Research on a machine learning-based adaptive and efficient screening model for psychological symptoms of community correctional prisoners.

Community correction institutions in China frequently employ the Symptom Checklist-90 (SCL-90) and the health survey brief (SF-12) as primary tools for psychological assessment of community correctional prisoners. However, in practical application, the SCL-90 Checklist faces issues such as complex item numbers, overall low cultural level of the subjects, and insufficient professional level of the administrators. The SF-12 health survey brief, as a preliminary screening tool, although only has 12 questions, to some extent simplifies the evaluation process and improves work efficiency, it is prone to missed screening. The research team collected 17-dimensional basic characteristic data and corresponding SCL-90 and SF-12 data from 25,480 samples of community correctional prisoners in Zhejiang Province, China. This study explored the application of multi-label multi-classification algorithms and oversampling techniques in building machine learning models to delve into the correlation between the psychological health risks of community correctional prisoners and their characteristic data. Inspired by computerized adaptive testing (CAT), we constructed an adaptive and efficient screening model for community correctional prisoners through experimental comparisons, based on the binary relevance algorithm with sample oversampling. This screening model personalize the assessment process by dynamically matching participants with the most relevant subset (s) of the nine dimensions of the SCL-90 based on their individual characteristics. Thus, adaptive dynamic simplification and personalized recommendation of the SCL-90 scale between question groups were achieved for the specific group of community correctional prisoners. As a screening tool for psychological symptoms of community correctional prisoners, this model significantly simplifies the number of questions compared to SCL-90, with a simplification rate of up to 65%. However, it achieves this simplification while maintaining excellent performance. The accuracy reached 0.66, with a sensitivity of 0.754, and an F1 score of 0.649. This innovation simplified the assessment process, reduced the assessment time, improved work efficiency, and enhanced the ability to judge the specificity of community correctional prisoners population. Compared to the SF-12, although the simplification rate and accuracy of the model are slightly lower than those of the SF-12, the sensitivity increased by 42.26%, and the F1 score improved by 15.28%. This means the model greatly reduces the possibility of missed screening, effectively preventing prisoners with abnormal psychological or mental states from losing control due to missed screening, and even committing suicide, self injury, or injuring others.

Using a dual-stream attention neural network to characterize mild cognitive impairment based on retinal images.

Mild cognitive impairment (MCI) is a critical transitional stage between normal cognition and dementia, for which early detection is crucial for timely intervention. Retinal imaging has been shown as a promising potential biomarker for MCI. This study aimed to develop a dual-stream attention neural network to classify individuals with MCI based on multi-modal retinal images. Our approach incorporated a cross-modality fusion technique, a variable scale dense residual model, and a multi-classifier mechanism within the dual-stream network. The model utilized a residual module to extract image features and employed a multi-level feature aggregation method to capture complex context information. Self-attention and cross-attention modules were utilized at each convolutional layer to fuse features from optical coherence tomography (OCT) and fundus modalities, resulting in multiple output losses. The neural network was applied to classify individuals with MCI, Alzheimer's disease, and control participants with normal cognition. Through fine-tuning the pre-trained model, we classified community-dwelling participants into two groups based on cognitive impairment test scores. To identify retinal imaging biomarkers associated with accurate prediction, we used the Gradient-weighted Class Activation Mapping technique. The proposed method achieved high precision rates of 84.96% and 80.90% in classifying MCI and positive test scores for cognitive impairment, respectively. Notably, changes in the optic nerve head on fundus photographs or OCT images among patients with MCI were not used to discriminate patients from the control group. These findings demonstrate the potential of our approach in identifying individuals with MCI and emphasize the significance of retinal imaging for early detection of cognitive impairment.

Cross-cultural factors influencing the adoption of virtual reality for practical learning.

