Fig 3 - available via license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Content may be subject to copyright.
Sequence diagram of in MVVM with LiveData approach [8]
Source publication
![Fig. 3 Sequence diagram of in MVVM with LiveData approach [8]](publication/341137937/figure/fig1/AS:887673543733261@1588649273862/Sequence-diagram-of-in-MVVM-with-LiveData-approach-8_Q320.jpg)
The development of highly specialized mobile applications and systems has risen for several years, as we observe rapid deployment of innovative biomedical sensors and wearable technologies. The experiences gathered over 10 years of mobile medical software development, provide practical recommendations for architectural concepts utilized in analytic...
Citations
... In most cases, they store different sensor data that can provide a new set of patient state features. Due to the mentioned facts, their biggest challenges are effective data storage and analysis (Chmielewski et al., 2021(Chmielewski et al., , 2020. For example, the experiment conducted by the system presented in stores biomedical sensor data gathered with MS Band 2, and a daily carried out system protocol provides about 200 MB of raw data for one patient. ...
... Trying to process data with such volume synchronously could lead to server failure and service unavailability. Such a scenario is not acceptable for all types of Figure 10 Hangfire UI -server overview Figure 11 Hangfire UI -jobs list services but medical monitoring in particular (Chmielewski et al., 2020). To overcome such a problem, the asynchronous processing procedure has been applied. ...
Diabetes is a chronic autoimmune disease affecting the blood glucose concentration due to a lack of the hormone insulin. The recent reports released by The World Health Organization say that it affects about 100 million people worldwide. People with diabetes have to constantly control their blood glucose levels and provide insulin to avoid situations with higher glucose levels. Multiple factors influence that level and should be monitored to help them to prevent such cases. Nowadays, Continuous Glucose Monitoring-CGM-systems provide such functionality. They monitor the patient blood glucose level by 24 hours per day, enhancing that information with additional data gathered with wearable sensors. This paper presents the BGDCaLP-Blood Glucose Data Collector and Level Predictor-solution that provides mechanisms to collect blood glucose level data and predict future patient levels based on the provided data.
... There are some papers about building biomedical systems but most of them are about the developed solution itself (Chmielewski et al., 2018), some generic reviews and problems investigation (Chan et al., 2012;Pantelopoulos and Bourbakis, 2010;Seneviratne et al., 2017), and code quality guidelines (Xudong Lu et al., 2005). Unfortunately, some of them are out of date (Jones et al., n.d.;Pantelopoulos and Bourbakis, 2010;Xudong Lu et al., 2005), others provide general development guidelines for not technical people (Jones et al., n.d.) or provide information about just some special cases (Chmielewski et al., 2018(Chmielewski et al., , 2020a. There is no work that would be a guideline for developing biomedical systems capable of wearable sensor data fusion and health events reasoning. ...
This paper focuses on sharing and discussing efficient system design and architectural concepts developed and tested acquisition and processing of biomedical data in large-scale systems for medical monitoring and reasoning. A major area of the research included utilization of wearable and mobile technologies to enable the collection of high volumes of inertial and biomedical data used to support data fusion and decision-making algorithms as well as medical inferencing. Although medical diagnostics and decision algorithms have not been the main aim of the research, this preliminary phase was crucial to test the capabilities of existing off-the-shelf technologies and the functional responsibilities of the system's logic components. Architecture variants contained several schemes for data processing moving the responsibility for signal feature extraction, data classification, and pattern recognition from wearable to mobile up to server facilities. Analysis of transmission and processing delays provided architecture variants pros and cons but most of all knowledge about possible applications in medical, military, and fitness domains. To construct, evaluate and test architectures, a set of alternative technology stacks was prepared and compared using and quantitative metrics. The major architecture characteristics (high availability, scalability, reliability) have been defined imposing asynchronous processing of sensor data, efficient data representation, iterative reporting, event-driven processing, and restricting pulling operations. Sensor data processing persists the original data on handhelds but is mainly aimed at extracting chosen set of signal features calculated for specific time windows-varying for analyzed signals and the sensor data acquisition rates. Long-term monitoring of patients required the development of mechanisms, which probe the patient and in case of detecting anomalies or dangerous characteristic changes tune the data acquisition process. This paper describes experiences connected with the design of scalable decision support tools and evaluation techniques for architectural concepts implemented within the mobile and server software.
