SICOBIO
Consolidation and Analysis System of Biomedical Information
William Enrique Parra Alba and Alexandra Pomares Quimbaya
Pontificia Universidad Javeriana, Carrera 7#40-62, Bogotá, Colombia
Keywords: Consolidation of Biomedical Information, Physiological Sensors, Remote Monitoring of Patients, Decision
Support, Data Analysis, Statistical Techniques, Fuzzy Logic.
Abstract: Improving information systems and guaranteeing the quality of health services are current challenges for
global governments. Financing problems and congestion in health care centers, increased demand for services
due to aging population, mobility issues and insecurity, risks of accidents and contagion of diseases in care
centers are some of the problems to be solved. In order to reduce these problems, it is necessary to take
advantage of technological advances to give alternative solutions. This paper presents SICOBIO, a system for
monitoring patients who finish their treatment at home, and facilitates the integration of devices and analysis
algorithms. The information from remote sensors is homogenized, consolidated, and analyzed in a
synchronous and asynchronous manner. Through synchronous analysis, the patient's risk is assessed and alerts
are generated to his caregivers, and through asynchronous analysis, decision-making is supported with
statistical and data mining techniques applied to historical information. SICOBIO was validated with devices
for remote monitoring of physical activity, heart rate and weight. It demonstrated its scalability to incorporate
new equipment and new algorithms of analysis and delivery of alerts. Its potential utility was validated
through the Technology Acceptance Model (TAM).
1 INTRODUCTION
Improving information systems (Tovar-Cuevas and
Arrivillaga-Quintero, 2014), and guaranteeing access
to health care services for citizens in conditions of
equality and quality, are some of the current
challenges (Scheil-Adlung, 2013). The increase in
life expectancy (Celler and Sparks, 2015),
employment informality, and the low percentage of
contributors to the funds that support the health care
services, compromise the sustainability of the system
(Zapata Jaramillo and Sánchez Villavicencio, 2012).
With the increase in life expectancy, the demand for
health services for the elderly is increased, who must
assume travel costs, insecure situations, risks of falls,
long travel times, long waiting times, and the risk of
contagion of diseases in health care centers (Morelos
Ramírez et al., 2014). These conditions that generate
higher costs to the system, can be avoided if the
patient is prevented from moving to the care centers
or if the patient is able to complete his recovery at
home, as long as he is guaranteed adequate follow-up
(González et al., 2012).
Previous works have devised different ways to
improve health care scheme, through the use of
technological innovation for home health care
(Chandler, 2014), (Winkley et al., 2012), (Dogali
Cetin et al., 2015), (Lamonaca et al., 2015),
(Rajkomar et al., 2015), (Baig et al., 2013).
Developed countries, such as France and Australia,
have implemented services for the remote monitoring
of patients supported in information technologies
(Basilakis et al., 2010). These systems report the
readings of vital signs to data processing centers,
where the information is used for monitoring and
control of patients.
For the home health care needs, diverse works
have been developed. Some focused on electronics
for the collection and transmission of data (Gallego
Londoño et al., 2010), (Toshiba Semiconductor &
Storage Products Company, 2013), (Jiménez
González et al., 2014) and others focused on their
analysis and processing (Villar-Montini, 2009),
(Dogali Cetin et al., 2015), (Skubic et al., 2015),
(Fanucci et al., 2013), (Guevara-Valdivia et al.,
2011), (Abo-Zahhad et al., 2014), (Cheng et al.,
2009), (Leite et al., 2011). The studies reviewed show
180
Alba, W. and Quimbaya, A.
SICOBIO - Consolidation and Analysis System of Biomedical Information.