Education is one area that was significantly affected by the COVID-19 pandemic with much of the education being transferred online. Many subjects that require hands-on experimental experience suffer when taught online. Education is also one area that many believe can benefit from the advances in virtual reality (VR) technology, particularly for remote, online learning. Furthermore, because the technology shows overall good results with hands-on experiential learning education, one possible way to overcome online education barriers is with the use of VR applications. Given that VR has yet to make significant inroads in education, it is essential to understand what factors will influence this technology's adoption and acceptance. In this work, we explore factors influencing the adoption of VR for hands-on practical learning around the world based on the Unified Theory of Acceptance and Use of Technology and three additional constructs. We also performed a cross-cultural analysis to examine the model fit for developed and developing countries and regions. Moreover, through open-ended questions, we gauge the overall feeling people in these countries have regarding VR for practical learning and how it compares with regular online learning.

Video Moment Retrieval With Noisy Labels.

Video moment retrieval (VMR) aims to localize the target moment in an untrimmed video according to the given nature language query. The existing algorithms typically rely on clean annotations to train their models. However, making annotations by human labors may introduce much noise. Thus, the video moment retrieval models will not be well trained in practice. In this article, we present a simple yet effective video moment retrieval framework via bottom-up schema, which is in end-to-end manners and robust to noisy label training. Specifically, we extract the multimodal features by syntactic graph convolutional networks and multihead attention layers, which are fused by the cross gates and the bilinear approach. Then, the feature pyramid networks are constructed to encode plentiful scene relationships and capture high semantics. Furthermore, to mitigate the effects of noisy annotations, we devise the multilevel losses characterized by two levels: a frame-level loss that improves noise tolerance and an instance-level loss that reduces adverse effects of negative instances. For the frame level, we adopt the Gaussian smoothing to regard noisy labels as soft labels through the partial fitting. For the instance level, we exploit a pair of structurally identical models to let them teach each other during iterations. This leads to our proposed robust video moment retrieval model, which experimentally and significantly outperforms the state-of-the-art approaches on standard public datasets ActivityCaption and textually annotated cooking scene (TACoS). We also evaluate the proposed approach on the different manual annotation noises to further demonstrate the effectiveness of our model.

Wrist pulse signal based vascular age calculation using mixed Gaussian model and support vector regression.

Purpose: Vascular age (VA) is the direct index to reflect vascular aging, so it plays a particular role in public health. How to obtain VA conveniently and cheaply has always been a research hotspot. This study proposes a new method to evaluate VA with wrist pulse signal. Methods: Firstly, we fit the pulse signal by mixed Gaussian model (MGM) to extract the shape features, and adopt principal component analysis (PCA) to optimize the dimension of the shape features. Secondly, the principal components and chronological age (CA) are respectively taken as the independent variables and dependent variable to establish support vector regression (SVR) model. Thirdly, the principal components are fed into the SVR model to predicted the vascular aging of each subject. The predicted value is regarded as the description of VA. Finally, we compare the correlation coefficients of VA with pulse width (PW), inflection point area ratio (IPA), Ratio b/a (RBA), augmentation index (AIx), diastolic augmentation index (DAI) and pulse transit time (PTT) with those of CA with these six indices. Results: Compared with the CA, the VA is closer to PW (r = 0.539, P < 0.001 to r = 0.589, P < 0.001 in men; r = 0.325, P < 0.001 to r = 0.400, P < 0.001 in women), IPA (r = - 0.446, P < 0.001 to r = - 0.534, P < 0.001 in men; r = - 0.623, P < 0.001 to r = - 0.660, P < 0.001 in women), RBA (r = 0.328, P < 0.001 to r = 0.371, P < 0.001 in women), AIx (r = 0.659, P < 0.001 to r = 0.738, P < 0.001 in men; r = 0.547, P < 0.001 to r = 0.573, P < 0.001 in women), DAI (r = 0.517, P < 0.001 to r = 0.532, P < 0.001 in men; r = 0.507, P < 0.001 to r = 0.570, P < 0.001 in women) and PTT (r = 0.526, P < 0.001 to r = 0.659, P < 0.001 in men; r = 0.577, P < 0.001 to r = 0.814, P < 0.001 in women). Conclusion: The VA is more representative of vascular aging than CA. The method presented in this study provides a new way to directly and objectively assess vascular aging in public health.

