ, Steinhubl SR. Novel wireless devices for cardiac monitoring. Circulation 2014; 130 (Suppl. 07) 573-81.
• 16 Najafi N, Auerbach A. Use and outcomes of telemetry monitoring on a medicine service. Arch Intern Med 2012; 172 (Suppl. 17) 1349-50.
• 17 Bonnici T, Charlton P, Alastruey J, Tarassenko L, Watkinson P, Beale R. Continuous physiological monitoring of ambulatory patients. Proceedings of MECBioeng 2014: 38
• 18 Welch J, Moon J, McCombie S. Early detection of the deteriorating patient: the case for a multi-parameter patient-worn monitor. Biomed Instrum Technol 2012; 46 s2 57-64.
• 19 Pimentel MAF, Clifton DA, Tarassenko L. Gaussian process clustering for the functional characterisation of vital-sign trajectories. Proceedings of IEEE Machine Learning for Signal Processing 2013; 1-6.
• 20 Clifton DA, Clifton L, Hugueny S, Wong D, Tarassenko L. An extreme function theory for novelty detection. IEEE J Sel Top Signal Process 2013; 7 (Suppl. 01) 28-37.
• 21 Pimentel MAF, Clifton DA, Clifton L, Watkinson PJ, Tarassenko L. Modelling physiological deterioration in post-operative patient vital-sign data. Med Biol Eng Comput 2013; 51: 869-77.
• 22 Duerichen R, Pimentel MAF, Clifton L, Schweikard A, Clifton DA. Multi-task Gaussian processes for multivariate physiological time-series analysis. IEEE Trans Biomed Eng 2015; Jan 62 (Suppl. 01) 314-22.
• 23 Clifton L, Clifton DA, Pimentel MAF, Watkinson PJ, Tarassenko L. Predictive monitoring of mobile patients by combining clinical observations with data from wearable sensors. IEEE J Biomed Health Inform 2014; 18 (Suppl. 03) 722-30.
• 24 Clifton L, Clifton DA, Zhang Y, Watkinson PJ, Tarassenko L, Yin H. Probabilistic novelty detection with support vector machines. IEEE Transactions on Reliability 2014; 63 (Suppl. 02) 455-67.
• 25 Clifton L, Clifton DA, Pimentel MAF, Watkinson PJ, Tarassenko L. Gaussian processes for person-alised e-health monitoring with wearable sensor. IEEE Trans Biomed Eng 2013; 60 (Suppl. 01) 193-7.
• 26 Clifton DA, Hugueny S, Clifton L, Tarassenko L. Extending the generalised Pareto distribution for novelty detection in high-dimensional spaces. J Signal Process Syst 2014; 74: 323-39.
• 27 Clifton DA, Hugueny S, Tarassenko L. Novelty detection with multivariate extreme value statistics. J Signal Process Syst 2011; 65: 371-89.
• 28 Clifton DA, Wong D, Clifton L, Pullinger R, Tarassenko L. A large-scale clinical validation of an integrated monitoring system in the Emergency Department. IEEE J Biomed Health Inform 2013; 17 (Suppl. 04) 835-42.
• 29 Heaton MJ, Peng RD. Flexible distributed lag models using random functions with application to estimating mortality displacement from heat-related deaths. J Agric Biol Environ Stat 2012; 17 (Suppl. 03) 313-31.
• 30 Heaton MJ, Sain SR, Monaghan AJ, Wilhelmi OV, Hayden MH. An analysis of an incomplete marked point pattern of heat-related 911 calls. J Am Stat Assoc 2015; 110 (Suppl. 509) 123-35.
• 31 Lloyd C, Gunter T, Osborne MA, Roberts SJ. Variational inference for Gaussian process modulated Poisson processes. Proceedings of International Conference on Machine Learning; 2015 p. 1-9
• 32 Lasko TA. Efficient inference of Gaussian process modulated renewal processes with application to medical event data. Proceedings of Uncertainty in Artificial Intelligence 2014 p. 36
• 33 Donnelly N, Hunniford T, Harper R, Flynn A, Kennedy A, Branagh D. et al. Demonstrating the accuracy of an in-hospital ambulatory patient monitoring solution in measuring respiratory rate. Conf Proc IEEE Eng Med Biol Soc 2013; 6711-5.