DOI: 10.5220/0006363701800187
In Proceedings of the 3rd International Conference on Information and Communication Technologies for Ageing Well and e-Health (ICT4AWE 2017), pages 180-187
ISBN: 978-989-758-251-6
Copyright © 2017 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
that the supply of sensors is not very wide in terms of
the number of signs that can be monitored, however,
there is interest and many ongoing investigations that
allow inferring a tendency to increase the sensors
supply. Companies such as Fitbit, iHealth Lab, MIR,
Activ8rlives, OMRON, NONIN, A&D Medical,
BEURER, Vitaphone, Movisens and others, have
vital signs reading equipment with the ability to
centralize data on their own servers. These
consolidated data can be consulted subsequently by
users through services provided by each
manufacturer.
Although diverse technologies have been
developed that integrate information from remote
sensors, no work has been identified that consolidates
and analyzes data from heterogeneous manufacturers.
Per the above, it has been found the opportunity to
build SICOBIO, a system for the consolidation and
analysis of biomedical information from
heterogeneous and remote devices. The system
provides synchronous and asynchronous analysis.
The synchronic analysis contains a fuzzy logic model
that qualifies the patient's risk level based on a
combination of vital signs, and generates alerts that
are prioritized and delivered to the patient's
caregivers. Asynchronous analysis contains statistical
and segmentation models for both individual follow-
up and follow-up at the patient population level.
These models of analysis facilitate decision-making
for medical personnel, and early discharge program
managers. In addition, they keep family members and
caregivers informed about the patient's health status.
To present SICOBIO, this article is distributed as
follows: section two, presents related works, and
section three, presents the detail of SICOBIO. First,
the general overview, then the technical architecture,
and then a section for the consolidation component
and one for the analysis component. Section four
presents the system validation. Finally, section five
presents the conclusions and future work.
2 RELATED WORKS
The related works will be analyzed from the
perspective of five proposed process for the remote
monitoring of patients (figure 1). The first process
consists in obtaining patient data that can be supplied
with tools such as commercial sensors, wearables,
custom sensors, and sensors installed on
smartphones.
Figure 1: Processes to remotely monitoring patients.
The second is the process of data transmission,
consisting of the technologies needed to send the data
obtained to a processing center. Some technologies
available are: GSM, 3G, 4G, WiFi, Bluetooth and
ZigBee. The third process is consolidation in which
formats of each manufacturer are interpreted, the data
is homogenized, and each patient information is
separated. The fourth process is the data analysis, in
which the data becomes useful information for the
different stakeholders in the care of the patient. The
final process is the delivery of results obtained from
the analysis of the information from the patient.
2.1 Sensors to Obtaining Biomedical
Signals
There are works at both, the research (Gallego
Londoño et al., 2010) and commercial level (Toshiba
Semiconductor & Storage Products Company, 2013).
On the commercial level, there are companies
dedicated to the sale of wearable physiological
sensors that can be worn as part of the dress or as
accessories in the form of bracelets. These sensors
measure different variables such as energy
expenditure, weight, steps performed, levels of skin
conductance and heart rate, among others.
With the arrival of new sensors in smartphones,
practical uses for health care has been discovered,
such as detection of falls, bad postures, eye problems
and even respiratory problems (Lamonaca et al.,
2015). Likewise, potential applications of these
devices have been discovered, such as portable
instruments for the visualization of vital signs
obtained in remote sensors (Jiménez González et al.,
2014), or as tools for the capture and transmission of
data obtained by not automatic means.
2.2 Home Health Care Systems
This section presents the group of works more related
to SICOBIO, and, for this reason, at the end of the
section a comparative analysis of these works is
presented.
SICOBIO - Consolidation and Analysis System of Biomedical Information
181
2.2.1 Telemedicine Systems
They are used for the tracking and remote monitoring
of patients. Some works are responsible for
monitoring heart disease through the use of
pacemakers or automatic defibrillator, with remote
satellite surveillance (Guevara-Valdivia et al., 2011)
(Villar-Montini, 2009). In those works, the patient
data is transmitted to a service center where it is
processed and analyzed by medical staff. From there,
an order can be transmitted to the device to adjust a
certain parameter (for example the heart rate)
(Fanucci et al., 2013). Data transmission is done in
two ways: the usual data, are transmitted in batch
once a day; and unusual data, according to the system
parameterization, is transmitted in near real time to
ensure the timely reaction of medical staff.
