Search

Author(s): Zhaozhi Qian [sup.1], Ahmed M. Alaa [sup.2], Mihaela van der Schaar [sup.1], Ari Ercole [sup.3] Author Affiliations: (1) grid.5335.0, 0000000121885934, University of Cambridge Centre for Mathematical Sciences, , Wilberforce [...]

Clinical decision making needs to be supported by evidence that treatments are beneficial to individual patients. Although randomized control trials (RCTs) are the gold standard for testing and introducing new drugs, due to the focus on specific questions with respect to establishing efficacy and safety vs. standard treatment, they do not provide a full characterization of the heterogeneity in the final intended treatment population. Conversely, real‐world observational data, such as electronic health records (EHRs), contain large amounts of clinical information about heterogeneous patients and their response to treatments. In this paper, we introduce the main opportunities and challenges in using observational data for training machine learning methods to estimate individualized treatment effects and make treatment recommendations. We describe the modeling choices of the state‐of‐the‐art machine learning methods for causal inference, developed for estimating treatment effects both in the cross‐section and longitudinal settings. Additionally, we highlight future research directions that could lead to achieving the full potential of leveraging EHRs and machine learning for making individualized treatment recommendations. We also discuss how experimental data from RCTs and Pharmacometric and Quantitative Systems Pharmacology approaches can be used to not only improve machine learning methods, but also provide ways for validating them. These future research directions will require us to collaborate across the scientific disciplines to incorporate models based on RCTs and known disease processes, physiology, and pharmacology into these machine learning models based on EHRs to fully optimize the opportunity these data present. [ABSTRACT FROM AUTHOR]

Background: Identifying people at risk of cardiovascular diseases (CVD) is a cornerstone of preventative cardiology. Risk prediction models currently recommended by clinical guidelines are typically based on a limited number of predictors with sub-optimal performance across all patient groups. Data-driven techniques based on machine learning (ML) might improve the performance of risk predictions by agnostically discovering novel risk predictors and learning the complex interactions between them. We tested (1) whether ML techniques based on a state-of-the-art automated ML framework (AutoPrognosis) could improve CVD risk prediction compared to traditional approaches, and (2) whether considering non-traditional variables could increase the accuracy of CVD risk predictions. Methods and findings: Using data on 423,604 participants without CVD at baseline in UK Biobank, we developed a ML-based model for predicting CVD risk based on 473 available variables. Our ML-based model was derived using AutoPrognosis, an algorithmic tool that automatically selects and tunes ensembles of ML modeling pipelines (comprising data imputation, feature processing, classification and calibration algorithms). We compared our model with a well-established risk prediction algorithm based on conventional CVD risk factors (Framingham score), a Cox proportional hazards (PH) model based on familiar risk factors (i.e, age, gender, smoking status, systolic blood pressure, history of diabetes, reception of treatments for hypertension and body mass index), and a Cox PH model based on all of the 473 available variables. Predictive performances were assessed using area under the receiver operating characteristic curve (AUC-ROC). Overall, our AutoPrognosis model improved risk prediction (AUC-ROC: 0.774, 95% CI: 0.768-0.780) compared to Framingham score (AUC-ROC: 0.724, 95% CI: 0.720-0.728, p < 0.001), Cox PH model with conventional risk factors (AUC-ROC: 0.734, 95% CI: 0.729-0.739, p < 0.001), and Cox PH model with all UK Biobank variables (AUC-ROC: 0.758, 95% CI: 0.753-0.763, p < 0.001). Out of 4,801 CVD cases recorded within 5 years of baseline, AutoPrognosis was able to correctly predict 368 more cases compared to the Framingham score. Our AutoPrognosis model included predictors that are not usually considered in existing risk prediction models, such as the individuals’ usual walking pace and their self-reported overall health rating. Furthermore, our model improved risk prediction in potentially relevant sub-populations, such as in individuals with history of diabetes. We also highlight the relative benefits accrued from including more information into a predictive model (information gain) as compared to the benefits of using more complex models (modeling gain). Conclusions: Our AutoPrognosis model improves the accuracy of CVD risk prediction in the UK Biobank population. This approach performs well in traditionally poorly served patient subgroups. Additionally, AutoPrognosis uncovered novel predictors for CVD disease that may now be tested in prospective studies. We found that the “information gain” achieved by considering more risk factors in the predictive model was significantly higher than the “modeling gain” achieved by adopting complex predictive models. [ABSTRACT FROM AUTHOR]

