Decision Support System for Health Continuous Vigilance in Industrial Environments

Abstract

Several European statistics confirm that a large number of people have fatal accidents every year in the workplace. For this reason, one of the most important European objectives is to reduce the number of industrial accidents significantly. Fasys Project, focused on factories of machining and assembly operations, aims to achieve this improvement promoting the use of technologies and giving, at the same time, a principal role to the worker. From now on, the worker, who has represented a neglected element in the factories, will be the center of attention. The increase of his security, and the enhancement of his working conditions and health, will be key elements for the Factory of the Future performance. In this paper, a health continuous vigilance system is proposed. The system includes both the monitoring to characterize workers activity and environment, and aspects related to prevention protocols. To manage it, several systems of collecting data are needed. They can be distributed around the factory and monitor, for example, personal and environment data, or get information, for example, from medical knowledge or previous medical information of the worker. Besides, due to the big amount of generated information, intelligent systems for massive data processing are needed. In this way, the information could be easily managed and classified, in order to obtain data from a specific situation that could be required.

Full text

Decision Support System for Health Continuous
Vigilance in Industrial Environments
María Martínez-Piqueras#1, Carlos Fernández-Llatas#2, Carlos Cebrián*3, Teresa Meneu#4
#ITACA - Health and Wellbeing Technologies
Universidad Politécnica de Valencia, Spain
1mamarpi2@itaca.upv.es
2carfell@itaca.upv.es
4tmeneu@itaca.upv.es
*TISSAT, S.A.
Parque Tecnológico de Valencia, Spain
3ccebrian@tissat.es
Abstract— Several European statistics confirm that a large
number of people have fatal accidents every year in the
workplace. For this reason, one of the most important European
objectives is to reduce the number of industrial accidents
significantly. Fasys Project, focused on factories of machining
and assembly operations, aims to achieve this improvement
promoting the use of technologies and giving, at the same time, a
principal role to the worker. From now on, the worker, who has
represented a neglected element in the factories, will be the
center of attention. The increase of his security, and the
enhancement of his working conditions and health, will be key
elements for the Factory of the Future performance.
In this paper, a health continuous vigilance system is proposed.
The system includes both the monitoring to characterize workers
activity and environment, and aspects related to prevention
protocols. To manage it, several systems of collecting data are
needed. They can be distributed around the factory and monitor,
for example, personal and environment data, or get information,
for example, from medical knowledge or previous medical
information of the worker. Besides, due to the big amount of
generated information, intelligent systems for massive data
processing are needed. In this way, the information could be
easily managed and classified, in order to obtain data from a
specific situation that could be required.
I. INTRODUCTION
The International Labour Organization (ILO) estimates that
160 million workers are victims of occupational accidents and
diseases every year [1]. The base of several associations is
that workers should be protected from sickness, disease and
injury arising from their employment. But currently two
million of people lose their lives every year from work-related
accidents and diseases. The suffering caused by such
accidents and illnesses to workers and their families is
innumerable. The standards on occupational safety and health
provide necessary tools for governments, employers, and
workers to establish such practices and to provide for full
safety at work. In 2003, ILO assumed a global strategy to
improve occupational safety and health, which included the
introduction of a preventive safety and health culture, the
promotion and development of relevant instruments, and
technical assistance [1].
One of the European objectives set for 2020 is the
25%reduction in the number of industrial accidents [2-3]. In
order to reduce accidents it is essential to pay attention to the
workers, their single workplaces and to their working
conditions. In this way, if workers had a safer environment,
the number of accidents could be significantly reduced,
implying therefore a reduction in costs. This economic saving
is very important to the general economy of the company. In
addition, these favourable environments make workers feel
more comfortable while they are in the factories, and thus the
efficiency is increased. As a consequence, it is possible to
obtain the maximum efficiency in the factory as a whole,
which also produces economic benefit for the company. From
a healthcare point of view, factories lack normally in an
amount of enough information to allow a holistic care of the
worker. Health data stored by companies are only a small
amount of data, usually stored once a year, and referred to the
physical condition of a person just in a particular moment [4].
For this reason, the future work has to be oriented on new
technological applications to get a factory safer and to reduce
significantly the number of accidents. To control the accidents
it is necessary to anticipate and estimate what can happen. So,
prevention will be a key point. To manage it, it is important to
collect and measure data during a period of time, in order to
evaluate their progress. Consequently, the perfect model
would be a factory in which the risks and health were
controlled at any time.
So, this paper describes the objective to turn the punctual
monitoring into a more frequent and personalized vigilance.
To achieve this goal, it is necessary a continuous and
individual monitoring, respectively. Collecting data of many
people during a long period of time requires collecting a big
amount of information. People are not able to process so much
information, so intelligent systems for massive data
processing are needed. Examples of this kind of systems are:
CEP [5], Process Mining [6], ECA [7]. These intelligent

