Search

Analysis of high-resolution inertial sensor and global navigation satellite system (GNSS) data collected by mobile and wearable devices is a relatively new methodology in forestry and safety research that provides opportunities for modeling work activities in greater detail than traditional time study analysis. The objective of this study was to evaluate whether smartwatch-based activity recognition models could quantify the activities of rigging crew workers setting and disconnecting log chokers on cable logging operations. Four productive cycle elements (travel to log,set choker,travel away,clear) were timed for choker setters and four productive cycle elements (travel to log,unhook,travel away,clear) were timed for chasers working at five logging sites in North Idaho. Each worker wore a smartwatch that recorded accelerometer data at 25 Hz. Random forest machine learning was used to develop predictive models that classified the different cycle elements based on features extracted from the smartwatch acceleration data using 15 sliding window sizes (1 to 15 s) and five window overlap levels (0%, 25%, 50%, 75%, and 90%). Models were compared using multiclass area under the Receiver Operating Characteristic (ROC) curve, or AUC. The best choker setter model was created using a 3-s window with 90% overlap and had sensitivity values ranging from 76.95% to 83.59% and precision values ranging from 41.42% to 97.08%. The best chaser model was created using a 1-s window with 90% overlap and had sensitivity values ranging from 71.95% to 82.75% and precision values ranging from 14.74% to 99.16%. These results have demonstrated the feasibility of quantifying forestry work activities using smartwatch-based activity recognition models, a basic step needed to develop real-time safety notifications associated with high-risk job functions and to advance subsequent, comparative analysis of health and safety metrics across stand, site, and work conditions.

Parkinson's disease (PD) is a neurodegenerative disease inducing dystrophy of the motor system. Automatic movement analysis systems have potential in improving patient care by enabling personalized and more accurate adjust of treatment. These systems utilize machine learning to classify the movement properties based on the features derived from the signals. Smartphones can provide an inexpensive measurement platform with their built-in sensors for movement assessment. This study compared three feature selection and nine classification methods for identifying PD patients from control subjects based on accelerometer and gyroscope signals measured with a smartphone during a 20-step walking test. Minimum Redundancy Maximum Relevance (mRMR) and sequential feature selection with both forward (SFS) and backward (SBS) propagation directions were used in this study. The number of selected features was narrowed down from 201 to 4-15 features by applying SFS and mRMR methods. From the methods compared in this study, the highest accuracy for individual steps was achieved with SFS (7 features) and Naive Bayes classifier (accuracy 75.3%), and the second highest accuracy with SFS (4 features) and k Nearest neighbours (accuracy 75.1%). Leave-one-subject-out cross-validation was used in the analysis. For the overall classification of each subject, which was based on the majority vote of the classified steps, k Nearest Neighbors provided the most accurate result with an accuracy of 84.5% and an error rate of 15.5%. This study shows the differences in feature selection methods and classifiers and provides generalizations for optimizing methodologies for smartphone-based monitoring of PD patients. The results are promising for further developing the analysis system for longer measurements carried out in free-living conditions. publishedVersion

BackgroundIn an intensive care units, experts in mechanical ventilation are not continuously at patient's bedside to adjust ventilation settings and to analyze the impact of these adjustments on gas exchange. The development of clinical decision support systems analyzing patients' data in real time offers an opportunity to fill this gap.ObjectiveThe objective of this study was to determine whether a machine learning predictive model could be trained on a set of clinical data and used to predict transcutaneous hemoglobin oxygen saturation 5 min (5min SpO2) after a ventilator setting change.Data sourcesData of mechanically ventilated children admitted between May 2015 and April 2017 were included and extracted from a high-resolution research database. More than 776,727 data rows were obtained from 610 patients, discretized into 3 class labels (< 84%, 85% to 91% and c92% to 100%).Performance metrics of predictive modelsDue to data imbalance, four different data balancing processes were applied. Then, two machine learning models (artificial neural network and Bootstrap aggregation of complex decision trees) were trained and tested on these four different balanced datasets. The best model predicted SpO2 with area under the curves < 0.75.ConclusionThis single center pilot study using machine learning predictive model resulted in an algorithm with poor accuracy. The comparison of machine learning models showed that bagged complex trees was a promising approach. However, there is a need to improve these models before incorporating them into a clinical decision support systems. One potentially solution for improving predictive model, would be to increase the amount of data available to limit over-fitting that is potentially one of the cause for poor classification performances for 2 of the three class labels.

Patterns of spatial behavior dictate how we use our infrastructure, encounter other people, or are exposed to services and opportunities. Understanding these patterns through the analysis of data commonly available through commodity smartphones has become an important arena for innovation in both academia and industry. The resulting datasets can quickly become massive, indicating the need for concise understanding of the scope of the data collected. Some data is obviously correlated (for example GPS location and which WiFi routers are seen). Codifying the extent of these correlations could identify potential new models, provide guidance on the amount of data to collect, and even provide actionable features. However, identifying correlations, or even the extent of correlation, is difficult because the form of the correlation must be specified. Fractal-based intrinsic dimensionality directly calculates the minimum number of dimensions required to represent a dataset. We provide an intrinsic dimensionality analysis of four smartphone datasets over seven input dimensions, and empirically demonstrate an intrinsic dimension of approximately two.

