Jenna Marie Reps

Can machine-learning improve cardiovascular risk prediction using routine clinical data?

Background Current approaches to predict cardiovascular risk fail to identify many people who would benefit from preventive treatment, while others receive unnecessary intervention. Machine-learning offers opportunity to improve accuracy by exploiting complex interactions between risk factors. We assessed whether machine-learning can improve cardiovascular risk prediction. Methods Prospective cohort study using routine clinical data of 378,256 patients from UK family practices, free from cardiovascular disease at outset. Four machine-learning algorithms (random forest, logistic regression, gradient boosting machines, neural networks) were compared to an established algorithm (American College of Cardiology guidelines) to predict first cardiovascular event over 10-years. Predictive accuracy was assessed by area under the ‘receiver operating curve’ (AUC); and sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) to predict 7.5% cardiovascular risk (threshold for initiating statins). Findings 24,970 incident cardiovascular events (6.6%) occurred. Compared to the established risk prediction algorithm (AUC 0.728, 95% CI 0.723–0.735), machine-learning algorithms improved prediction: random forest +1.7% (AUC 0.745, 95% CI 0.739–0.750), logistic regression +3.2% (AUC 0.760, 95% CI 0.755–0.766), gradient boosting +3.3% (AUC 0.761, 95% CI 0.755–0.766), neural networks +3.6% (AUC 0.764, 95% CI 0.759–0.769). The highest achieving (neural networks) algorithm predicted 4,998/7,404 cases (sensitivity 67.5%, PPV 18.4%) and 53,458/75,585 non-cases (specificity 70.7%, NPV 95.7%), correctly predicting 355 (+7.6%) more patients who developed cardiovascular disease compared to the established algorithm. Conclusions Machine-learning significantly improves accuracy of cardiovascular risk prediction, increasing the number of patients identified who could benefit from preventive treatment, while avoiding unnecessary treatment of others.

Comparing stochastic differential equations and agent-based modelling and simulation for early-stage cancer.

There is great potential to be explored regarding the use of agent-based modelling and simulation as an alternative paradigm to investigate early-stage cancer interactions with the immune system. It does not suffer from some limitations of ordinary differential equation models, such as the lack of stochasticity, representation of individual behaviours rather than aggregates and individual memory. In this paper we investigate the potential contribution of agent-based modelling and simulation when contrasted with stochastic versions of ODE models using early-stage cancer examples. We seek answers to the following questions: (1) Does this new stochastic formulation produce similar results to the agent-based version? (2) Can these methods be used interchangeably? (3) Do agent-based models outcomes reveal any benefit when compared to the Gillespie results? To answer these research questions we investigate three well-established mathematical models describing interactions between tumour cells and immune elements. These case studies were re-conceptualised under an agent-based perspective and also converted to the Gillespie algorithm formulation. Our interest in this work, therefore, is to establish a methodological discussion regarding the usability of different simulation approaches, rather than provide further biological insights into the investigated case studies. Our results show that it is possible to obtain equivalent models that implement the same mechanisms; however, the incapacity of the Gillespie algorithm to retain individual memory of past events affects the similarity of some results. Furthermore, the emergent behaviour of ABMS produces extra patters of behaviour in the system, which was not obtained by the Gillespie algorithm.

Discovering sequential patterns in a UK general practice database

The wealth of computerised medical information becoming readily available presents the opportunity to examine patterns of illnesses, therapies and responses. These patterns may be able to predict illnesses that a patient is likely to develop, allowing the implementation of preventative actions. In this paper sequential rule mining is applied to a General Practice database to find rules involving a patients age, gender and medical history. By incorporating these rules into current health-care a patient can be highlighted as susceptible to a future illness based on past or current illnesses, gender and year of birth. This knowledge has the ability to greatly improve health-care and reduce health-care costs.

Illness beliefs predict mortality in patients with diabetic foot ulcers

Background Patients’ illness beliefs have been associated with glycaemic control in diabetes and survival in other conditions. Objective We examined whether illness beliefs independently predicted survival in patients with diabetes and foot ulceration. Methods Patients (n = 169) were recruited between 2002 and 2007. Data on illness beliefs were collected at baseline. Data on survival were extracted on 1st November 2011. Number of days survived reflected the number of days from date of recruitment to 1st November 2011. Results Cox regressions examined the predictors of time to death and identified ischemia and identity beliefs (beliefs regarding symptoms associated with foot ulceration) as significant predictors of time to death. Conclusions Our data indicate that illness beliefs have a significant independent effect on survival in patients with diabetes and foot ulceration. These findings suggest that illness beliefs could improve our understanding of mortality risk in this patient group and could also be the basis for future therapeutic interventions to improve survival.

