Personalized Assistive Technologies for Motor-Impaired Students: A Case of Learning Process Mining.

Motor disability includes the lack of sensation, movement, or coordination, and Assistive Technologies (AT) can help overcome these challenges. Motor-disabled students need different ATs and configurations depending on courses and individual needs, and some solutions can be expensive. Some affordable AT has roots in gaming but can also be used for other purposes. However, there is little research on how they can be combined to define a personalized setting. Therefore, we performed a literature review to identify challenges and solutions to support students with motor disabilities in using information systems. The result defines a framework for identifying personalized settings. The usability of the result was demonstrated by performing a self-experimentation study of the first author, who has a motor disability. The results show its utility while learning process mining using the Graphical User Interface (GUI) and code-based tools. We identified challenges in using different User Interface (UI) elements, which can be used as a guideline for designers of process mining tools as well as other information systems to support diversity.

Adversarial Lagrangian integrated contrastive embedding for limited size datasets.

Certain datasets contain a limited number of samples with highly various styles and complex structures. This study presents a novel adversarial Lagrangian integrated contrastive embedding (ALICE) method for small-sized datasets. First, the accuracy improvement and training convergence of the proposed pre-trained adversarial transfer are shown on various subsets of datasets with few samples. Second, a novel adversarial integrated contrastive model using various augmentation techniques is investigated. The proposed structure considers the input samples with different appearances and generates a superior representation with adversarial transfer contrastive training. Finally, multi-objective augmented Lagrangian multipliers encourage the low-rank and sparsity of the presented adversarial contrastive embedding to adaptively estimate the coefficients of the regularizers automatically to the optimum weights. The sparsity constraint suppresses less representative elements in the feature space. The low-rank constraint eliminates trivial and redundant components and enables superior generalization. The performance of the proposed model is verified by conducting ablation studies by using benchmark datasets for scenarios with small data samples.

Online action proposal generation using spatio-temporal attention network.

Temporal action proposal generation aims to generate temporal boundaries containing action instances. In real-time applications such as surveillance cameras, autonomous driving, and traffic monitoring, the online localization and recognition of human activities occurring in short temporal intervals are important areas of research. Existing approaches of temporal action proposal generation consider only the offline and frame-level feature aggregation along the temporal dimension. Those offline methods also generate many redundant irrelevant proposal regions in the frames as temporal boundaries. This leads to higher computational cost along with slow processing speed which is not suitable for online tasks. In this study, we propose a novel spatio-temporal attention network for online action proposal generation as opposed to existing offline proposal generation methods. Our novel proposed approach incorporates the inter-dependency between the spatial and temporal context information of each incoming video clip to generate more relevant online temporal action proposals. First, we propose a windowed spatial attention module to capture the inter-spatial relationship between the features of incoming frames. The windowed spatial network produces more robust clip-level feature representation and efficiently deals with noisy features such as occlusion or background scenes. Second, we introduce a temporal attention module to capture relevant temporal dynamic information mutually to the localized spatial information to model the long inter-frame temporal relationship since most online real life videos are untrimmed in nature. By applying these two attention modules sequentially, the novel proposed spatio-temporal network model is able to generate precise action boundaries at a particular instant of time. In addition, the model generates fewer discriminative temporal action proposals while maintaining a low computational cost and high processing speed suitable for online settings.

dfgcompare: a library to support process variant analysis through Markov models.

BACKGROUND: Data-driven process analysis is an important area that relies on software support. Process variant analysis is a sort of analysis technique in which analysts compare executed process variants, a.k.a. process cohorts. This comparison can help to identify insights for improving processes. There are a few software supports to enable process cohort comparison based on the frequencies of process activities and performance metrics. These metrics are effective in cohort analysis, but they cannot support cohort comparison based on the probability of transitions among states, which is an important enabler for cohort analysis in healthcare. RESULTS: This paper defines an approach to compare process cohorts using Markov models. The approach is formalized, and it is implemented as an open-source python library, named dfgcompare. This library can be used by other researchers to compare process cohorts. The implementation is also used to compare caregivers' behavior when prescribing drugs in the Stockholm Region. The result shows that the approach enables the comparison of process cohorts in practice. CONCLUSIONS: We conclude that dfgcompare supports identifying differences among process cohorts.

Low-shot transfer with attention for highly imbalanced cursive character recognition.

