2.2 Early Warning Score Systems Requirements
Early warning score systems are used to improve the process of recording, scoring and
responding to changes in measured vital signs of patients. The following paragraphs
present the principal requirements of EWS systems to ensure an efficient detection and
response to clinical deterioration:
• Personalization
The patient vital sings data are the key of successful medical decisions. Each
patient requires a personalized control depending on his situation [7]. For example a
patient with a disease that causes hyperthermia, the temperature is a critical data which
must be visualized with more precision; every second. While other non-critical
parameters can be calculated only every hour for example.
• Medical staff engagement
The early warning scores systems are highly user-dependent and depend on the
appropriate response and actions of the Medical Emergency Teams [4].
• Need for expert’s decision
The warning scores cannot replace specialist’s decision and the importance on
knowing individual patients, they cannot also replace the background to the observations that are recorded [2].
NEWS correlated poorly with the patient’s clinical status within the first 24 h postoperatively [2].
2.3 Related Works
In [8], authors present a solution which takes benefits from the concept of Edge
computing and fog computing in the context of IoT based Early Warning Scores
systems. The solution provides high level services in a Geo-distributed fashion at the
edge of the network. The proof of concept is demonstrated with smart e-Health Gateway called UT-GATE implemented for an IoT-based remote health monitoring system.
The demonstration includes the data flow processes from data acquisition at sensor
nodes to the cloud and the end-users. However, the data has a static interval of
acquisition which is defined at the development stage. As a result, the solution does not
give the possibility of personalization. The smart e-Health Gateway proposed in [9]
provides local storage to perform real-time local data processing and mining. When a
patient’s vital signal is processed, reliable IoT systems are provided to facilitate faulttolerant healthcare services. Zhang et al. proposes in [10] a patient-centric cyberphysical system named Health-CPS aiming to ensure convenient Healthcare service.
The Health-CPS depends on Cloud computing and data analytics to handle the big data
related issues of different healthcare applications. The system is composed of several
layers such as data collection layer, data management layer and data-oriented service
layer. The system collects data in a unified standard. It supports distributed storage and
parallel processing.
Self-adaptative Early Warning Scoring System for Smart Hospital
19
Early warning score systems are used to improve the process of recording, scoring and
responding to changes in measured vital signs of patients. The following paragraphs
present the principal requirements of EWS systems to ensure an efficient detection and
response to clinical deterioration:
• Personalization
The patient vital sings data are the key of successful medical decisions. Each
patient requires a personalized control depending on his situation [7]. For example a
patient with a disease that causes hyperthermia, the temperature is a critical data which
must be visualized with more precision; every second. While other non-critical
parameters can be calculated only every hour for example.
• Medical staff engagement
The early warning scores systems are highly user-dependent and depend on the
appropriate response and actions of the Medical Emergency Teams [4].
• Need for expert’s decision
The warning scores cannot replace specialist’s decision and the importance on
knowing individual patients, they cannot also replace the background to the observations that are recorded [2].
NEWS correlated poorly with the patient’s clinical status within the first 24 h postoperatively [2].
2.3 Related Works
In [8], authors present a solution which takes benefits from the concept of Edge
computing and fog computing in the context of IoT based Early Warning Scores
systems. The solution provides high level services in a Geo-distributed fashion at the
edge of the network. The proof of concept is demonstrated with smart e-Health Gateway called UT-GATE implemented for an IoT-based remote health monitoring system.
The demonstration includes the data flow processes from data acquisition at sensor
nodes to the cloud and the end-users. However, the data has a static interval of
acquisition which is defined at the development stage. As a result, the solution does not
give the possibility of personalization. The smart e-Health Gateway proposed in [9]
provides local storage to perform real-time local data processing and mining. When a
patient’s vital signal is processed, reliable IoT systems are provided to facilitate faulttolerant healthcare services. Zhang et al. proposes in [10] a patient-centric cyberphysical system named Health-CPS aiming to ensure convenient Healthcare service.
The Health-CPS depends on Cloud computing and data analytics to handle the big data
related issues of different healthcare applications. The system is composed of several
layers such as data collection layer, data management layer and data-oriented service
layer. The system collects data in a unified standard. It supports distributed storage and
parallel processing.
Self-adaptative Early Warning Scoring System for Smart Hospital
19