Improved risk predictions via sparse imputation of patient conditions in electronic medical records

Electronic Medical Records (EMR) are increasingly used for risk prediction. EMR analysis is complicated by missing entries. There are two reasons - the “primary reason for admission” is included in EMR, but the co-morbidities (other chronic diseases) are left uncoded, and, many zero values in the data are accurate, reflecting that a patient has not accessed medical facilities. A key challenge is to deal with the peculiarities of this data - unlike many other datasets, EMR is sparse, reflecting the fact that patients have some, but not all diseases. We propose a novel model to fill-in these missing values, and use the new representation for prediction of key hospital events. To “fill-in” missing values, we represent the feature-patient matrix as a product of two low rank factors, preserving the sparsity property in the product. Intuitively, the product regularization allows sparse imputation of patient conditions reflecting common comorbidities across patients. We develop a scalable optimization algorithm based on Block coordinate descent method to find an optimal solution. We evaluate the proposed framework on two real world EMR cohorts: Cancer (7000 admissions) and Acute Myocardial Infarction (2652 admissions). Our result shows that the AUC for 3 months admission prediction is improved significantly from (0.741 to 0.786) for Cancer data and (0.678 to 0.724) for AMI data. We also extend the proposed method to a supervised model for predicting of multiple related risk outcomes (e.g. emergency presentations and admissions in hospital over 3, 6 and 12 months period) in an integrated framework. For this model, the AUC averaged over outcomes is improved significantly from (0.768 to 0.806) for Cancer data and (0.685 to 0.748) for AMI data.

Do the numbers and locations of road-killed anuran carcasses accurately reflect impacts of vehicular traffic?

Road-killed animals are easy and inexpensive to survey, and may provide information about species distributions, abundances, and mortality rates. As with any sampling method, however, we need to explore methodological biases in such data. First, how does an animal's behavior (e.g., use of the center vs. periphery of the road) influence its vulnerability to vehicular traffic? Second, how rapidly do post-mortem processes (scavenging by other animals, destruction or displacement by subsequent vehicles) change the numbers and locations of roadkills? Our surveys of anurans on a highway in tropical Australia show that different anuran species are distributed in different ways across the width of the road, and that locations of live versus dead animals sometimes differ within a species. Experimental trials show that location on the road affects the probability of being hit by a vehicle, with anurans in the middle of the road begin hit 35% more often than anurans on the edges; thus, center-using species are more likely to be hit than edge-using taxa. The magnitude of post-mortem displacement and destruction by subsequent vehicles depended on anuran species and body size. The mean parallel displacement distance was 122.7 cm, and carcasses of thin-skinned species exhibited greater post-mortem destruction. Scavenging raptors removed 73% of carcasses, most within a few hours of sunrise. Removal rates were biased with respect to size and species. Overall, our studies suggest that investigators should carefully evaluate potential biases before using roadkill counts to estimate underlying animal abundances or mortality rates.

Association between perception of fault for the crash and function, return to work and health status 1 year after road traffic injury: a registry-based cohort study

Objectives To establish the association between the patient's perception of fault for the crash and 12-month outcomes after non-fatal road traffic injury.Setting Two adult major trauma centres, one regional trauma centre and one metropolitan trauma centre in Victoria, Australia.Participants 2605 adult, orthopaedic trauma patients covered by the state's no-fault third party insurer for road traffic injury, injured between September 2010 and February 2014.Outcome measures EQ-5D-3L, return to work and functional recovery (Glasgow Outcome Scale—Extended score of upper good recovery) at 12 months postinjury.Results After adjusting for key confounders, the adjusted relative risk (ARR) of a functional recovery (0.57, 95% CI 0.46 to 0.69) and return to work (0.92, 95% CI 0.86 to 0.99) were lower for the not at fault compared to the at fault group. The ARR of reporting problems on EQ-5D items was 1.20–1.35 times higher in the not at fault group. Conclusions Patients who were not at fault, or denied being at fault despite a police report of fault, experienced poorer outcomes than the at fault group. Attributing fault to others was associated with poorer outcomes. Interventions to improve coping, or to resolve negative feelings from the crash, could facilitate better outcomes in the future.