systems classify data and generate alarms associated to the
worker. Thanks to these alerts and all the other environmental
and personal data stored, it is possible to predict health threats.
Thus, it is possible to act in the most appropriate way for each
worker in particular. These preventive actions, adapted for
each worker, are represented by workflows [8]. In order to
configure templates of these prevention actions, it is required
to develop a visual and intuitive interface that allows experts
to directly do it. In addition, the created protocol must be
automatically executable in computer systems. To manage it,
a specific system has to turn the design of the plan into an
executable format.
Nowadays, the concept of absolutely safe and healthy
environments [9] is increasingly used. In order to get the main
objective, that is, the reduction of industrial accidents, Fasys
Project [10] aims to an absolutely safe and healthy factory,
developing knowledge and technology to guarantee both the
safety and permanent wellbeing of the worker in the factories
of machining, handling and assembly operations of the future.
Through this, the workers will become the key factors of
competitiveness and differentiation of the new productive
model. To solve the lack of the continuous and personal
vigilance and the personalization of the preventive actions,
Fasys proposes a general scheme of a decision support system
for health continuous vigilance in industrial environments.
This scheme includes blocks focused on monitoring,
collecting and managing data, creating diagnosis and
establishing prevention plans. With the purpose of
interrelating this modules, in the project emerges the need to
develop an architecture which connects and relates all of
them.
II. MATERIALS AND METHODS
Nowadays, the number of sensors for monitoring personal
health data is increasing. In addition, sensors that collect
environmental parameters in industrial factories are being
introduced more and more. The problem encountered so far,
and intended to be solved in this project, is that these data are
not connected. The information is only collected in order to
produce isolated diagnosis, but not common results, and the
collected data become less relevant if they are not treated
together. The final decision, in a dynamic environment like a
factory, could be more precise if results came from a study of
a diverse set of parameters.
According to Fasys project, the first step to improve the health
and safety in factories is to increase the personal and
environmental monitored data. Consequently, there is a need
to develop a system able to store all this information.
Nowadays, NOMHAD system is an application able to stored
part of this required information: workers' personal parameters
such as blood pressure, pulse and oxygen saturation. The
system performs a prioritization and an intelligent
management of alarms. These alarms are based on the rules
and protocols accepted by health professionals and by the
health system. This service will combine the information,
through the prioritized alarm list, with the generation of
specific summaries about the state and evolution of the person.
This enables a more efficient management of events.
With the purpose of completing the personal data stored in
NOMHAD, Fasys project proposes a connection to the current
health system. This connection provides a register about the
health state of the person during his lifetime, which collects
data such as his diseases, surgically interventions or pains.
This health system is commonly known as EHR (Electronic
Health Record) [11].
Finally, once the information has been monitored and
classified, the next point is focused on the intervention. With
the aim of representing prevention protocols for this
intervention, workflows are developed. Given the workers
singularity, the adaptation of the prevention protocols is
needed for each one of them. In this way, the elimination of
the occupational hazard is much more effective.
III. RESULTS
Several studies confirm that, in European Union,
approximately 5,500 people per year have fatal accidents in
the workplace [1]. These accidents cost a high price for the
EU and affect all sectors of the economy, mainly enterprises
with less than 50 workers. It has been checked that prevent
work accidents has more benefits than just reducing damages
[1]. In addition, from the European point of view, the
Factories of the Future (FoF) will have fabrication
environments highly dynamics, what entails that workers will
be involved progressively in more diverse situations [10].
The increase in the number of accidents that currently occurs
in factories, added to the European ideas in the factories of the
future, makes workers the key roles in the industrial
environments. The worker health and safety become a central
element in the production process, relating it to the
performance, productivity and efficiency. Consequently, there
is a need to generate systems for health continuous vigilance
of workers.
The Figure1 presents the scheme, developed in Fasys, for a
health continuous vigilance system. This system is based on
five main parts: Monitoring Module; Response Medical
Center; Differential Diagnosis Module; Prevention Plans
Module; and Intervention Module. Each of these parts can
be influenced by a number of external variables and
parameters such as the Electronic Healthcare Record
Given the big amount of generated information in this model,
it is necessary to process all the collected data, since such
amount of information would not be easily understandable by
health professionals. Services and intelligent devices that have
been generated will provide a classification of the monitored
data. Some data will be set inside a normal range, and others
will be out of the settled limits, generating alarms due to this.
Furthermore, this classification will help the doctor to
organize and evaluate all workers’ data and at the same time it
will be able to act more precisely against a particular
diagnostic.