Recognition of ancient Korean-Chinese cursive character (Hanja) is a challenging problem mainly because of large number of classes, damaged cursive characters, various hand-writing styles, and similar confusable characters. They also suffer from lack of training data and class imbalance issues. To address these problems, we propose a unified Regularized Low-shot Attention Transfer with Imbalance τ-Normalizing (RELATIN) framework. This handles the problem with instance-poor classes using a novel low-shot regularizer that encourages the norm of the weight vectors for classes with few samples to be aligned to those of many-shot classes. To overcome the class imbalance problem, we incorporate a decoupled classifier to rectify the decision boundaries via classifier weight-scaling into the proposed low-shot regularizer framework. To address the limited training data issue, the proposed framework performs Jensen-Shannon divergence based data augmentation and incorporate an attention module that aligns the most attentive features of the pretrained network to a target network. We verify the proposed RELATIN framework using highly-imbalanced ancient cursive handwritten character datasets. The results suggest that (i) the extreme class imbalance has a detrimental effect on classification performance; (ii) the proposed low-shot regularizer aligns the norm of the classifier in favor of classes with few samples; (iii) weight-scaling of decoupled classifier for addressing class imbalance appeared to be dominant in all the other baseline conditions; (iv) further addition of the attention module attempts to select more representative features maps from base pretrained model; (v) the proposed (RELATIN) framework results in superior representations to address extreme class imbalance issue.

Atrial Fibrillation Prediction With Residual Network Using Sensitivity and Orthogonality Constraints.

Atrial fibrillation (AF) is the most prevalent cardiac arrhythmia. The atrial beat is irregular during AF, which causes blood flow hardly. This may cause blood clot formation and cardioembolic strokes. Computer-aided devices may assist cardiologists in diagnosing heart rhythm disorders better. From this viewpoint, we attempt to identify the premature atrial complexes (PACs) to predict the occurrence of AF by using electrocardiogram (ECG) spectrograms. Convolutional neural networks (CNN) models such as ResNet and Wide-ResNet are used to predict the prelude of AF. Regularization constraints are used to deal with the imbalanced and small number of samples in the minority premature AF class. Sensitivity regularization investigates small variations in premature AF samples. It highlights more representative features that distinguish the PACs from the normal rhythm. On the other hand, orthogonality regularization removes the interference between negatively correlated feature weights. It places constraints on capturing similar patterns with slight differences. This constraint allows convergence to a better feature representation with fewer weight redundancies. We propose a combination of sensitivity and orthogonality penalty terms to the cost function of ResNet to decrease the overfitting and obtain a superior representation. The re-sampling class distribution method is also utilized to mitigate the issue of imbalanced data. The proposed method shows better AF prediction for highly imbalanced data with a small number of samples.

Evaluating a Clinical Decision Support System for Drug-Drug Interactions.

Janusmed is a clinical decision support system, developed by the Stockholm County Council that supports physicians in identifying drug-drug interactions. To determine how Janusmed is used in and affects the clinical practice, an evaluation study is currently being carried out that analyzes multiple data sources through descriptive statistics. The study focuses on how Janusmed affects the behavior of the physicians, in particular, to what extent physicians reconsider their prescription decisions based on warnings from Janusmed.

Thermal analysis of the dentine tubule under hot and cold stimuli using fluid-structure interaction simulation.

The objective of this study is to compare the thermal stress changes in the tooth microstructures and the hydrodynamic changes of the dental fluid under hot and cold stimuli. The dimension of the microstructures of eleven cats' teeth was measured by scanning electron microscopy, and the changes in thermal stress during cold and hot stimulation were calculated by 3D fluid-structure interaction modeling. Evaluation of results, following data validation, indicated that the maximum velocities in cold and hot stimuli were - 410.2 ± 17.6 and + 205.1 ± 8.7 µm/s, respectively. The corresponding data for maximum thermal stress were - 20.27 ± 0.79 and + 10.13 ± 0.24 cmHg, respectively. The thermal stress caused by cold stimulus could influence almost 2.9 times faster than that caused by hot stimulus, and the durability of the thermal stress caused by hot stimulus was 71% greater than that by cold stimulus under similar conditions. The maximum stress was on the tip of the odontoblast, while the stress in lateral walls of the odontoblast and terminal fibril was very weak. There is hence a higher possibility of pain transmission with activation of stress-sensitive ion channels at the tip of the odontoblast. The maximum thermal stress resulted from the cold stimulus is double that produced by the hot stimulus. There is a higher possibility of pain transmission in the lateral walls of the odontoblast and terminal fibril by releasing mediators during the cold stimulation than the hot stimulation. These two reasons can be associated with a greater pain sensation due to intake of cold liquids.