Refine
Year of publication
Document Type
- Article (218) (remove)
Language
- English (218) (remove)
Keywords
- (Strict) sign-regularity (1)
- 1D-CNN (1)
- 3D urban planning (1)
- AAL (2)
- Aboriginal people (1)
- Abstract interpretation (1)
- Accelerometer calibration (1)
- Accelerometer sensor (1)
- Accelerometers (1)
- Actuators (1)
Institute
- Fakultät Architektur und Gestaltung (1)
- Fakultät Bauingenieurwesen (11)
- Fakultät Elektrotechnik und Informationstechnik (1)
- Fakultät Informatik (7)
- Fakultät Maschinenbau (3)
- Fakultät Wirtschafts-, Kultur- und Rechtswissenschaften (27)
- Institut für Angewandte Forschung - IAF (30)
- Institut für Optische Systeme - IOS (9)
- Institut für Strategische Innovation und Technologiemanagement - IST (6)
- Institut für Systemdynamik - ISD (27)
Black-box variational inference (BBVI) is a technique to approximate the posterior of Bayesian models by optimization. Similar to MCMC, the user only needs to specify the model; then, the inference procedure is done automatically. In contrast to MCMC, BBVI scales to many observations, is faster for some applications, and can take advantage of highly optimized deep learning frameworks since it can be formulated as a minimization task. In the case of complex posteriors, however, other state-of-the-art BBVI approaches often yield unsatisfactory posterior approximations. This paper presents Bernstein flow variational inference (BF-VI), a robust and easy-to-use method flexible enough to approximate complex multivariate posteriors. BF-VI combines ideas from normalizing flows and Bernstein polynomial-based transformation models. In benchmark experiments, we compare BF-VI solutions with exact posteriors, MCMC solutions, and state-of-the-art BBVI methods, including normalizing flow-based BBVI. We show for low-dimensional models that BF-VI accurately approximates the true posterior; in higher-dimensional models, BF-VI compares favorably against other BBVI methods. Further, using BF-VI, we develop a Bayesian model for the semi-structured melanoma challenge data, combining a CNN model part for image data with an interpretable model part for tabular data, and demonstrate, for the first time, the use of BBVI in semi-structured models.
Apnea is a sleep disorder characterized by breathing interruptions during sleep, impacting cardiorespiratory function and overall health. Traditional diagnostic methods, like polysomnography (PSG), are unobtrusive, leading to noninvasive monitoring. This study aims to develop and validate a novel sleep monitoring system using noninvasive sensor technology to estimate cardiorespiratory parameters and detect sleep apnea. We designed a seamless monitoring system integrating noncontact force-sensitive resistor sensors to collect ballistocardiogram signals associated with cardiorespiratory activity. We enhanced the sensor’s sensitivity and reduced the noise by designing a new concept of edge-measuring sensor using a hemisphere dome and mechanical hanger to distribute the force and mechanically amplify the micromovement caused by cardiac and respiration activities. In total, we deployed three edge-measuring sensors, two deployed under the thoracic and one under the abdominal regions. The system is supported with onboard signal preprocessing in multiple physical layers deployed under the mattress. We collected the data in four sleeping positions from 16 subjects and analyzed them using ensemble empirical mode decomposition (EMD) to avoid frequency mixing. We also developed an adaptive thresholding method to identify sleep apnea. The error was reduced to 3.98 and 1.43 beats/min (BPM) in heart rate (HR) and respiration estimation, respectively. The apnea was detected with an accuracy of 87%. We optimized the system such that only one edge-measuring sensor can measure the cardiorespiratory parameters. Such a reduction in the complexity and simplification of the instruction of use shows excellent potential for in-home and continuous monitoring.
We quantify the effects of GATT/WTO membership on trade and welfare. Using an extensive database covering manufacturing trade for 186 countries over the period 1980–2016, we find that the average partial equilibrium impact of GATT/WTO membership on trade among member countries is large, positive, and significant. We contribute to the literature by estimating country-specific estimates and find them to vary widely across the countries in our sample with poorer members benefitting more. Using these estimates, we simulate the general equilibrium effects of GATT/WTO on welfare, which are sizable and heterogeneous across members. We show that countries not experiencing positive trade effects from joining GATT/WTO can still gain in terms of welfare, due to lower import prices and higher export demand.
Healthy and good sleep is a prerequisite for a rested mind and body. Both form the basis for physical and mental health. Healthy sleep is hindered by sleep disorders, the medically diagnosed frequency of which increases sharply from the age of 40. This chapter describes the formal specification of an on-course practical implementation for a non-invasive system based on biomedical signal processing to support the diagnosis and treatment of sleep-related diseases. The system aims to continuously monitor vital data during sleep in a patient’s home environment over long periods by using non-invasive technologies. At the center of the development is the MORPHEUS Box (MoBo), which consists of five main conceptualizations: the MoBo core, the MoBo-HW, the MoBo algorithm, the MoBo API, and the MoBo app. These synergistic elements aim to support the diagnosis and treatment of sleep-related diseases. Although there are related developments in individual aspects concerning the system, no comparative approach is known that gives a similar scope of functionality, deployment flexibility, extensibility, or the possibility to use multiple user groups. With the specification provided in this chapter, the MORPHEUS project sets a good platform, data model, and transmission strategies to bring an innovative proposal to measure sleep quality and detect sleep diseases from non-invasive sensors.