Fig. 1 General scheme for a health continuous vigilance
The content of the blocks shown in Figure1 is:
- Where a group of personal data is collected is the
Monitoring Module.
- All these personal data, obtained from the monitoring and a
group of environment variables from several sensors in the
factory, are joined together in the Response Medical Center.
As environment variables, one can understand parameters
such as, environment temperature or humidity, that is,
particular characteristics from the workplaces at which the
worker can stay during a work journey.
The Response Medical Center allows to filter and organize the
population depending on the changeable rules and on the user
role. So, it is in this module where the first amount of data is
collected, creating, as a result, personalized records of the
workers and establishing alerts which make easier the task of
health professionals.
From this point on, next steps are already focused on getting
an action line according to the problem detected.
- The information stored in the previous module is not enough
to make a complete diagnosis. So, data from other sources are
needed such as:
• Data from a medical base of knowledge (it contains
relations among diseases, risks, medical tests, medical
recommendations, etc).
• Personal data from the health system, which include the
previous medical history and it is known as Electronic
Healthcare Record (EHR).
• Trend analyzer data. This system is in charge of detecting
how some parameters of a person are changing during the
pass of time. These parameters can be added to the
absolute values in order to get a more complete
evaluation of the person.
• Evaluation module results. It can be defined as a
“photograph of the person” in a particular moment, with
no need to detect a problem.
These four mentioned sources are the subsystems shown in
the general scheme, Figure 1, which provide important
information to the main blocks. To manage all this
information, Fasys has developed the Differential Diagnosis
Module, shown in Figure1. This module, through intelligent
systems, helps in decision making by health personal.
- The next step is to reach the Prevention Plans Module,
where it is defined how to act. The measures to be taken can
be of two types: on one hand a medical diagnosis and on the
other hand a technical diagnosis, for instance a redesign of the
workplace. It is important to remark that these measures are
not exclusive. According to this, different levels of action can
be established. That is, from very complex levels to more
simple levels such as, for example, reminder panels.
In addition, the prevention actions carried out in this module
can be conducted at three levels. At the first level, the system
reacts automatically. When one of the collected data reaches a
condition that the professional wants to be controlled, there is
an automatic reaction. This associated reaction can be the
activation of an alarm, a protocol in a situation of risk, etc.
These automatic reactions are achieved using ECA rules-
Event, Condition, Action. At the second level, health
professionals receive the alerts and react to individual
workers. The reaction of professionals can be the assignation
of a prevention plan developed before, or the assignation of a
prevention plan modified for the worker situation in
particular. These ways that define the processes are called
workflows. The third level is in charge of providing
knowledge for the other two levels, improving the protocols,
adapting them to new situations and personalizing the
recommendations. Innovative intelligent tools are used to
manage it.
- Finally, the Intervention Module is responsible for
performing the particular actuation selected for the problem in
question.
Fasys system is considered cyclic and of a continuous learning,
in a way that, after the Intervention Module, it starts again
from the Monitoring Module.
Another important aspect to take into account is the personal
privacy. As a consequence of this, only a few people will have
access to the EHR (Electronic Healthcare Record), to the
personal variables, and to the personal diagnosis.
From all five main parts, it is going to be emphasized the
Response Medical Center. All collected data in this module
has to be processed by an application called NOMHAD. In the
immediate future, ICT (Information Communication