We investigate the problem of estimating the causal effect of a treatment on individual subjects from observational data; this is a central problem in various application domains, including healthcare, social sciences, and online advertising. Within the Neyman–Rubin potential outcomes model, we use the Kullback–Leibler (KL) divergence between the estimated and true distributions as a measure of accuracy of the estimate, and we define the information rate of the Bayesian causal inference procedure as the (asymptotic equivalence class of the) expected value of the KL divergence between the estimated and true distributions as a function of the number of samples. Using Fano's method, we establish a fundamental limit on the information rate that can be achieved by any Bayesian estimator, and show that this fundamental limit is independent of the selection bias in the observational data. We characterize the Bayesian priors on the potential (factual and counterfactual) outcomes that achieve the optimal information rate. We go on to propose a prior adaptation procedure (which we call the information-based empirical Bayes procedure) that optimizes the Bayesian prior by maximizing an information-theoretic criterion on the recovered causal effects rather than maximizing the marginal likelihood of the observed (factual) data. Building on our analysis, we construct an information-optimal Bayesian causal inference algorithm. This algorithm embeds the potential outcomes in a vector-valued reproducing Kernel Hilbert space, and uses a multitask Gaussian process prior over that space to infer the individualized causal effects. We show that for such a prior, the proposed information-based empirical Bayes method adapts the smoothness of the multitask Gaussian process to the true smoothness of the causal effect function by balancing a tradeoff between the factual bias and the counterfactual variance. We conduct experiments on a well-known real-world dataset and show that our model significantly outperforms the state-of-the-art causal inference models. [ABSTRACT FROM AUTHOR]

Modeling continuous-time physiological processes that manifest a patient's evolving clinical states is a key step in approaching many problems in healthcare. In this paper, we develop the Hidden Absorbing Semi-Markov Model (HASMM): a versatile probabilistic model that is capable of capturing the modern electronic health record (EHR) data. Unlike existing models, the HASMM accommodates irregularly sampled, temporally correlated, and informatively censored physiological data, and can describe non-stationary clinical state transitions. Learning the HASMM parameters from the EHR data is achieved via a novel forward-filtering backward-sampling Monte-Carlo EM algorithm that exploits the knowledge of the end-point clinical outcomes (informative censoring) in the EHR data, and implements the E-step by sequentially sampling the patients' clinical states in the reversetime direction while conditioning on the future states. Real-time inferences are drawn via a forward-filtering algorithm that operates on a virtually constructed discrete-time embedded Markov chain that mirrors the patient's continuous-time state trajectory. We demonstrate the prognostic utility of the HASMM in a critical care prognosis setting using a real-world dataset for patients admitted to the Ronald Reagan UCLA Medical Center. In particular, we show that using HASMMs, a patient's clinical deterioration can be predicted 8-9 hours prior to intensive care unit admission, with a 22% AUC gain compared to the Rothman index, which is the state-of-the-art critical care risk scoring technology. [ABSTRACT FROM AUTHOR]

