The wide-spread use of wearable devices and mobile apps in the Internet of Things (IoT) environments makes effectively capture of life-logging personal health data come true. A long-term collection of these health dat...
详细信息
ISBN:
(纸本)9781509001552
The wide-spread use of wearable devices and mobile apps in the Internet of Things (IoT) environments makes effectively capture of life-logging personal health data come true. A long-term collection of these health data will benefit to interdisciplinary healthcare research and collaboration. But most wearable devices and mobile apps in the market focus on personal fitness plan and lack of compatibility and extensibility to each other. Existing IoT based platforms rarely achieve a successful heterogeneous life-logging data aggregation. Also, the demand on high security increases difficulties of designing reliable platform for integrating and managing multi-resource life-logging health data. This paper investigates the possibility of collecting and aggregating life-logging data with the use of wearable devices, mobile apps and social media. It compares existing personal health data collection solutions and identifies essential needs of designing a life-logging data aggregator in the IoT environments. An integrated data collection solution with high secure standard is proposed and deployed on a state-of-the-art interdisciplinary healthcare platform: MHA [15] by integrating five life-logging resources: Fitbit, Moves, Facbook, Twitter, etc. The preliminary experiment demonstrates that it successfully record, store and reuse the unified and structured personal health information in a long term, including activities, location, exercise, sleep, food, heat rate and mood.
As a key health indictor, daily physical activity (PA) data has great significance on diagnosis and treatment of many chronic diseases. Numerous studies have been carried out for accurately monitoring and assessing ph...
详细信息
ISBN:
(纸本)9781509001552
As a key health indictor, daily physical activity (PA) data has great significance on diagnosis and treatment of many chronic diseases. Numerous studies have been carried out for accurately monitoring and assessing physical activity. Most attentions of these studies focus on designing standalone highly accurate wearable sensors or investigating advance machine learning algorithms to train these PA data in a controlled environment. But the wide use of cost-effective wearable devices and mobile apps makes it possible to monitor and access PA into a more open and connective Internet of Things (IoT) environment. Yet, it still lacks of a systemic survey on how to effectively transfer classic PA monitoring and assessment (PAMA) technologies into a heterogeneous device connected IoT environment. In an effect to understand the development of IoT technologies in PAMA, this paper reviews current research of PAMA technologies from an IoT layer-based perspective, and also identifies research challenges and future trends. A main contribution of this review paper is that it is first attempt to categorize classic PAMA technologies into an IoT architecture systematically.
Clustering algorithms have been popularly applied in tissue segmentation in MRI. However, traditional clustering algorithms could not take advantage of some prior knowledge of data even when it does exist. In this pap...
详细信息
Clustering algorithms have been popularly applied in tissue segmentation in MRI. However, traditional clustering algorithms could not take advantage of some prior knowledge of data even when it does exist. In this paper, we propose a new approach to tissue segmentation of 3D brain MRI using semi-supervised spectral clustering. Spectral clustering algorithm is more powerful than traditional clustering algorithms since it models the voxel-to-voxel relationship as opposed to voxel-to-cluster relationships. In the semi-supervised spectral clustering, two types of instance-level constraints: must-link and cannot-link as background prior knowledge are incorporated into spectral clustering, and the self-tuning parameter is applied to avoid the selection of the scaling parameter of spectral clustering. The semi-supervised spectral clustering is an effective tissue segmentation method because of its advantages in (1) better discovery of real data structure since there is no cluster shape restriction, (2) high quality segmentation results as it can obtain the global optimal solutions in the relaxed continuous domain by eigen-decomposition and combines the pairwise constraints information. Experimental results on simulated and real MRI data demonstrate its effectiveness.
In many biomedical applications, it is often desired to simulate, analyse and visualise the dynamics of a particular patient based on a patient-specific musculoskeletal model. However, reconstructing a patient-specifi...
详细信息
In many biomedical applications, it is often desired to simulate, analyse and visualise the dynamics of a particular patient based on a patient-specific musculoskeletal model. However, reconstructing a patient-specific model directly from medical images is highly labour intensive, and impractical in the clinical context. A more efficient method is to derive it from an atlas musculoskeletal model using patient-specific hints. In this paper, Laplacian mesh processing is introduced to deform an atlas model to a patient-specific model, based on patient-specific landmarks extracted from two orthogonal clinical images and using least-squares error optimization. Muscle attachment landmarks and motion landmarks in the atlas are also transformed as part of the process. Drift and inter-surface penetrations are prevented by supplementary inter-surface landmarks. Mesh simplification and reconstruction are used to avoid out-of-memory failures that may result from trying to deform models at high resolution.
暂无评论