Internet-connected devices, especially mobile devices such as smartphones, have become widely accessible in the past decade. Interaction with such devices has evolved into frequent and short-duration usage, and this phenomenon has resulted in a pervasive popularity of casual games in the game sector. On the other hand, development of casual games has become easier than ever as a result of the advancement of development tools. With the resulting fierce competition, now both acquisition and retention of users are the prime concerns in the field. In this study, we focus on churn prediction of mobile and online casual games. While churn prediction and analysis can provide important insights and action cues on retention, its application using play log data has been primitive or very limited in the casual game area. Most of the existing methods cannot be applied to casual games because casual game players tend to churn very quickly and they do not pay periodic subscription fees. Therefore, we focus on the new players and formally define churn using observation period (OP) and churn prediction period (CP). Using the definition, we develop a standard churn analysis process for casual games. We cover essential topics such as pre-processing of raw data, feature engineering including feature analysis, churn prediction modeling using traditional machine learning algorithms (logistic regression, gradient boosting, and random forests) and two deep learning algorithms (CNN and LSTM), and sensitivity analysis for OP and CP. Play log data of three different casual games are considered by analyzing a total of 193,443 unique player records and 10,874,958 play log records. While the analysis results provide useful insights, the overall results indicate that a small number of well-chosen features used as performance metrics might be sufficient for making important action decisions and that OP and CP should be properly chosen depending on the analysis goal.

Accurate and on-demand animal population counts are the holy grail for wildlife conservation organizations throughout the world because they enable fast and responsive adaptive management policies. While the collection of image data from camera traps, satellites, and manned or unmanned aircraft has advanced significantly, the detection and identification of animals within images remains a major bottleneck since counting is primarily conducted by dedicated enumerators or citizen scientists. Recent developments in the field of computer vision suggest a potential resolution to this issue through the use of rotation-invariant object descriptors combined with machine learning algorithms. Here we implement an algorithm to detect and count wildebeest from aerial images collected in the Serengeti National Park in 2009 as part of the biennial wildebeest count. We find that the per image error rates are greater than, but comparable to, two separate human counts. For the total count, the algorithm is more accurate than both manual counts, suggesting that human counters have a tendency to systematically over or under count images. While the accuracy of the algorithm is not yet at an acceptable level for fully automatic counts, our results show this method is a promising avenue for further research and we highlight specific areas where future research should focus in order to develop fast and accurate enumeration of aerial count data. If combined with a bespoke image collection protocol, this approach may yield a fully automated wildebeest count in the near future.

Bioimpedance spectroscopy (BIS) measurement errors may be caused by parasitic stray capacitance, impedance mismatch, cross-talking or their very likely combination. An accurate detection and identification is of extreme importance for further analysis because in some cases and for some applications, certain measurement artifacts can be corrected, minimized or even avoided. In this paper we present a robust method to detect the presence of measurement artifacts and identify what kind of measurement error is present in BIS measurements. The method is based on supervised machine learning and uses a novel set of generalist features for measurement characterization in different immittance planes. Experimental validation has been carried out using a database of complex spectra BIS measurements obtained from different BIS applications and containing six different types of errors, as well as error-free measurements. The method obtained a low classification error (0.33%) and has shown good generalization. Since both the features and the classification schema are relatively simple, the implementation of this pre-processing task in the current hardware of bioimpedance spectrometers is possible.

The double-yolked (DY) egg is quite popular in some Asian countries because it is considered as a sign of good luck, however, the double yolk is one of the reasons why these eggs fail to hatch. The usage of automatic methods for identifying DY eggs can increase the efficiency in the poultry industry by decreasing egg loss during incubation or improving sale proceeds. In this study, two methods for DY duck egg identification were developed by using computer vision technology. Transmittance images of DY and single-yolked (SY) duck eggs were acquired by a CCD camera to identify them according to their shape features. The Fisher's linear discriminant (FLD) model equipped with a set of normalized Fourier descriptors (NFDs) extracted from the acquired images and the convolutional neural network (CNN) model using primary preprocessed images were built to recognize duck egg yolk types. The classification accuracies of the FLD model for SY and DY eggs were 100% and 93.2% respectively, while the classification accuracies of the CNN model for SY and DY eggs were 98% and 98.8% respectively. The CNN-based algorithm took about 0.12 s to recognize one sample image, which was slightly faster than the FLD-based (about 0.20 s). Finally, this work compared two classification methods and provided the better method for DY egg identification.