Background: Risk prediction is crucial in many areas of medical practice, such as cardiac transplantation, but existing clinical risk-scoring methods have suboptimal performance. We develop a novel risk prediction algorithm and test its performance on the database of all patients who were registered for cardiac transplantation in the United States during 1985-2015. Methods and findings: We develop a new, interpretable, methodology (ToPs: Trees of Predictors) built on the principle that specific predictive (survival) models should be used for specific clusters within the patient population. ToPs discovers these specific clusters and the specific predictive model that performs best for each cluster. In comparison with existing clinical risk scoring methods and state-of-the-art machine learning methods, our method provides significant improvements in survival predictions, both post- and pre-cardiac transplantation. For instance: in terms of 3-month survival post-transplantation, our method achieves AUC of 0.660; the best clinical risk scoring method (RSS) achieves 0.587. In terms of 3-year survival/mortality predictions post-transplantation (in comparison to RSS), holding specificity at 80.0%, our algorithm correctly predicts survival for 2,442 (14.0%) more patients (of 17,441 who actually survived); holding sensitivity at 80.0%, our algorithm correctly predicts mortality for 694 (13.0%) more patients (of 5,339 who did not survive). ToPs achieves similar improvements for other time horizons and for predictions pre-transplantation. ToPs discovers the most relevant features (covariates), uses available features to best advantage, and can adapt to changes in clinical practice. Conclusions: We show that, in comparison with existing clinical risk-scoring methods and other machine learning methods, ToPs significantly improves survival predictions both post- and pre-cardiac transplantation. ToPs provides a more accurate, personalized approach to survival prediction that can benefit patients, clinicians, and policymakers in making clinical decisions and setting clinical policy. Because survival prediction is widely used in clinical decision-making across diseases and clinical specialties, the implications of our methods are far-reaching. [ABSTRACT FROM AUTHOR]

Objective: In this paper, we develop a personalized real-time risk scoring algorithm that provides timely and granular assessments for the clinical acuity of ward patients based on their (temporal) lab tests and vital signs; the proposed risk scoring system ensures timely intensive care unit admissions for clinically deteriorating patients. Methods: The risk scoring system is based on the idea of sequential hypothesis testing under an uncertain time horizon. The system learns a set of latent patient subtypes from the offline electronic health record data, and trains a mixture of Gaussian Process experts, where each expert models the physiological data streams associated with a specific patient subtype. Transfer learning techniques are used to learn the relationship between a patient's latent subtype and her static admission information (e.g., age, gender, transfer status, ICD-9 codes, etc). Results: Experiments conducted on data from a heterogeneous cohort of 6321 patients admitted to Ronald Reagan UCLA medical center show that our score significantly outperforms the currently deployed risk scores, such as the Rothman index, MEWS, APACHE, and SOFA scores, in terms of timeliness, true positive rate, and positive predictive value. Conclusion: Our results reflect the importance of adopting the concepts of personalized medicine in critical care settings; significant accuracy and timeliness gains can be achieved by accounting for the patients’ heterogeneity. Significance: The proposed risk scoring methodology can confer huge clinical and social benefits on a massive number of critically ill inpatients who exhibit adverse outcomes including, but not limited to, cardiac arrests, respiratory arrests, and septic shocks. [ABSTRACT FROM PUBLISHER]

In this paper, we characterize the achievable (linear) degrees of freedom (DoFs) of the K-user single-input single-output (SISO) interference channel with blind interference alignment (BIA) using staggered antenna switching. In this scheme, each transmitter is equipped with one conventional antenna, and each receiver is equipped with one reconfigurable (multimode) antenna. Assuming that the channel is known to the receivers only, we show that by using linear BIA with staggered antenna switching at the receiver, a sum DoF of \text{2}K/(K+\text{2}) is achievable. This result implies that in using BIA, we can double the DoF achieved by orthogonal multiple-access schemes for a large number of users. Moreover, we propose a systematic algorithm to generate the transmit beamforming vectors and the reconfigurable antenna switching patterns, showing that it can achieve the \text{2}K/(K+\text{2})$ sum DoF for any K$. Finally, we apply this algorithm to the four-user SISO interference channel and demonstrate that \text{4}/\text{3} sum DoF is indeed achievable. [ABSTRACT FROM PUBLISHER]