KeratoScreen: Early Keratoconus Classification With Zernike Polynomial Using Deep Learning.

PURPOSE: We aimed to investigate the usefulness of Zernike coefficients (ZCs) for distinguishing subclinical keratoconus (KC) from normal corneas and to evaluate the goodness of detection of the entire corneal topography and tomography characteristics with ZCs as a screening feature input set of artificial neural networks. METHODS: This retrospective study was conducted at the Affiliated Eye Hospital of Wenzhou Medical University, China. A total of 208 patients (1040 corneal topography images) were evaluated. Data were collected between 2012 and 2018 using the Pentacam system and analyzed from February 2019 to December 2021. An artificial neural network (KeratoScreen) was trained using a data set of ZCs generated from corneal topography and tomography. Each image was previously assigned to 3 groups: normal (70 eyes; average age, 28.7 ± 2.6 years), subclinical KC (48 eyes; average age, 24.6 ± 5.7 years), and KC (90 eyes; average age, 25.9 ± 5.4 years). The data set was randomly split into 70% for training and 30% for testing. We evaluated the precision of screening symptoms and examined the discriminative capability of several combinations of the input set and nodes. RESULTS: The best results were achieved using ZCs generated from corneal thickness as an input parameter, determining the 3 categories of clinical classification for each subject. The sensitivity and precision rates were 93.9% and 96.1% in subclinical KC cases and 97.6% and 95.1% in KC cases, respectively. CONCLUSIONS: Deep learning algorithms based on ZCs could be used to screen for early KC and for other corneal ectasia during preoperative screening for corneal refractive surgery.

Design and Preliminary Realization of a Screening and Early Warning Health Management System for Populations at High Risk for Depression.

Depression has a high incidence in the world. Based on the concept of preventive treatment of disease of traditional Chinese medicine, timely screening and early warning of depression in populations at high risk for this condition can avoid, to a certain extent, the dysfunctions caused by depression. This work studied a method to collect information on depression, generate a database of depression features, design algorithms for screening populations at high risk for depression and creating an early warning model, develop an early warning short-message service (SMS) platform, and implement a scheme of depression screening and an early warning health management system. The implementation scheme included mobile application (app), cloud form, screening and early warning model, cloud platform, and computer software. Multiple modules jointly realized the screening, early warning, and management of the health functions of individuals at high risk for depression. At the same time, function modules such as mobile app and cloud form for collecting depression health information, early warning SMS platform, and health management software were designed, and the functions of the modules were preliminarily developed. Finally, the black-box test and white-box test were used to assess the system's functions and ensure the reliability of the system. Through the integration of mobile app and computer software, this study preliminarily realized the screening and early warning health management of a population at high risk for depression.

MagicChem: a MR system based on needs theory for chemical experiments.

Real chemical experiments may be dangerous or pollute the environment; meanwhile, the preparation of drugs and reagents is time-consuming. Due to the above-mentioned reasons, few experiments can be actually operated by students, which is not conducive to the chemistry learning and the phenomena principle understanding. Recently, due to the impact of Covid-19, many schools adopt online teaching, which is even more detrimental to students' learning of chemistry. Fortunately, MR(mixed reality) technology provides us with the possibility of solving the safety issues and breaking the space-time constraints, while the theory of human needs (Maslow's hierarchical needs) provides us with a way to design a comfortable and stimulant MR system with realistic visual presentation and interaction. The paper combines with the theory of human needs to propose a new needs model for virtual experiment. Based on this needs model, we design and develop a comprehensive MR system called MagicChem, which offers a robust 6-DoF interactive and illumination consistent experimental space with virtual-real occlusion, supporting realistic visual interaction, tangible interaction, gesture interaction with touching, voice interaction, temperature interaction, olfactory interaction and virtual human interaction. User study shows that MagicChem satisfies the needs model better than other MR experimental environments that partially meet the needs model. In addition, we explore the application of the needs model in VR environment. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s10055-021-00560-z.