As a complement to the systems of remote
monitoring of patients, technologies for home
monitoring has been developed (Winkley et al.,
2012). These systems identify people´s behavior
patterns, and over time may infer unusual behaviors
for the generation of reminder, warning or danger
messages (Agreda and Gonzalez, 2014).
2.2.2 Systems for Signals Analysis
These works deal with both, the collection and
transmission of data, as well as their analysis and
processing. In order to obtain vital signs, portable
technologies are being developed to obtain data from
the patient, without them being conscious (Baig et al.,
2013). The data obtained by these devices are sent to
a central server for processing and analysis (Dogali
Cetin et al., 2015).
Due to the risk level of some patients, it is
necessary to have redundant communication channels
to ensure that the warning signals arrive to their
receiver and guarantee an immediate reaction from
medical staff to any eventuality. In this meaning,
prototypes have been developed (Abo-Zahhad et al.,
2014) for ECG (Electrocardiogram), SPO2 (Blood
Oxygen Saturation), temperature and blood pressure
sensors, which report data to a central computer for
diagnosis of chronic diseases. These systems use two
channels depending on the type of information they
are going to transmit. For normal readings, they use
the standard network. In the case of atypical readings,
they use GSM / GPRS networks to transmit alert
messages to emergency services and caregivers.
2.2.3 Systems for Decision Support
These systems obtain data and send it to a centralized
system that maintains a knowledge base from which
medical personnel can take decisions. For these
systems rule engines are used in which physicians can
set parameters according to the patient (Basilakis et
al., 2010). When the monitoring system detects
signals that exceed the configured thresholds, the
system can generate alerts.
For the analysis of data, there are works that
through data mining techniques, perform
synchronous and asynchronous analysis. In (Chauhan
et al., 2010), they apply a cluster algorithm for the
analysis of databases with medical information of
patients with cancer. In (Dudik et al., 2015), they
illustrate a comparative analysis of some density
based cluster algorithms for the analysis of patients
with swallowing problems. In (Duclos-Gosselin et
al., 2015), they propose data mining techniques to
solve the problem of contracting nosocomial
pneumonia, and in (Tomar and Agarwal, 2013), they
present the utility of classification, cluster,
association and regression algorithms in the health
care. They propose using predictive mining
techniques, to identify patients at high risk of disease,
and / or to validate different treatment options.
2.2.4 Alert Generation Systems
These systems evaluate the consolidated data, and
identify unusual conditions, from which generate
notifications to those interested in patient care. In
intra-hospital systems, they generate warning signals
when monitoring instruments detect abnormal signs,
and they are integrated with other systems using
protocols such as HL7 (Cheng et al., 2009). The
generated alerts are analyzed and qualified by
healthcare professionals regarding their clinical
relevance, to determine if the alert is good (Skubic et
al., 2015). With these systems, it is possible to
determine when a patient is stable, and ready to
release the bed that can be occupied by a patient who
requires it most urgently (Leite et al., 2011).
2.2.5 Comparative Analysis of Home Health
Care Systems
From related works, a comparative analysis was made
from the perspective of the following criteria:
Real-time transmission: Indicates whether the
system has support for real-time or batch
transmission. Protocol: Classification of protocols
used by the sensor to send the obtained data.
Wearable: Indicates whether the sensor can be part of
the dress. Monitored Signs: Provides vital signs data
monitored by the sensor.
ICT4AWE 2017 - 3rd International Conference on Information and Communication Technologies for Ageing Well and e-Health
182
Table 1: Comparative Systems for Remote Patient Monitoring.