Breast cancer screening policies attempt to achieve timely diagnosis by regularly screening healthy women via various imaging tests. Various clinical decisions are needed to manage the screening process: selecting initial screening tests, interpreting test results, and deciding if further diagnostic tests are required. Current screening policies are guided by clinical practice guidelines (CPGs), which represent a “one-size-fits-all” approach, designed to work well (on average) for a population, and can only offer coarse expert-based patient stratification that is not rigorously validated through data. Since the risks and benefits of screening tests are functions of each patient's features, personalized screening policies tailored to the features of individuals are desirable. To address this issue, we developed ConfidentCare: a computer-aided clinical decision support system that learns a personalized screening policy from electronic health record (EHR) data. By a “personalized screening policy,” we mean a clustering of women's features, and a set of customized screening guidelines for each cluster. ConfidentCare operates by computing clusters of patients with similar features, then learning the “best” screening procedure for each cluster using a supervised learning algorithm. The algorithm ensures that the learned screening policy satisfies a predefined accuracy requirement with a high level of confidence for every cluster. By applying ConfidentCare to real-world data, we show that it outperforms the current CPGs in terms of cost efficiency and false positive rates: a reduction of 31 $\%$ in the false positive rate can be achieved. [ABSTRACT FROM PUBLISHER]

In this paper, we introduce the exposed secondary-user (SU) problem in underlay cognitive radio systems, where both the secondary-to-primary and primary-to-secondary channels have a line-of-sight (LoS) component. Based on a Rician model for the LoS channels, we show, both analytically and numerically, that LoS interference hinders the achievable SU capacity when interference constraints are imposed at the primary user (PU) receiver. This is caused by the poor dynamic range of interference channel fluctuations when a dominant LoS component exists. To improve the capacity of such a system, we propose the use of an electronically steerable parasitic array radiator (ESPAR) antenna at the secondary terminals. An ESPAR antenna involves a single radio frequency (RF) chain and has a reconfigurable radiation pattern that is controlled by assigning arbitrary weights to $M$ orthonormal basis radiation patterns via altering a set of reactive loads. By viewing the orthonormal patterns as multiple virtual dumb antennas, we randomly vary their weights over time, creating artificial channel fluctuations that can perfectly eliminate the undesired impact of LoS interference. This scheme is termed as random aerial beamforming (RAB) and is well suited for compact and low-cost mobile terminals as it uses a single RF chain. Moreover, we investigate the exposed-SU problem in a multiuser setting, showing that LoS interference hinders multiuser interference diversity and affects the growth rate of the SU capacity as a function of the number of users. Using RAB, we show that LoS interference can, in fact, be exploited to improve multiuser diversity by boosting the effective number of users. [ABSTRACT FROM AUTHOR]

GSM network operators are required to free up some of the spectrum allocated for voice traffic in order to serve 3G and 4G users. Re-farming 900 MHz and 1800 MHz bands into HSPA, W-CDMA (UMTS) and LTE is a challenge faced by operators with growing GSM voice traffic as 3G systems require 5 MHz bandwidth which is 25 times more bandwidth than a typical GSM channel. 4G systems (LTE) may even use 10 MHz to show superiority over the 5 MHz 3G systems. It is required to cope with the GSM voice traffic with significantly less bandwidth in order to accommodate 3/4G users. Orthogonal Sub Channel (OSC) is a novel technique offering efficient hardware utilization that can result in doubling GSM voice capacity. This is achieved by allowing two users to share the same physical radio resources. OSC can be used in phasing-out GSM networks to release significant frequency spectrum for 3/4G users. In this work, a thorough analysis on OSC performance gain is presented. The hard capacity gain offered by OSC is calculated as a function of the OSC penetration and compared with system level simulations. It is shown that hard capacity can be maintained with less number of TRXs (transceivers), and consequently, narrower spectrum. This paves the way for re-farming the released GSM spectrum into 3/4G traffic. [ABSTRACT FROM AUTHOR]