A Convenient and Low-Cost Model of Depression Screening and Early Warning Based on Voice Data Using for Public Mental Health.

Depression is a common mental health disease, which has great harm to public health. At present, the diagnosis of depression mainly depends on the interviews between doctors and patients, which is subjective, slow and expensive. Voice data are a kind of data that are easy to obtain and have the advantage of low cost. It has been proved that it can be used in the diagnosis of depression. The voice data used for modeling in this study adopted the authoritative public data set, which had passed the ethical review. The features of voice data were extracted by Python programming, and the voice features were stored in the format of CSV files. Through data processing, a big database, containing 1479 voice feature samples, was generated for modeling. Then, the decision tree screening model of depression was established by 10-fold cross validation and algorithm selection. The experiment achieved 83.4% prediction accuracy on voice data set. According to the prediction results of the model, the patients can be given early warning and intervention in time, so as to realize the health management of personal depression.

A rapid density method for taxi passengers hot spot recognition and visualization based on DBSCAN.

With the development of city size and vehicle interconnection, visual analysis technology is playing a very important role in the course of city calculation and city perception. A Reasonable visual model can effectively present the feature of city. In order to solve the problem of traditional density algorithm that cluster the large scale data slowly and cannot find cluster centers to adapt taxi track data. The DBSCAN+ (density-based spatial clustering of applications with noise plus) algorithm that can split data and extract maximum density clusters under the large scale data was proposed in the paper. The passenger points should be cleaned from the original point of the passenger trajectory data firstly, and then the massive passenger points are sliced and clustered cyclically. In the clustering process, the cluster centers can be extracted based on maximum density, and finally the clustering results are visualized according to the results. The experimental results show that compared with other popular methods, the proposed method has significant advantages in clustering speed, precision and visualization for large-scale city passenger hotspots. Moreover, it provides important decisions for further urban planning and promotes the traffic efficiency.

Semi-Supervised Perception Augmentation for Aerial Photo Topologies Understanding.

Intelligently understanding the sophisticated topological structures from aerial photographs is a useful technique in aerial image analysis. Conventional methods cannot fulfill this task due to the following challenges: 1) the topology number of an aerial photo increases exponentially with the topology size, which requires a fine-grained visual descriptor to discriminatively represent each topology; 2) identifying visually/semantically salient topologies within each aerial photo in a weakly-labeled context, owing to the unaffordable human resources required for pixel-level annotation; and 3) designing a cross-domain knowledge transferal module to augment aerial photo perception, since multi-resolution aerial photos are taken asynchronistically in practice. To handle the above problems, we propose a unified framework to understand aerial photo topologies, focusing on representing each aerial photo by a set of visually/semantically salient topologies based on human visual perception and further employing them for visual categorization. Specifically, we first extract multiple atomic regions from each aerial photo, and thereby graphlets are built to capture the each aerial photo topologically. Then, a weakly-supervised ranking algorithm selects a few semantically salient graphlets by seamlessly encoding multiple image-level attributes. Toward a visualizable and perception-aware framework, we construct gaze shifting path (GSP) by linking the top-ranking graphlets. Finally, we derive the deep GSP representation, and formulate a semi-supervised and cross-domain SVM to partition each aerial photo into multiple categories. The SVM utilizes the global composition from low-resolution counterparts to enhance the deep GSP features from high-resolution aerial photos which are partially-annotated. Extensive visualization results and categorization performance comparisons have demonstrated the competitiveness of our approach.

Design and Evaluation of an Augmented Reality-Based Exergame System to Reduce Fall Risk in the Elderly.