Work Real time Protocol Wear
able
Monitored Signs Gateway Third-
party
Analy
sis
(Guevara et
al., 2011)
YES, and
Batch
Wireless
GPRS/SMS
YES
Heart rate
(Defibrillator)
CardioMessen
ger (Germany)
NO YES
(Fanucci et
al., 2013)
YES, and
Batch
Bluetooth/
ADSL/GSM
NO Chronic Heart Failure
(CHF)
Personal
Computer
NO YES
(Dogali Cetin
et al., 2015)
YES WiPort
Ethernet
NO Temperature / Blood
pressure
Wireless
Ethernet
NO NO
(Abo-Zahhad
et al., 2014)
YES, and
Batch
GSM/GPRS
NO Temperature / Blood
pressure
Mobile-Care
Unit
NO YES
(Basilakis et
al., 2010)
NO Internet NO Chronic Obstructive
Pulmonary Disease
Phone NO YES
(Skubic et al.,
2015)
YES Infrared NA Early Detection of
Diseases
WSN NO YES
(Leite et al.,
2011)
YES SMS, Email NA Systolic and diastolic
blood pressure.
NA NO YES
Gateway: Device used to obtain data from the
sensor and responsible for sending it to a processing
or display device. Consolidate third-party data: If the
system is enabled to consolidate and homogenize
third-party data. System of Analysis of Information: If
it contains a server for processing and analysis of the
information collected.
From the comparative work, it was possible to
establish that there are gateways to proprietary
technologies such as CardioMessenger and for
opened technologies through smartphones, laptops
and sensor networks. GSM/GPRS mobile
communication protocols are used in critical signal
tracking systems such as ECGs and pacemakers.
Most works, transmit data in "near real time", and
only few use wearables. Most perform analysis on the
data obtained in their proprietary formats, but none
consolidates data from third-party technologies, a
topic that will be developed in the next chapter.
3 SICOBIO
3.1 General Outline of the Solution
Figure 2 presents the general outline of the proposed
solution for monitoring patients at home. The block
"A. Means for Home Care ", represents the patients in
their home, with their different monitoring devices,
and considering different protocols that can have
these equipment, for the transmission of the readings
to the server. Block "B. Consolidation and Analysis",
is composed of interfaces that integrate third-party
servers, and interfaces that directly receive signals
from a gateway. These interfaces that are installed in
the server must implement the functionalities of
interpretation of the delivery formats and of data
validation and homogenization. When the data is
consolidated, the control is passed to the analysis
server, in which synchronous analysis are performed
for early identification of risk signs, and
asynchronous analysis, over historical data of a
patient, or patient set. If the system identifies a risk
signal, it sends a warning signal to the group of people
responsible for patient care.
3.2 Technical Architecture of
SICOBIO
SICOBIO has been designed under a layered
architecture (figure 3). The obtention layer represents
the devices and components that obtain and transmit
patient data. The consolidation layer contains the
components that process the data received from the
obtention component, and the analysis layer contains
the models for analysis and alert generation.
Data collection can be achieved using wearable
physiological sensors, monitoring systems, sensors
on smart phones, or manually entered information.
Some sensors studied send the data of the readings to
proprietary servers from which they must be
downloaded by an adapter that reports them to
SICOBIO for consolidation and analysis.
SICOBIO - Consolidation and Analysis System of Biomedical Information
183
Figure 2: Outline of a Home Health Care Support Scheme.
When a new message arrives, the service passes it
to a reception controller in which the source format is
identified, and the appropriate structures are
instantiated for its treatment. After obtaining the data
or the data set, the system identifies the type of signal
to which each datum belongs, and performs the tasks
of homogenization and validation, to ensure that the
data is in the units accepted by the system. The system
persists the data in the database and sends an analysis
request message to the analysis component which
receives the request with the processed data and
depending on the received data, it instantiates the
appropriate analysis method. For example, if you
receive data on systolic blood pressure and oximetry,
the analysis based on fuzzy logic is instantiate to
obtain a risk rating. The methods of analysis in
SICOBIO, have been cataloged thus:
Synchronous analysis: identifies risk signals and
generates alerts that are sent to the alert controller,
which retrieves the data of those interested in
receiving the notification according to the level of
urgency (High, Medium, Low). It also instantiates the
specialized component in the delivery of the alert and
sends the alert. For a high urgency level, it uses the
SMS component, for a medium urgency level, it uses
the e-mail component, and for a low urgency level,
the persistence component.