Introduction: Identifying people at risk of cardiovascular diseases (CVD) is a cornerstone of preventative cardiology. Risk prediction models currently recommended by clinical guidelines are based on a limited number of predictors with sub-optimal performance across all patient groups. Hypothesis: We tested whether a state-of-the-art method machine learning (ML) method (AutoPrognosis) can improve the performance of risk predictions and discover novel CVD risk predictors. Methods: AutoPrognosis is an algorithmic tool that automatically selects and tunes ensembles of ML models and learns the relative importance of different risk predictors. Using data on 473 variables collected for 423,604 participants without CVD at baseline in UK Biobank, we evaluated the accuracy of AutoPrognosis in predicting the participants' future CVD events. We compared the predictive accuracy of AutoPrognosis with a standard risk prediction score (Framingham score), a Cox proportional hazards (PH) model based on 7 conventional risk factors, and a Cox PH model based on all variables in the UK Biobank. Predictive performances were assessed using area under the receiver operating characteristic curve (AUC-ROC). Results: Our AutoPrognosis model improved risk prediction (AUC-ROC: 0.764, 95% CI: 0.758-0.770) compared to Framingham score (AUC-ROC: 0.707, 95% CI: 0.703-0.711, p < 0.001), the Cox model that uses the 7 conventional risk factors (AUC-ROC: 0.721, 95% CI: 0.717-0.725, p < 0.001), and the Cox model that uses all of the UK Biobank variables (AUC-ROC: 0.741, 95% CI: 0.736-0.746, p < 0.001). Our AutoPrognosis model included predictors that are not usually considered in existing risk prediction models, such as the individuals' usual walking pace and their self-reported overall health rating. Furthermore, our model substantially improved risk prediction in individuals with history of diabetes (AUC-ROC: 0.756, 95% CI: 0.745-0.767) compared to the Framingham score (AUC-ROC: 0.634, 95% CI: 0.625-0.673, p < 0.001). Conclusions: Our AutoPrognosis improves the accuracy of CVD risk prediction in the UK Biobank population. Additionally, AutoPrognosis uncovered novel predictors for CVD disease that may now be tested in prospective studies. [ABSTRACT FROM AUTHOR]

• G-G haplotype in IL-17A is negatively associated with HCC development in SOF, Peg-IFNα-2a, RBV treated CHC G4 patients. • Serum AFP is only associated with HCC development in Peg-IFNα-2a, RBV treated CHC G4 patients. • SOF might be linked with HCC development in some CHC G4 patients. • Haplotyping of IL-17A genes is crucial to personalize the proper treatment protocol for CHC G4 patient. Suspect has been directed towards some direct acting antivirals (DAAs) due to their reported association with hepatocellular carcinoma (HCC) development in chronic hepatitis C (CHC) patients. The mechanisms behind HCC development, following CHC treatment, were not well understood and may be linked to genetic variabilities in different patients which affect several cytokine productions involved in angiogenesis and inflammation. Of these variabilities, is the genetic polymorphisms in the interleukin-17 (IL-17) A receptor gene. Being an important pleiotropic cytokine, this study aimed to investigate the association between haplotypes in IL-17A receptor rs2275913 and rs3819024 and development of HCC in CHC patients treated with either triple therapy (sofosbuvir (SOF), pegylated interferon-alpha-2a (Peg-IFNα-2a) & ribavirin(RBV)) or with dual therapy (Peg-IFNα-2a&RBV). A cohort of 100 CHC patients was recruited in this study. Samples were tested for single nucleotide polymorphism (SNPs) in IL-17A receptor (rs2275913 and rs3819024) using TaqMan Genotyping assay. Our results showed that the presence of G-G haplotype in IL-17A (rs2275913& rs3819024) is inversely associated with HCC development in patients receiving triple therapy. While, high serum AFP levels are directly associated with HCC development in patients receiving triple therapy. However, in patients receiving dual therapy, HCC development was only associated with high serum alpha fetoprotein (AFP) levels and was not correlated to any specific allele in our studied SNPs. Such results highlight the importance of IL17A receptor gene haplotyping in the prediction of HCC development in patients receiving triple therapy. These results will aid in performing tailored, personalized strategy for CHC treatment. [ABSTRACT FROM AUTHOR]