STAR-Lite: A light-weight scalable self-taught learning framework for older adults’ activity recognition

Recognizing activities for older adults is challenging as we observe a variety of activity patterns caused due to aging (e.g., limited dexterity, limb control, slower response time) or/and underlying health conditions (e.g., dementia). However, existing literature with deep learning methods has successfully recognized activities when the dataset contains high-quality annotations and is captured in a controlled environment. On the contrary, data captured in a real-world environment, especially with older adults exhibiting memory-related symptoms, varying psychological and mental health status, reliance on caregivers to perform daily activities, and unavailability of domain-specific annotators, makes obtaining quality data with annotations challenging; leaving us with limited labeled data and abundant unlabeled data. In this paper, we hypothesize that projecting the labeled data representations comprising a specific set of activities onto a new representation space characterized by the unlabeled data comprising activities beyond the limited activities in the labeled dataset would help us rely less on the annotated data to improve activity detection performance. Motivated by this, we propose STAR-Lite, a self-taught learning framework that involves a pre-training framework to prepare the new representation space considering activities beyond the initial labels in the labeled dataset. STAR-Lite projects the labeled data representations on the new representation space characterized by unlabeled data labels and learns higher-level representations of the labeled dataset while optimizing inter- and intra- class distances without explicitly using a computation hungry similarity-based approach. We demonstrate that our proposed approach, STAR-Lite (a) improves activity recognition performance in a supervised setting and (b) is feasible for real-world deployment. To enhance the feasibility of deploying STAR-Lite on devices with limited memory resources, we explore model compression techniques such as pruning and quantization and propose a novel layer-wise pruning-rate optimization technique that effectively compresses the network while preserving the model performance. The evaluation was performed using the Alzheimer’s Activity Recognition dataset (AAR) captured from 25 individuals living in a retirement community center with IRB approval (#Y18NR12035) using an in-house SenseBox infrastructure while concurrently assessing the clinical evaluation of the participants for dementia, and independent living. Our extensive evaluation reveals that STAR-Lite can detect activities with an F1-score of 85.12% despite 62% reduction in model size and 5% improvement of execution time on a resource constrained device.     

Multimodal Attentive Fusion Network for audio-visual event recognition

Event classification is inherently sequential and multimodal. Therefore, deep neural models need to dynamically focus on the most relevant time window and/or modality of a video. In this study, we propose the Multimodal Attentive Fusion Network (MAFnet), an architecture that can dynamically fuse visual and audio information for event recognition. Inspired by prior studies in neuroscience, we couple both modalities at different levels of visual and audio paths. Furthermore, the network dynamically highlights a modality at a given time window relevant to classify events. Experimental results in AVE (Audio-Visual Event), UCF51, and Kinetics-Sounds datasets show that the approach can effectively improve the accuracy in audio-visual event classification. Code is available at: https://github.com/numediart/MAFnet     

Rapidly finding CAD features using database optimization

Automatic feature recognition aids downstream processes such as engineering analysis and manufacturing planning. Not all features can be defined in advance; a declarative approach allows engineers to specify new features without having to design algorithms to find them. Naive translation of declarations leads to executable algorithms with high time complexity. Database queries are also expressed declaratively; there is a large literature on optimizing query plans for efficient execution of database queries. Our earlier work investigated applying such technology to feature recognition, using a testbed interfacing a database system (SQLite) to a CAD modeler (CADfix). Feature declarations were translated into SQL queries which are then executed.The current paper extends this approach, using the PostgreSQL database, and provides several new insights: (i) query optimization works quite differently in these two databases, (ii) with care, an approach to query translation can be devised that works well for both databases, and (iii) when finding various simple common features, linear time performance can be achieved with respect to model size, with acceptable times for real industrial models. Further results also show how (i) lazy evaluation can be used to reduce the work performed by the CAD modeler, and (ii) estimating the time taken to compute various geometric operations can further improve the query plan. Experimental results are presented to validate our main conclusions.     

Pattern analysis in daily physical activity data for personal health management

Purpose: sedentary lifestyles have resulted in an increasing number of people who are at increased risk of various conditions and diseases, including overweight, obesity, and metabolic syndromes. Our objective was to systematically record the daily life journal on a platform to increase the self-awareness and improve the sedentary lifestyle and to assist clinicians in understanding and facilitating patients’ daily physical activity.Method: we developed a portable activity pattern recognition system designed to automatically recognize the daily activity habits of users, and provide visualized life logs on the wellness self-management platform for patients and clinicians. Based on the participants’ and the clinician’s comments, appropriate modifications were made.Results: persuading people to improve their activities during non-working hours can enhance the general physical activity. Since users’ smartphones automatically monitor their energy expenditure, healthcare professionals can use these data to assist their patients in addressing health problems stemming from the obesity or metabolic syndromes, thus empowering users to avert or delay the progression of diabetes, cardiovascular disease and other complications.Discussion and conclusions: the clinical pilot study showed the feasibility of applying this persuasive technology to improve the physical activity of overweight people. The limitation of the study is the need for Wi-Fi and 3G environments and a smartphone.     