Technologies) will have an outstanding role in the health
sector. This will allow the improvement of the current
processes, making them more accessible and efficient.
Important efforts have been performed to extend its use in the
health sector.
This module receives automatically monitored data from all
workers. This information is treated to prioritize and manage
more efficiently the attention and the available resources for
the factory population.
So far, the options managed by NOMHAD are specifically the
following:
• To create a patient. The professional will fill the
administrative worker data and the medical relevant data
for a future evaluation. Patients will also be assigned to
health professionals.
• Reception and Display of Monitoring. The system stores
the monitoring data of the person. It stores them into a
database related to the personal health record in order to
be used by health professionals in the future. The
monitoring data are displayed in the right way, either in
graphical form, numerical, image, etc
The data stored in the system are processed on arrival. A
set of several rules, defined by the doctor and adapted to
the personal profile, are applied. These rules allow the
system to detect potential anomalies found in the data, in
order to take decisions.
For the definition of the rules, health professionals will
have a tool to help themselves with this task.
• To configure alerts: The system will have an alert module
that, based on the monitoring data and the limits defined
as optimal by a professional, will be able to detect
whether the data stored are acceptable or not.
• The possibility of assigning questionnaires: The patient
mood could be extracted from these questionnaires.
Health professional will choose the questionnaire and will
have the possibility to personalize it depending on the
needs of each worker in particular.
These questionnaires will be available in the future in
case the professionals want a subsequent consulting or
validations.
• Monitoring: The measurements can be obtained with
usual external devices, whose information is introduced
afterwards in the system or it can be used integrated
elements into the system (controlled, for example, by
Bluetooth and transmitting the captured data directly to
the system). The design of the system can be extended to
introduce new devices.
The following shows the main screen of the application.
Fig. 2 NOMHAD system
When the general modules and the relation among them are
defined, an architecture must be created, that is to say, a way
to guarantee the connection and interoperability among them.
To get information from the worker environment and his
personal parameters, it is necessary to interconnect sensors
and services in a fault tolerance and decentralized way. This
process, complex and highly interconnected, can be solved
using Choreography of Services [12-8]. This means that the
choreographied processes are independent and can
communicate each other to define execution flows. This
model makes easier the connection and disconnection of
services dynamically, and at the same time it is capable of
using different kind of sensors and configurations. This
approach is shown in Figure 3.
The use of choreography to interconnect services requires also
the use of a common exchange language to allow the services
to understand each other. This can be performed by an
architecture which includes a Semantic layer in the
Choreographer. The reason to do that is to improve the
intercommunication among sensors, actuators and services of
the system.
The ontologies [12] are a solution to describe concepts
formally. Concretely, an ontology is a formal and explicit
specification of a shared conceptualization. It provides a
common vocabulary that can be used to model the kind of
objects and/or concepts and its properties and relations. The
reasoners [12] are software applications allowing the semantic
seek in the ontologic description. Using this technology, it is
possible to describe semantically the sensors and data services,
giving them the ability of having a more complete
understanding of the collected data and the services actions.
The use of services of Ontologies and Reasoning Systems to
describe the data coming from the sensors, makes possible to
get a more precise interpretation and to detect automatically
the sensors and services available at any time.

Fig. 3 Fasys Architecture
In order to illustrate graphically the action to perform and the
standards describing the flow followed by actions, it is
possible to use workflows. They are a formalization of the
process to be automated. Some workflow languages can be
executed automatically. This is known as a workflow
interpretation. The automatic interpretation of a workflow is
done by a workflow engine, which can complete the actions
explained in a workflow, in the order and with the derivation
rules specified in it. Workflows can be employed by people
who are not experts in programming for the health area. For
that reason and thanks to these modules, health professionals
are able to design and modify the protocols to be executed
automatically.
A Services Orchestrator [8] is included in the architecture and
moreover, it is connected to the Choreographer which accepts
the use of Workflows.
Finally, the objective of this paper is to show the current
situation of the health data warehouses related to Fasys project.
Nowadays, there is a health system ready to be used by health
professionals. It stores the medical data of the patient from the
point of view of the assistance process, and it is owned by the
current healthcare system. This repository is called EHR
(Electronic Healthcare Record).
With the purpose of improving the characterization of the
person and his environment, EHR data have been increased.
This new amount of information is stored, together with the
EHR information, in other repositories. These repositories are
known as PHR (Personal Health Record) [13] and can collect
data such as habits, preferences, information about the family,
moods, customs or nutritional profile. A PHR adapted to the
needs of Fasys requirements is being developed. This kind of
repositories is owned by the person, who has the option to
share it with people he chooses.
When an employee goes to work in a factory for the first time,
health professionals ask him to download his previous EHR,
in order to have the personal file (PHR) more complete for the
final diagnosis.
In general, current PHR contains a summarized version of
EHR ready for patients and, in some cases, home monitoring
data. PHR developed by Fasys is based on the following
aspects:
• It is focused on workplace health.
• It allows patient to introduce data (automatically or
manually)
• It allows an exchange of information with the healthcare
system (EHR)
• It includes an option to generate summaries to share
information with others PHR. One of the advantages, for
example, is when a worker goes to work in other factory.
If the new factory has the Fasys system, his PHR could be
downloaded in the system of the new factory in order to
have a more complete file.
• Stored Data can also be extracted for consultations in case
health professionals need to do. Consequently, there must
be an Access Control. With this control, it is ensured that
these personal data can only be seen by authorized
people. If the data have to be used for statistical studies, it
must be made anonymous. So, the results of the studies
will not be related to people in particular.
In addition, and with the objective of validate the developed
work in the different stages of the project, several meetings
with experts have been done. On the one hand, the first
meetings had the objective to clarify the main points to be
considered for a health vigilance scheme. And on the other
hand, the last meetings had the mission to validate the
developed scheme of health vigilance. Their point of view is
vital to perform a good scheme of a support decision and a
health continuous vigilance, directing it to solve the real needs
that contain each covered area.
IV. CONCLUSIONS
According to the project objectives, a scheme of health
continuous vigilance has been designed. It provides a
workable solution in order to improve the current healthcare
system in the factories of machining, handling and assembly
operations. It is possible to obtain a more continuous
monitoring of the worker, improving his own health and
making the factory safer and healthier. To achieve this, it is
necessary to increase the number of variables obtained from
the environment of the worker, from personal parameters, and
to combine them with the medical knowledge and the
actuation protocols.