Falls are a major public health concern in today's aging society. Virtual reality (VR) technology is a promising method for reducing fall risk. However, the absence of representations of the user's body in a VR environment lessens the spatial sense of presence. In terms of user experience, augmented reality (AR) can provide a higher degree of presence and embodiment than VR. We developed an AR-based exergame system that is specifically designed for the elderly to reduce fall risk. Kinect2.0 was used to capture and generate 3D models of the elderly and immerse them in an interactive virtual environment. The software included three functional modules: fall risk assessment, cognitive-motor intervention (CMI) training, and training feedback. The User Experience Questionnaire (UEQ-S) was used to evaluate user experience. Twenty-five elders were enrolled in the study. It was shown that the average scores for each aspect were: pragmatic quality score (1.652 ± 0.868); hedonic quality score (1.880 ± 0.962); and overall score was 1.776 ± 0.819. The overall score was higher than 0.8, which means that the system exhibited a positive user experience. After comparing the average score in a dataset product of UEQ-S Data Analysis Tool, it was found that the pragmatic quality aspect was categorized as good, while the hedonic quality aspect was categorized as excellent. It revealed a positive evaluation from users.

A novel ensemble of random forest for assisting diagnosis of Parkinson's disease on small handwritten dynamics dataset.

BACKGROUND: Parkinson's disease (PD) is a neurodegenerative disease of the elderly, which leads to patients' motor and non-motor disabilities and affects patients' quality of daily life. Timely and effective detection of PD is a key step to medical intervention. Recently, computer aided methods for PD detection have gained lots of attention in artificial intelligence domain. METHODS: This paper proposed a novel ensemble learning model fusing Random Forest (RF) classifiers and Principal Component Analysis (PCA) technique to differentiate PD patients from healthy controls (HC). Six different RF models were separately constructed to generate the corresponding class probability vectors which represent an individual's category predictions on 6 different handwritten exams, and the final prediction result for an individual was obtained through voting strategy of all RF models. Stratified k-fold cross validation was performed to split the exam datasets and evaluate the classification performances. RESULTS: Experimental results prove that our proposed ensemble model on six handwritten exams has achieved better classification performances than a single RF based method on a single handwritten exam. Our ensemble of RF model based on multiple handwritten exams has promising accuracy (89.4 %), specificity (93.7 %), sensitivity (84.5 %) and F1-score (87.7 %). Compared with Logistic Regression (LR) and Support Vector Machines (SVM), the ensemble model based on RF can achieve better classification results. CONCLUSION: A computer-assisted PD diagnosis model on small handwritten dynamics dataset is proposed, and it provides a potential way for assisting diagnosis of PD in clinical setting.

Dream-Experiment: A MR User Interface with Natural Multi-channel Interaction for Virtual Experiments.

This paper studies a set of MR technologies for middle school experimental teaching environments and develops a multi-channel MR user interface called Dream-Experiment. The goal of Dream-Experiment is to improve the traditional MR user interface, so that users can get a real, natural 3D interactive experience like real experiments, but without danger and pollution. In terms of visual presentation, we design multi-camera collaborative registration to realize robust 6-DoF MR interactive space, and also define a complete rendering pipeline to provide improved processing of virtual-real objects' occlusion including translucent devices. In the virtual-real interaction, we provide six interaction modes that support visual interaction, tangible interaction, virtual-real gestures with touching, voice, thermal feeling, and olfactory feeling. After users' testing, we find that Dream-Experiment has better interactive efficiency and user experience than traditional MR environments.

ScatterNet: A Deep Subjective Similarity Model for Visual Analysis of Scatterplots.

Similarity measuring methods are widely adopted in a broad range of visualization applications. In this work, we address the challenge of representing human perception in the visual analysis of scatterplots by introducing a novel deep-learning-based approach, ScatterNet, captures perception-driven similarities of such plots. The approach exploits deep neural networks to extract semantic features of scatterplot images for similarity calculation. We create a large labeled dataset consisting of similar and dissimilar images of scatterplots to train the deep neural network. We conduct a set of evaluations including performance experiments and a user study to demonstrate the effectiveness and efficiency of our approach. The evaluations confirm that the learned features capture the human perception of scatterplot similarity effectively. We describe two scenarios to show how ScatterNet can be applied in visual analysis applications.