Signalling Paediatric Side Effects using an Ensemble of Simple Study Designs

BackgroundChildren are frequently prescribed medication ‘off-label’, meaning there has not been sufficient testing of the medication to determine its safety or effectiveness. The main reason this safety knowledge is lacking is due to ethical restrictions that prevent children from being included in the majority of clinical trials.ObjectiveThe objective of this paper is to investigate whether an ensemble of simple study designs can be implemented to signal acutely occurring side effects effectively within the paediatric population by using historical longitudinal data. The majority of pharmacovigilance techniques are unsupervised, but this research presents a supervised framework.MethodsMultiple measures of association are calculated for each drug and medical event pair and these are used as features that are fed into a classifier to determine the likelihood of the drug and medical event pair corresponding to an adverse drug reaction. The classifier is trained using known adverse drug reactions or known non-adverse drug reaction relationships. ResultsThe novel ensemble framework obtained a false positive rate of 0.149, a sensitivity of 0.547 and a specificity of 0.851 when implemented on a reference set of drug and medical event pairs. The novel framework consistently outperformed each individual simple study design.ConclusionThis research shows that it is possible to exploit the mechanism of causality and presents a framework for signalling adverse drug reactions effectively.

Comparing data-mining algorithms developed for longitudinal observational databases

Longitudinal observational databases have become a recent interest in the post marketing drug surveillance community due to their ability of presenting a new perspective for detecting negative side effects. Algorithms mining longitudinal observation databases are not restricted by many of the limitations associated with the more conventional methods that have been developed for spontaneous reporting system databases. In this paper we investigate the robustness of four recently developed algorithms that mine longitudinal observational databases by applying them to The Health Improvement Network (THIN) for six drugs with well document known negative side effects. Our results show that none of the existing algorithms was able to consistently identify known adverse drug reactions above events related to the cause of the drug and no algorithm was superior.

Refining adverse drug reaction signals by incorporating interaction variables identified using emergent pattern mining

PURPOSE To develop a framework for identifying and incorporating candidate confounding interaction terms into a regularised cox regression analysis to refine adverse drug reaction signals obtained via longitudinal observational data. METHODS We considered six drug families that are commonly associated with myocardial infarction in observational healthcare data, but where the causal relationship ground truth is known (adverse drug reaction or not). We applied emergent pattern mining to find itemsets of drugs and medical events that are associated with the development of myocardial infarction. These are the candidate confounding interaction terms. We then implemented a cohort study design using regularised cox regression that incorporated and accounted for the candidate confounding interaction terms. RESULTS The methodology was able to account for signals generated due to confounding and a cox regression with elastic net regularisation correctly ranking the drug families known to be true adverse drug reactions above those that are not. This was not the case without the inclusion of the candidate confounding interaction terms, where confounding leads to a non-adverse drug reaction being ranked highest. CONCLUSIONS The methodology is efficient, can identify high-order confounding interactions and does not require expert input to specify outcome specific confounders, so it can be applied for any outcome of interest to quickly refine its signals. The proposed method shows excellent potential to overcome some forms of confounding and therefore reduce the false positive rate for signal analysis using longitudinal data.

A Novel Semisupervised Algorithm for Rare Prescription Side Effect Discovery

Drugs are frequently prescribed to patients with the aim of improving each patients medical state, but an unfortunate consequence of most prescription drugs is the occurrence of undesirable side effects. Side effects that occur in more than one in a thousand patients are likely to be signaled efficiently by current drug surveillance methods, however, these same methods may take decades before generating signals for rarer side effects, risking medical morbidity or mortality in patients prescribed the drug while the rare side effect is undiscovered. In this paper, we propose a novel computational metaanalysis framework for signaling rare side effects that integrates existing methods, knowledge from the web,metric learning, and semisupervised clustering. The novel framework was able to signal many known rare and serious side effects for the selection of drugs investigated, such as tendon rupture when prescribed Ciprofloxacin or Levofloxacin, renal failure with Naproxen and depression associated with Rimonabant. Furthermore, for the majority of the drugs investigated it generated signals for rare side effects at a more stringent signaling threshold than existing methods and shows the potential to become a fundamental part of post marketing surveillance to detect rare side effects.

Attributes for causal inference in electronic healthcare databases

Side effects of prescription drugs present a serious issue. Existing algorithms that detect side effects generally require further analysis to confirm causality. In this paper we investigate attributes based on the Bradford-Hill causality criteria that could be used by a classifying algorithm to definitively identify side effects directly. We found that it would be advantageous to use attributes based on the association strength, temporality and specificity criteria.

Investigating distance metric learning in semi-supervised fuzzy c-means clustering

The idea behind distance metric learning (DML) is to accentuate the distance relations found in the training data, maintaining whether the data patterns are similar or dissimilar. In this paper, we investigate in using DML (GDML, LMNN, MCML and NCA) in semi-supervised Fuzzy c-means clustering and apply them on a real, biomedical dataset and on UCI datasets. We used a cross validation setting with varying amount of labelled data to test our methodology. Out of eight datasets, statistical significant improvement was found on five datasets using ssFCM with DML. This shows that DML can improve ssFCM clustering for some datasets. Further analysis using 2D PCA projection and sum of squared distances before and after DML transformation of the original data are carried out. Interestingly, DML was found to worsen ssFCM clustering in the NTBC dataset with hierarchical clusters.