• 34 OBS Medical Ltd.. Visensia Mobile and its innovative software for automated respiration rate calculation. Available: http://www.obsmedical.com/news/article/visensia-mobile-and-its-innovative-software-for-automated-respiration-rate. Accessed: 30-Nov-2014.
• 35 Pimentel MAF, Charlton PH, Clifton DA. Probabilistic estimation of respiratory rate from wearable sensors. In: Wearable Electronic Sensors. Mukhopadhyay S. editor. Smart Sensors, Measurement and Instrumentation. 2015. 15: 241-62.
• 36 Pimentel MAF, Clifton DA, Clifton L, Tarassenko L. Probabilistic estimation of respiratory rate using Gaussian processes. Conf Proc IEEE Eng Med Biol Soc 2013; 2902-5.
• 37 Orphanidou C, Bonnici T, Charlton P, Clifton DA, Vallance D, Tarassenko L. Signal quality indices for the electrocardiogram and photoplethysmo-gram: derivation and applications to wireless monitoring. IEEE J Biomed Health Inform 2015; 19 (Suppl. 03) 832-8.
• 38 Sempeles S. Continuous wireless monitoring device passes test in first hospital use. J Clin Eng 2013; 38 (Suppl. 03) 86-7.
• 39 Denny JC. Mining electronic health records in the genomics era. PLoS Comput Biol 2012; 8 (Suppl. 12) e1002823.
• 40 Jensen PB, Jensen LJ. Brunak Mining electronic health records: towards better research applications and clinical care. Nat Rev Genet 2012; 13 (Suppl. 06) 395-405.
• 41 Simpao AF, Ahumada LM, Galvez JA, Rehman MA. A review of analytics and clinical informatics in health care. J Med Syst 2014; 38 (Suppl. 04) 45.
• 42 Bright TJ, Wong A, Dhurjati R, Bristow E, Bastian L, Coeytaux RR. et al. Effect of clinical decision-support systems: a systematic review. Ann Intern Med 2012; 157 (Suppl. 01) 29-43.
• 43 Dewey FE, Grove ME, Pan C, Goldstein BA, Bernstein JA, Chaib H. et al. Clinical interpretation and implications of whole-genome sequencing. J Am Med Assoc 2014; 311 (Suppl. 10) 1035-45.
• 44 Paxton C, Niculescu-Mizil A, Saria S. Developing predictive models using electronic medical records: challenges and pitfalls. AMIA Annu Symp Proc 2013; 1109-15.
• 45 Hripcsak G, Albers DJ. Next-generation phenotyping of electronic health records. J Am Med Inform Assoc 2013; 20 (Suppl. 01) 117-21.
• 46 Gottesman O, Kuivaniemi H, Tromp G, Faucett WA, Li R, Manolio TA. et al. The Electronic Medical Records and Genomics (eMERGE) Network: past, present, and future. Genet Med 2013; 15 (Suppl. 10) 761-71.
• 47 Rasmussen-Torvik LJ, Stallings SC, Gordon AS, Almoguera B, Basford MA, Bielinski SJ. et al. Design and anticipated outcomes of the eMERGE-PGx project: a multicenter pilot for preemptive pharmacogenomics in electronic health record systems. Clin Pharmacol Ther 2014; 96 (Suppl. 04) 482-9.
• 48 Wang Z, Shah AD, Tate AR, Denaxas S, Shawe-Taylor J, Hemingway H. Extracting diagnoses and investigation results from unstructured text in electronic health records by semi-supervised machine learning. PLoS One 2012; 7 (Suppl. 01) e30412
• 49 Shivade C, Raghavan P, Fosler-Lussier E, Embi PJ, Elhadad N, Honson SB. et al. A review of approaches to identifying patient phenotype cohorts using electronic health records. J Am Med Inform Assoc 2014; 21 (Suppl. 02) 221-30.