Detailed Surface Geometry and Albedo Recovery from RGB-D Video under Natural Illumination.

This article presents a novel approach for depth map enhancement from an RGB-D video sequence. The basic idea is to exploit the photometric information in the color sequence to resolve the inherent ambiguity of shape from shading problem. Instead of making any assumption about surface albedo or controlled object motion and lighting, we use the lighting variations introduced by casual object movement. We are effectively calculating photometric stereo from a moving object under natural illuminations. One of the key technical challenges is to establish correspondences over the entire image set. We, therefore, develop a lighting insensitive robust pixel matching technique that out-performs optical flow method in presence of lighting variations. An adaptive reference frame selection procedure is introduced to get more robust to imperfect lambertian reflections. In addition, we present an expectation-maximization framework to recover the surface normal and albedo simultaneously, without any regularization term. We have validated our method on both synthetic and real datasets to show its superior performance on both surface details recovery and intrinsic decomposition.

A Cascade Ensemble Learning Model for Human Activity Recognition with Smartphones.

Human activity recognition (HAR) has gained lots of attention in recent years due to its high demand in different domains. In this paper, a novel HAR system based on a cascade ensemble learning (CELearning) model is proposed. Each layer of the proposed model is comprised of Extremely Gradient Boosting Trees (XGBoost), Random Forest, Extremely Randomized Trees (ExtraTrees) and Softmax Regression, and the model goes deeper layer by layer. The initial input vectors sampled from smartphone accelerometer and gyroscope sensor are trained separately by four different classifiers in the first layer, and the probability vectors representing different classes to which each sample belongs are obtained. Both the initial input data and the probability vectors are concatenated together and considered as input to the next layer's classifiers, and eventually the final prediction is obtained according to the classifiers of the last layer. This system achieved satisfying classification accuracy on two public datasets of HAR based on smartphone accelerometer and gyroscope sensor. The experimental results show that the proposed approach has gained better classification accuracy for HAR compared to existing state-of-the-art methods, and the training process of the model is simple and efficient.

Toward Visual Avatars that Dress You Well and Impact Your Health.

This article discusses what would be needed to realize visual avatars that could dress us well and potentially impact our health by suggesting clothes appropriate for our activities.

Visual Object Tracking Based on Cross-Modality Gaussian-Bernoulli Deep Boltzmann Machines with RGB-D Sensors.

Visual object tracking technology is one of the key issues in computer vision. In this paper, we propose a visual object tracking algorithm based on cross-modality featuredeep learning using Gaussian-Bernoulli deep Boltzmann machines (DBM) with RGB-D sensors. First, a cross-modality featurelearning network based on aGaussian-Bernoulli DBM is constructed, which can extract cross-modality features of the samples in RGB-D video data. Second, the cross-modality features of the samples are input into the logistic regression classifier, andthe observation likelihood model is established according to the confidence score of the classifier. Finally, the object tracking results over RGB-D data are obtained using aBayesian maximum a posteriori (MAP) probability estimation algorithm. The experimental results show that the proposed method has strong robustness to abnormal changes (e.g., occlusion, rotation, illumination change, etc.). The algorithm can steadily track multiple targets and has higher accuracy.

Customized Body Mapping to Facilitate the Ergonomic Design of Sportswear.

A successful high-performance sportswear design that considers human factors should result in a significant increase in thermal comfort and reduce energy loss. The authors describe a body-mapping approach that facilitates the effective ergonomic design of sportswear. Their general framework can be customized based on the functional requirements of various sports and sportswear, the desired combination and selection of mapping areas for the human body, and customized quantitative data distribution of target physiological indicators.