Infrastructure-making in interwar India was a dynamic, multilayered process involving roads and vehicles in urban and rural sites. One of their strongest playgrounds was Bombay Presidency and the Central Provinces in central and western India. Focusing on this region in the interwar period, this paper analyzes the varied relationship between peasant households and town-centred modernizing agents in the making of road transport infrastructures. The central argument of this paper is about the persistence of bullock carts over motor cars in the region. This persistence was grounded in the specific regional environment, the effects of the 1930s economic depression, and the priorities of social classes. Pinpointing these connections, the paper highlights that “modernization” of infrastructure was not a simple, linear process of progressivist change, nor did it mean the survival of apparently “old” technologies in the modern era. Instead, the paper pays attention to conflicting social complexities, implications, and meanings of the connection between infrastructure and modernity that modernization assumptions often overlook. Here, the paper shows how technological change occurred as a result of real, material class interests pulling infrastructural technology in different directions. This was where and why arguments of road-motor lobbyists and cart advocates eventually clashed, and Gandhian social workers resisted motor transport in defense of peasant interests.
This study aims to adapt CEFR in developing an integrative approach-based teaching material model for a pre-basic BISOL class. The method used in this research is the development research design by Borg and Gall. This study was development research. The stages are identification of the problem, formulation of a hypothetical draft model; feasibility testing by experts; product revision; and test product effectiveness. The data were collected through survey techniques, interviews, and documentation. The needs identification results revealed data encompassing 10 themes, 5 tasks per theme, and diverse evaluations comprising theory, in-class practice, and real-world field assignments, both on an individual and group basis. These identified needs require alignment with CEFR A1 for the development of BISOL learning. These findings were subsequently incorporated into the design of the teaching material model, and the results indicated that tailoring CEFR to BISOL as an integrative language teaching material model was feasible for application in the classroom, as assessed by experts. The implications suggest that integrating CEFR into BISOL is highly feasible for the development of teaching materials, and teachers can leverage this instructional model to enhance students' proficiency in the Indonesian language.
In this work, a storage study was conducted to find suitable packaging material for tomato powder storage. Experiments were laid out in a single factor completely randomized design (CRD) to study the effect of packaging materials on lycopene, vitamin C moisture content, and water activity of tomato powder; The factor (packaging materials) has three levels (low‐density polyethylene bag, polypropylene bottle, wrapped with aluminum foils, and packed in low‐density polyethylene bag) and is replicated three times. During the study, a twin layer solar tunnel dried tomato slices of var. Galilea was used. The dried tomato slices were then ground and packed (40 g each) in the packaging materials and stored at room temperature. Samples were drawn from the packages at 2‐month interval for quality analysis and SAS (version 9.2) software was used for statistical analysis. From the result, higher retention of lycopene (80.13%) and vitamin C (49.32%) and a nonsignificant increase in moisture content and water activity were observed for tomato powder packed in polypropylene bottles after 6 months of storage. For low‐density polyethylene packed samples and samples wrapped with aluminum foil and packed in a low‐density polyethylene bag, 57.06% and 60.45% lycopene retention and 42.9% and 49.23% Vitamin C retention were observed, respectively, after 6 months of storage. Considering the results found, it can be concluded that lycopene and vitamin C content of twin layer solar tunnel dried tomato powder can be preserved at ambient temperature storage by packing in a polypropylene bottle with a safe range of moisture content and water activity levels for 6 months.
The aim of this paper is to find out in how accommodation providers in the Seychelles perceive climate change and what mitigation and adaptation measures they can provide. In order to answer these questions, a qualitative mixed-method-approach, comprised of twenty semi-structured interviews, an online-survey and participant observation was used. Results show that accommodation providers especially perceive the effects of climate change that directly affect their business and that they have already partly implemented some mitigation and adaptation measures. However, strategies and regulations are needed at the Seychelles’ government level and on a global level to actually achieve CO2 neutral travel.
Unintrusive health monitoring systems is important when continuous monitoring of the patient vital signals is required. In this paper, signals obtained from accelerometers placed under a bed are processed with ballistocardiography algorithms and compared with synchronized electrocardiographic signals.
Cardiovascular diseases (CVD) are leading contributors to global mortality, necessitating advanced methods for vital sign monitoring. Heart Rate Variability (HRV) and Respiratory Rate, key indicators of cardiovascular health, are traditionally monitored via Electrocardiogram (ECG). However, ECG's obtrusiveness limits its practicality, prompting the exploration of Ballistocardiography (BCG) as a non-invasive alternative. BCG records the mechanical activity of the body with each heartbeat, offering a contactless method for HRV monitoring. Despite its benefits, BCG signals are susceptible to external interference and present a challenge in accurately detecting J-Peaks. This research uses advanced signal processing and deep learning techniques to overcome these limitations. Our approach integrates accelerometers for long-term BCG data collection during sleep, applying Discrete Wavelet Transforms (DWT) and Ensemble Empirical Mode Decomposition (EEMD) for feature extraction. The Bi-LSTM model, leveraging these features, enhances heartbeat detection, offering improved reliability over traditional methods. The study's findings indicate that the combined use of DWT, EEMD, and Bi-LSTM for J-Peak detection in BCG signals is effective, with potential applications in unobtrusive long-term cardiovascular monitoring. Our results suggest that this methodology could contribute to HRV monitoring, particularly in home settings, enhancing patient comfort and compliance.