Future steps will be focused on the detailed definition of some
modules of the scheme of health vigilance that have to be
completed. The optimal way to fit all input and output data
should be studied in depth. It is important to remark the proper
connection with other modules and smaller subsystems which
they are related with.
Up to now, the application included into the Response
Medical Center collects data from personal monitoring. In the
future, this application will be improved, introducing relevant
data for the project, such as: environment variables (room
temperature, humidity…) or type of machine used (which is
related to one kind of strain or another).
Besides the introduction of new parameters, other two
innovations for this application are being studied:
• Possibility to carry out videoconferences between the
doctor and the worker. This action will improve the
continuous monitoring. On the one hand, it will be useful
for external consultations, in case the doctor is not in the
factory. And on the other hand, it will be useful to raise
remote queries to specialists.
• Mobility. The goal is to build a little version of the
application. It can be used on small devices like a tablet.
Thus, the information will be available anywhere,
carrying out a supervision of the processes and a
management of alerts in real time.
REFERENCES
[1] International Labour Organization
http://www.ilo.org/global/standards/subjects-covered-by-international-
labour-standards/occupational-safety-and-health/lang--en/index.htm
(Last Access: November 2011)
[2] Balogh I, Orbaek P, Winkel J, Nordander C, Ohlsson K, Ektor-
Andersen J, et al. (2001). “Questionnairebased mechanical exposure
indices for large population studies-reliability, internal consistency and
predictive validity”. Scand J Work Environ Health; 27(1):41–48.
[3] Leijon O, Wiktorin C, Harenstam A, Karlqvist L, MOA Research
Group (2002). “Validity of a self-administered questionnaire for
assessing physical workloads in a general population”. J Occup
Environ Med; 44(8):724– 735.
[4] J. Stranks, 2006. The manager's guide to health & safety at work.
London: Kogan Page Limited
[5] Segev Wasserkrug, Avigdor Gal, Opher Etzion, and Yulia Turchin.
“Complex event processing over uncertain data”. In Proceedings of the
second international conference on Distributed event-based systems,
DEBS '08, pages 253{264, New York, NY, USA, 2008. ACM.
[6] Carlos Fernández, Juan Pablo Lázaro, and Jose Miguel Benedí.
“Workfow mining application to ambient intelligence behavior
modeling”. In Universal Access in Human-Computer Interaction,
volume 5615 of Lecture Notes in Computer Science, pages 160{167.
Springer, 2009.
[7] E. Behrends, O. Fritzen, W. May, and D. Schubert. “An ECA Engine
for Deploying Heterogeneous Component Languages in the Semantic
Web”. In Web Reactivity (EDBT Workshop), Springer LNCS 4254,
2006.
[8] Carlos Fernández-Llatas, Juan B. Mocholí, Carlos Sánchez, Pilar Sala,
Juan Carlos Naranjo “Process choreography for Interaction simulation
in Ambient Assisted Living environments” The 12th Mediterranean
Conference on Medical and Biological Engineering and Computing
MEDICON 2010 2010
[9] J.M. Stellman (ed.), 1998. Encyclopaedia of occupational health and
safety, Volumen 1;Volumen 5. Geneva: International Labour Office
[10] Fasys: “Fábrica Absolutamente Segura y Saludable”.
Available:http://www.fasys.es/en/
[11] Himss, http://www.himss.org/ASP/topics_ehr.asp (Last Access:
November 2011)
[12] Carlos Fernández-Llatas-Llatas, Juan B. Mocholí, Agustín Moyano,
Teresa Meneu “Semantic Process Choreography for Distributed Sensor
Management” International Workshop on Semantic Sensor Web - IC3K
2010 2010
[13] PHR Reviews. Available: http://www.phrreviews.com/