Designing and preparing of quercetin surface‐imprinted material and its molecular recognition characteristics

Quercetin is an important compound of flavonoids. In this work, quercetin molecule surface‐imprinted material with high performance was prepared using a novel surface‐imprinting technique of “synchronously graft‐polymerizing and imprinting.” The modified micron‐sized silica gel particles containing amino groups were used as matrix, methacrylic acid (MAA) was used as functional monomer, and N,N′‐Methylenebisacrylamide (MBA) was used as crosslinker. In dimethyl formamide solution of quercetin, MAA molecules arranged automatically around the template quercetin molecule by right of hydrogen bonding interactions of two type, ordinary hydrogen bond and π‐type hydrogen bond. By initiating the surface‐initiating system of – , the graft/cross‐linking polymerization of MAA on SiO₂ particles and the quercetin molecule surface‐imprinting were simultaneously carried out, forming quercetin molecule surface‐imprinted material MIP‐PMAA/SiO₂. With another two flavonoids, rutin and genistein, as contrasting substances, the molecule recognition character of the quercetin molecule surface‐imprinted material MIP‐PMAA/SiO₂ was investigated with batch and column methods. The experimental results show that the imprinted material MIP‐PMAA/SiO₂ possesses special recognition selectivity and excellent binding affinity for quercetin molecule. The binding capacity of MIP‐PMAA/SiO₂ for quercetin is 0.325 mmol/g, and its selectivity coefficients for quercetin relative to rutin and genistein are 7.69 and 4.40, respectively. The main conditions of imprinting process affect the property of MIP‐PMAA/SiO₂ greatly, and the optimal molar ratio of monomer MAA to crosslinker MBA is 7 : 1 and appropriate molar ratio of monomer MAA to template quercetin is equal to 6 : 1. © 2014 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2014 , 131, 41112.     

天然气集输过程环境因素识别与评价

天然气是一种易燃、易爆的危险物质,集输过程工艺比较复杂,环境因素多,易发生火灾、爆炸等事故,造成环境污染、人员伤亡和财产损失。文章在借鉴国内外研究成果的基础上,以某天然气站线项目的天然气集输管道为例,结合天然气集输过程的工艺特点,采用输入输出分析法和矩阵法等现代理论和方法,从人、机、环境等三方面来研究天然气集输过程的环境因素。     

基于氢键的阴离子比色探针的合成和性质研究

以5-溴水杨醛和4-苯基-3-氨基硫脲合成了一种新型希夫碱类阴离子比色探针,研究了该探针对常见阴离子的比色响应特性。结果表明,在DMSO体系中加入氟离子、乙酸根离子和磷酸二氢根离子后,探针的最大吸波长从352 nm红移到398 nm,肉眼可观察到溶液的颜色从淡黄色变成了黄色,而其它的阴离子如氯离子、溴离子和碘离子的加入基本不引起溶液紫外-可见吸收波长的任何变化。因此,这种探针可以用来识别氟离子、乙酸根离子和磷酸二氢根离子。     

异斯特维醇衍生物的应用研究进展

异斯特维醇是由天然产物甜菊糖苷经酸解得到,其衍生物在生物活性方面的应用研究较多。本文对异斯特维醇衍生物在生物活性、催化剂、分子识别和超分子组装等领域的应用进行了综述。     

马路划线漆

阐述了马路划线漆的种类、特点与施工要点。重点介绍了水性马路划线漆的配方设计与施工要点,以及玻璃珠与马路划线漆的配套要求。     

In Vitro Selection of a Single-Stranded DNA Molecular Recognition Element against Atrazine

Widespread use of the chlorotriazine herbicide, atrazine, has led to serious environmental and human health consequences. Current methods of detecting atrazine contamination are neither rapid nor cost-effective. In this work, atrazine-specific single-stranded DNA (ssDNA) molecular recognition elements (MRE) were isolated. We utilized a stringent Systematic Evolution of Ligands by Exponential Enrichment (SELEX) methodology that placed the greatest emphasis on what the MRE should not bind to. After twelve rounds of SELEX, an atrazine-specific MRE with high affinity was obtained. The equilibrium dissociation constant (K_d) of the ssDNA sequence is 0.62 ± 0.21 nM. It also has significant selectivity for atrazine over atrazine metabolites and other pesticides found in environmentally similar locations and concentrations. Furthermore, we have detected environmentally relevant atrazine concentrations in river water using this MRE. The strong affinity and selectivity of the selected atrazine-specific ssDNA validated the stringent SELEX methodology and identified a MRE that will be useful for rapid atrazine detection in environmental samples.