This paper presents a general methodology for high-dimensional pattern regression on medical images via machine learning techniques. Compared with pattern classification studies, pattern regression considers the problem of estimating continuous rather than categorical variables, and can be more challenging. It is also clinically important, since it can be used to estimate disease stage and predict clinical progression from images. In this work, adaptive regional feature extraction approach is used along with other common feature extraction methods, and feature selection technique is adopted to produce a small number of discriminative features for optimal regression performance. Then the Relevance Vector Machine (RVM) is used to build regression models based on selected features. To get stable regression models from limited training samples, a bagging framework is adopted to build ensemble basis regressors derived from multiple bootstrap training samples, and thus to alleviate the effects of outliers as well as facilitate the optimal model parameter selection. Finally, this regression scheme is tested on simulated data and real data via cross-validation. Experimental results demonstrate that this regression scheme achieves higher estimation accuracy and better generalizing ability than Support Vector Regression (SVR).

Background Technological interventions such as mobile apps, Web-based social networks, and wearable trackers have the potential to influence physical activity; yet, only a few studies have examined the efficacy of an intervention bundle combining these different technologies. Objective This study aimed to pilot test an intervention composed of a social networking mobile app, connected with a wearable tracker, and investigate its efficacy in improving physical activity, as well as explore participant engagement and the usability of the app. Methods This was a pre-post quasi-experimental study with 1 arm, where participants were subjected to the intervention for a 6-month period. The primary outcome measure was the difference in daily step count between baseline and 6 months. Secondary outcome measures included engagement with the intervention and system usability. Descriptive and inferential statistical tests were conducted; posthoc subgroup analyses were carried out for participants with different levels of steps at baseline, app usage, and social features usage. Results A total of 55 participants were enrolled in the study; the mean age was 23.6 years and 28 (51%) were female. There was a nonstatistically significant increase in the average daily step count between baseline and 6 months (mean change=14.5 steps/day, P =.98, 95% CI –1136.5 to 1107.5). Subgroup analysis comparing the higher and lower physical activity groups at baseline showed that the latter had a statistically significantly higher increase in their daily step count (group difference in mean change from baseline to 6 months=3025 steps per day, P =.008, 95% CI 837.9-5211.8). At 6 months, the retention rate was 82% (45/55); app usage decreased over time. The mean system usability score was 60.1 (SD 19.2). Conclusions This study showed the preliminary efficacy of a mobile social networking intervention, integrated with a wearable tracker to promote physical activity, particularly for less physically active subgroups of the population. Future research should explore how to address challenges faced by physically inactive people to provide tailored advices. In addition, users’ perspectives should be explored to shed light on factors that might influence their engagement with the intervention.

BackgroundApproximately 10% of admissions to acute-care hospitals are associated with an adverse event. Analysis of incident reports helps to understand how and why incidents occur and can inform policy and practice for safer care. Unfortunately our capacity to monitor and respond to incident reports in a timely manner is limited by the sheer volumes of data collected. In this study, we aim to evaluate the feasibility of using multiclass classification to automate the identification of patient safety incidents in hospitals.MethodsText based classifiers were applied to identify 10 incident types and 4 severity levels. Using the one-versus-one (OvsO) and one-versus-all (OvsA) ensemble strategies, we evaluated regularized logistic regression, linear support vector machine (SVM) and SVM with a radial-basis function (RBF) kernel. Classifiers were trained and tested with “balanced” datasets (n_Type = 2860, n_SeverityLevel = 1160) from a state-wide incident reporting system. Testing was also undertaken with imbalanced “stratified” datasets (n_Type = 6000, n_SeverityLevel =5950) from the state-wide system and an independent hospital reporting system. Classifier performance was evaluated using a confusion matrix, as well as F-score, precision and recall.ResultsThe most effective combination was a OvsO ensemble of binary SVM RBF classifiers with binary count feature extraction. For incident type, classifiers performed well on balanced and stratified datasets (F-score: 78.3, 73.9%), but were worse on independent datasets (68.5%). Reports about falls, medications, pressure injury, aggression and blood products were identified with high recall and precision. “Documentation” was the hardest type to identify. For severity level, F-score for severity assessment code (SAC) 1 (extreme risk) was 87.3 and 64% for SAC4 (low risk) on balanced data. With stratified data, high recall was achieved for SAC1 (82.8–84%) but precision was poor (6.8–11.2%). High risk incidents (SAC2) were confused with medium risk incidents (SAC3).ConclusionsBinary classifier ensembles appear to be a feasible method for identifying incidents by type and severity level. Automated identification should enable safety problems to be detected and addressed in a more timely manner. Multi-label classifiers may be necessary for reports that relate to more than one incident type.Electronic supplementary materialThe online version of this article (doi:10.1186/s12911-017-0483-8) contains supplementary material, which is available to authorized users.

No abstract

No abstract