Asynchronous analysis: Analysis performed on
historical information of a patient so that the treating
physician can see the evolution of the patient
regarding a given treatment and on consolidated
information about several patients to help understand
phenomena related to the population.
All the system configuration, such as data of
patients, physicians, caregivers, equipment, units of
measurement, pathologies, programs, users of the
system and standardized and classified readings are
found in the early discharge component (figure 3).
As follows, the components that make up each of
the layers of the system and a description of their
responsibilities are presented:
3.2.1 Obtention Layer
This layer is responsible for obtaining and
transmitting data. Its main components are:
Native components: responsible for transmitting
the data to the consolidation service in its native
format. Adapter components: responsible for
connecting to third-party servers, retrieving
information from the readings of a device, and
transmitting them to the consolidation service.
Manual reporting components: responsible for the
recording of readings obtained manually, and
transmit them to the consolidation service.
3.2.2 Consolidation Layer
This layer is responsible for the homogenization and
consolidation of data. Its main components are:
Data receiving component: Web service in charge of
receiving the message. Controller component:
Responsible for resolving the format of the message
and delivering it to the appropriate component for
processing. Formats processing components:
Responsible for processing the corresponding format.
Signal processing components: Responsible for
retrieving the data of the processed format, and apply
the homogenization and validation on each data
reported. In homogenization, the data is passed to the
unit of measure accepted by the system, and in the
validation, the data is marked as normal or atypical.
(Example, a SPO2 > 100, is an atypical value). Data
persistence component:
ICT4AWE 2017 - 3rd International Conference on Information and Communication Technologies for Ageing Well and e-Health
184
Figure 3: Architecture of SICOBIO.
Responsible for persisting the homogenized and
validated data. Sending analysis requests component:
Responsible for sending a message to the analysis
component, with details of the data or set of data that
has been processed.
3.2.3 Analysis Layer
Layer responsible for synchronous and asynchronous
analysis, delivery of alerts and parameterization of
the system. Its main components are:
Distribution component: Receives the analysis
requests, and passes them to the controller
component. Controller component: Determines the
type of synchronous (light) analysis applicable to
received data. Asynchronous controller component:
Controls the processes of analysis of historical data.
Data analysis components: It is an abstract and
extensible component, responsible for performing the
synchronous and asynchronous analysis, and
generating the patient's risk rating. It supports
decision making to medical staff and functional
administrators through trend charts, scores, boxes and
whiskers, point clouds, value tables and segmentation
algorithms. Alerts component: It is an abstract and
extensible component, in charge of selecting the most
appropriate channel for sending the alerts per the
level of risk of the patient. Data presentation
component: Displays the result of the data analysis.
Early discharge component: Registers early
discharge processes, and maintains system
configuration (doctors, caregivers, equipment, etc.).
4 VALIDATION
Validation is presented in two sections, one for the
consolidation model, and the other by the analysis
model.
4.1 Consolidation Model Validation
For this validation, we performed tests with three
Fitbit devices that measure weight (aria), physical
activity (flex), and heart rate (HR). The three devices
were associated to different people, and data
collection was maintained between the first and
second semester of 2016.
The data was synchronized, using an adapter that
takes the readings from the Fitbit server, and reports
them to the consolidation component. The adapter
was designed to take readings with a frequency of five
minutes. This validation obtained the following
results: 121 data was processed in batches,
corresponding to readings taken between January 1,
2016 and April 30, 2016 through flex device. On May
2016, 30 records of daily heart rate readings were
processed, taken with the HR device. A quantitative
verification of the values of the physical activity data
loaded for each one of the dates in the database
PostgreSQL was carried out, with respect to the data
that the fitbit server registered for the same dates and
the difference was zero.
