A Novel Kernel Based Approach to Arbitrary Length Symbolic Data with Application to Type 2 Diabetes Risk

Overview

Journal Sci Rep

Specialty Science

Date 2022 Mar 24

PMID 35322076

Authors

Nnanyelugo Nwegbu

Santosh Tirunagari

David Windridge

Affiliations

Soon will be listed here.

Abstract

Predictive modeling of clinical data is fraught with challenges arising from the manner in which events are recorded. Patients typically fall ill at irregular intervals and experience dissimilar intervention trajectories. This results in irregularly sampled and uneven length data which poses a problem for standard multivariate tools. The alternative of feature extraction into equal-length vectors via methods like Bag-of-Words (BoW) potentially discards useful information. We propose an approach based on a kernel framework in which data is maintained in its native form: discrete sequences of symbols. Kernel functions derived from the edit distance between pairs of sequences may then be utilized in conjunction with support vector machines to classify the data. Our method is evaluated in the context of the prediction task of determining patients likely to develop type 2 diabetes following an earlier episode of elevated blood pressure of 130/80 mmHg. Kernels combined via multi kernel learning achieved an F1-score of 0.96, outperforming classification with SVM 0.63, logistic regression 0.63, Long Short Term Memory 0.61 and Multi-Layer Perceptron 0.54 applied to a BoW representation of the data. We achieved an F1-score of 0.97 on MKL on external dataset. The proposed approach is consequently able to overcome limitations associated with feature-based classification in the context of clinical data.

References

Haasdonk B . Feature space interpretation of SVMs with indefinite kernels. IEEE Trans Pattern Anal Mach Intell. 2005; 27(4):482-492. DOI: 10.1109/TPAMI.2005.78. View

Barber S, Davies M, Khunti K, Gray L . Risk assessment tools for detecting those with pre-diabetes: a systematic review. Diabetes Res Clin Pract. 2014; 105(1):1-13. DOI: 10.1016/j.diabres.2014.03.007. View

Jia D, Zhang D, Li N . Pulse waveform classification using support vector machine with Gaussian time warp edit distance kernel. Comput Math Methods Med. 2014; 2014:947254. PMC: 3934457. DOI: 10.1155/2014/947254. View

Soguero-Ruiz C, Hindberg K, Mora-Jimenez I, Rojo-Alvarez J, Skrovseth S, Godtliebsen F . Predicting colorectal surgical complications using heterogeneous clinical data and kernel methods. J Biomed Inform. 2016; 61:87-96. DOI: 10.1016/j.jbi.2016.03.008. View

Basile A, Ritchie M . Informatics and machine learning to define the phenotype. Expert Rev Mol Diagn. 2018; 18(3):219-226. PMC: 6080627. DOI: 10.1080/14737159.2018.1439380. View

Loosli G, Canu S, Ong C . Learning SVM in Kreĭn Spaces. IEEE Trans Pattern Anal Mach Intell. 2015; 38(6):1204-16. DOI: 10.1109/TPAMI.2015.2477830. View

Grant P, Cosentino F . The 2019 ESC Guidelines on diabetes, pre-diabetes, and cardiovascular diseases developed in collaboration with the EASD: New features and the ‘Ten Commandments’ of the 2019 Guidelines are discussed by Professor Peter J. Grant and Professor Francesco.... Eur Heart J. 2019; 40(39):3215-3217. DOI: 10.1093/eurheartj/ehz687. View

Norgeot B, Glicksberg B, Trupin L, Lituiev D, Gianfrancesco M, Oskotsky B . Assessment of a Deep Learning Model Based on Electronic Health Record Data to Forecast Clinical Outcomes in Patients With Rheumatoid Arthritis. JAMA Netw Open. 2019; 2(3):e190606. PMC: 6484652. DOI: 10.1001/jamanetworkopen.2019.0606. View

Dai W, Brisimi T, Adams W, Mela T, Saligrama V, Paschalidis I . Prediction of hospitalization due to heart diseases by supervised learning methods. Int J Med Inform. 2014; 84(3):189-97. PMC: 4314395. DOI: 10.1016/j.ijmedinf.2014.10.002. View

10.

Knowler W, Barrett-Connor E, Fowler S, Hamman R, Lachin J, Walker E . Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002; 346(6):393-403. PMC: 1370926. DOI: 10.1056/NEJMoa012512. View

11.

Bisaso K, Anguzu G, Karungi S, Kiragga A, Castelnuovo B . A survey of machine learning applications in HIV clinical research and care. Comput Biol Med. 2017; 91:366-371. DOI: 10.1016/j.compbiomed.2017.11.001. View

12.

Marteau P, Gibet S . On recursive edit distance kernels with application to time series classification. IEEE Trans Neural Netw Learn Syst. 2014; 26(6):1121-33. DOI: 10.1109/TNNLS.2014.2333876. View

13.

Perveen S, Shahbaz M, Keshavjee K, Guergachi A . Prognostic Modeling and Prevention of Diabetes Using Machine Learning Technique. Sci Rep. 2019; 9(1):13805. PMC: 6760163. DOI: 10.1038/s41598-019-49563-6. View

14.

Pham T, Tran T, Phung D, Venkatesh S . Predicting healthcare trajectories from medical records: A deep learning approach. J Biomed Inform. 2017; 69:218-229. DOI: 10.1016/j.jbi.2017.04.001. View

15.

Chu J, Dong W, He K, Duan H, Huang Z . Using neural attention networks to detect adverse medical events from electronic health records. J Biomed Inform. 2018; 87:118-130. DOI: 10.1016/j.jbi.2018.10.002. View

16.

Wang Y, Wu P, Liu Y, Weng C, Zeng D . Learning Optimal Individualized Treatment Rules from Electronic Health Record Data. Proc (IEEE Int Conf Healthc Inform). 2017; 2016:65-71. PMC: 5423731. DOI: 10.1109/ICHI.2016.13. View

17.

Nguyen B, Pham H, Tran H, Nghiem N, Nguyen Q, Do T . Predicting the onset of type 2 diabetes using wide and deep learning with electronic health records. Comput Methods Programs Biomed. 2019; 182:105055. DOI: 10.1016/j.cmpb.2019.105055. View

18.

Li H, Jiang T . A class of edit kernels for SVMs to predict translation initiation sites in eukaryotic mRNAs. J Comput Biol. 2005; 12(6):702-18. DOI: 10.1089/cmb.2005.12.702. View

19.

Wang F . Adaptive semi-supervised recursive tree partitioning: The ART towards large scale patient indexing in personalized healthcare. J Biomed Inform. 2015; 55:41-54. DOI: 10.1016/j.jbi.2015.01.009. View

20.

Singh A, Nadkarni G, Gottesman O, Ellis S, Bottinger E, Guttag J . Incorporating temporal EHR data in predictive models for risk stratification of renal function deterioration. J Biomed Inform. 2014; 53:220-8. PMC: 4520404. DOI: 10.1016/j.jbi.2014.11.005. View