Aggregation on the fly: Reducing traffic for big data in the cloud

As a leading framework for processing and analyzing big data, MapReduce is leveraged by many enterprises to parallelize their data processing on distributed computing systems. Unfortunately, the all-to-all data forwarding from map tasks to reduce tasks in the traditional MapReduce framework would generate a large amount of network traffic. The fact that the intermediate data generated by map tasks can be combined with significant traffic reduction in many applications motivates us to propose a data aggregation scheme for MapReduce jobs in cloud. Specifically, we design an aggregation architecture under the existing MapReduce framework with the objective of minimizing the data traffic during the shuffle phase, in which aggregators can reside anywhere in the cloud. Some experimental results also show that our proposal outperforms existing work by reducing the network traffic significantly.

Joint optimization of rule placement and traffic engineering for QoS provisioning in software defined network

Software-Defined Network (SDN) is a promising network paradigm that separates the control plane and data plane in the network. It has shown great advantages in simplifying network management such that new functions can be easily supported without physical access to the network switches. However, Ternary Content Addressable Memory (TCAM), as a critical hardware storing rules for high-speed packet processing in SDN-enabled devices, can be supplied to each device with very limited quantity because it is expensive and energy-consuming. To efficiently use TCAM resources, we propose a rule multiplexing scheme, in which the same set of rules deployed on each node apply to the whole flow of a session going through but towards different paths. Based on this scheme, we study the rule placement problem with the objective of minimizing rule space occupation for multiple unicast sessions under QoS constraints. We formulate the optimization problem jointly considering routing engineering and rule placement under both existing and our rule multiplexing schemes. Via an extensive review of the state-of-the-art work, to the best of our knowledge, we are the first to study the non-routing-rule placement problem. Finally, extensive simulations are conducted to show that our proposals significantly outperform existing solutions.

Robust network traffic classification

As a fundamental tool for network management and security, traffic classification has attracted increasing attention in recent years. A significant challenge to the robustness of classification performance comes from zero-day applications previously unknown in traffic classification systems. In this paper, we propose a new scheme of Robust statistical Traffic Classification (RTC) by combining supervised and unsupervised machine learning techniques to meet this challenge. The proposed RTC scheme has the capability of identifying the traffic of zero-day applications as well as accurately discriminating predefined application classes. In addition, we develop a new method for automating the RTC scheme parameters optimization process. The empirical study on real-world traffic data confirms the effectiveness of the proposed scheme. When zero-day applications are present, the classification performance of the new scheme is significantly better than four state-of-the-art methods: random forest, correlation-based classification, semi-supervised clustering, and one-class SVM.

Design of an optimal ANFIS traffic signal controller by using cuckoo search for an isolated intersection

An optimal design of Adaptive Neuro-Fuzzy Inference System (ANFIS) traffic signal controller is presented in this paper. The proposed controller aims to adjust a set of green times for traffic lights in a single intersection with the purpose of minimizing travel delay time and traffic congestion. The ANFIS controller is trained, to learned how to set green times for each traffic phase. This intelligent controller uses the Cuckoo Search (CS) algorithm to tune its parameters during the learning pried. Evaluating the performance of the proposed controller in comparison with the performance of a FLS controller (FLC) with predefined rules and membership functions, and also three fixed-Time controllers, illustrates the better performance of the optimal ANFIS controller against the other benchmark controllers.

Stabilizing sparse cox model using statistic and semantic structures in electronic medical records

Stability in clinical prediction models is crucial for transferability between studies, yet has received little attention. The problem is paramount in high dimensional data, which invites sparse models with feature selection capability. We introduce an effective method to stabilize sparse Cox model of time-to-events using statistical and semantic structures inherent in Electronic Medical Records (EMR). Model estimation is stabilized using three feature graphs built from (i) Jaccard similarity among features (ii) aggregation of Jaccard similarity graph and a recently introduced semantic EMR graph (iii) Jaccard similarity among features transferred from a related cohort. Our experiments are conducted on two real world hospital datasets: a heart failure cohort and a diabetes cohort. On two stability measures – the Consistency index and signal-to-noise ratio (SNR) – the use of our proposed methods significantly increased feature stability when compared with the baselines.

Computational intelligence-based traffic signal timing optimization

 Traffic congestion has explicit effects on productivity and efficiency, as well as side effects on environmental sustainability and health. Controlling traffic flows at intersections is recognized as a beneficial technique, to decrease daily travel times. This thesis applies computational intelligence to optimize traffic signals' timing and reduce urban traffic.