• 50 Kohane IS, Churchill SE, Murphy SN. A translational engine at the national scale: informatics for integrating biology and the bedside. J Am Med Inform Assoc 2012; 19 (Suppl. 02) 181-5.
• 51 Kohane I. Deeper, longer phenotyping to accelerate the discovery of the genetic architectures of diseases. Genome Biol 2014; 15 (Suppl. 05) 115.
• 52 Loscalzo J, Kohane I, Barabasi AL. Human disease classification in the postgenomic era: a complex systems approach to human pathobiology. Mol Syst Biol 2007; 3: 124.
• 53 Ho JC, Ghosh J, Steinhubl S, Stewart WF, Denny JC, Malin BA, Sun J. Limestone: High-throughput candidate phenotype generation via tensor factorization. J Biomed Inform 2014; 52: 199-211.
• 54 Doshi-Velez F, Ge Y, Kohane I. Comorbidity clusters in autism spectrum disorders: an electronic health record time-series analysis. Pediatrics 2014; 133 (Suppl. 01) e54-63.
• 55 Schulam P, Wigley F, Saria S. Clustering longitudinal clinical marker trajectories from electronic health data: applications to phenotyping and endo-type discovery. Proceedings of AAAI Conference on Artificial Intelligence; 2015 p. 1-9
• 56 Lasko TA, Denn JC, Levy MA. Computational phenotype discovery using unsupervised feature learning over noisy, sparse, and irregular clinical data. PloS One 2013; 8 (Suppl. 06) e66341.
• 57 Huang SH, LePendu P, Iyer SV, Tai-Seale M, Carrell D, Shah NH. Toward personalizing treatment for depression: predicting diagnosis and severity. J Med Inform Assoc 2014; 21 (Suppl. 06) 1069-75.
• 58 Wiens J, Guttag J, Horvitz E. Patient risk stratification for hospital-associated C. diff as a time-series classification task. Proceedings of Neural Information Processing Systems 2012 p. 476-84
• 59 van der Heijden M, Velikova M, Lucas PJ. Learning Bayesian networks for clinical time series analysis. J Biomed Inform 2014; 48: 94-105.
• 60 Li L, Ruau DJ, Patel CJ, Weber SC, Chen R, Tatonetti NP. et al. Disease risk factors identified through shared genetic architecture and electronic medical records. Sci Transl Med 2014; 6 (Suppl. 234) 234ra57.
• 61 Van de Vijver MJ, He YD, van ’t Veer L. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med 2002; 347 (Suppl. 25) 1999-2009.
• 62 Ding H, Wang C, Huang K, Machiraju R. iGPSe: a visual analytic system for integrative genomic based cancer patient stratification. BMC Bioinformatics 2014; 15: 203.
• 63 Das J, Gayvert KM, Yu H. Predicting cancer prognosis using functional genomics data sets. Cancer Inform 2014; 13 S5 85-8.
• 64 Didelot X, Bowden R, Wilson DJ, Peto TE, Crook DW. Transforming clinical microbiology with bacterial genome sequencing. Nat Rev Genet 2012; 13 (Suppl. 09) 601-12.
• 65 Stoesser N, Batty EM, Eyre DW, Morgan M, Wyllie DH, Del Ojo Elias C. et al. Predicting antimicrobial susceptibilities for Escherichia coli and Klebsiella pneumoniae isolates using whole genomic sequence data. J Antimicrob Chemother 2013; 68 (Suppl. 10) 2234-44.
• 66 Gordon NC, Price JR, Cole K, Everitt R, Morgan M, Finney J. et al. Prediction of Staphylococcus aureus antimicrobial resistance by whole-genome sequencing. J Clin Microbiol 2014; 52 (Suppl. 04) 1182-91.
• 67 Niehaus KE, Walker TM, Crook DW, Peto TE, Clifton DA. Machine learning for the prediction of antibacterial susceptibility in Mycobacterium tuberculosis. Proceedings of IEEE Biomedical Health Informatics; 2014 p. 618-21