The validation of the second semester (from June
1 to November 7, 2016) occurred in an interval of 160
days, during which, 27 readings were obtained from
the flex device, 112 readings from the HR device, and
57 readings from the aria device. These records were
processed and uploaded to the database, in which they
were verified and validated using the same technique
used for the data of the first semester.
A second validation process was performed by
capturing vital signs data from an application
developed in Android. For these tests, data was
entered in different units (for example: Celsius
degrees, Fahrenheit degrees, Reaumur degrees, beats
per minute, beats per hour), and the system correctly
performed the transformations to the accepted units.
SICOBIO - Consolidation and Analysis System of Biomedical Information
185
Temperature (Celsius degrees), heart rate (beats per
minute). Likewise, depending on the vital sign, and
its unit scale, tests were performed to prove that the
system adequately marks the typical or normal data,
and differentiated them from atypical data, or out of
the range expected for a given unit of measure.
4.2 Analysis Model Validation
This model was validated from a technical and
functional point of view. For the technical validation,
vital sign data was sent from a smartphone. The
system consolidated and sent the processed data to the
analysis model in which synchronous analyses were
applied for systolic blood pressure and oximetry
signals. Using a fuzzy logic model, the system rated
risk and generated different types of alert. The
calculations performed by the system were validated
against previously calculated control data, and
identical results were obtained.
The functional validation of the analysis model
was performed using the Technology Acceptance
Model- TAM (Venkatesh and Bala, 2008). The
Direction of Aging, the Office of Biomedical
Management of the San Ignacio University Hospital
and the Office of Public Health of the Pontificia
Universidad Javeriana were involved.
The scores resulting from the validation process
were rated on a scale from 1 to 7 (1-Strongly
Disagree, 2-Moderately Disagree, 3-Somewhat
Disagree, 4-Neutral, 5-Somewhat Agree, 6-
Moderately Agree, 7-Strongly agree). The average
rating was 5.7 / 7.0. The aging direction rated it at 5.7
/ 7, the biomedical management office rated it at 5.9 /
7 and the public health office rated it at 5.5 / 7. The
quality output item was best validated by the
Direction of Aging. Likewise, the evaluations of
behavioral intention and perceived utility have had a
positive evaluation, which also validates the practical
utility of the system.
5 CONCLUSIONS AND FUTURE
WORK
According to the related works, it can be seen the
interest that exists for the development of systems of
remote monitoring of patients. In the short term, it is
expected that a wide range of commercial and non-
commercial devices will be available to obtain
biomedical signals.
The SICOBIO model offers scalability for the
integration of new devices, for the incorporation of
new analysis algorithms, and for the integration of
technologies for the delivery of alerts. Its scalability
was validated through transparent incorporation of
devices for the measurement of the heart rate,
physical activity, and weight measurement. The
architecture proposed and the technologies used for
the implementation of the prototype, allow the
deployment of the system in a high availability
environment. Our approach is the first one including
scalability in consolidation, analysis and delivery of
alerts. This way, the system can evolve as the
technology evolves.
As future work, it is suggested the integration of
new devices, new analysis models, and new
components for the delivery of alerts (e.g. Whatsapp,
automatic calls) that take advantage of the scalability
offered by the system design. Likewise, the
exploitation of historical information through mining
predictive models is suggested. It is suggested to
enrich the prototype of the mobile APP so that,
through the sensors installed on a smartphone, it
obtains and transmits data of the patient's activity, and
/ or as a channel of communication between the
doctor and the caregiver.
ACKNOWLEDGEMENTS
This research was carried out by the Center of
Excellence and Appropriation in Big Data and Data
Analytics (CAOBA). It is leading by the Pontificia
Universidad Javeriana Colombia and it was funded by
the Ministry of Information Technologies and
Telecommunications of the Republic of Colombia
(MinTIC) through the Colombian Administrative
Department of Science, Technology and Innovation
(COLCIENCIAS) within contract No. FP44842-
anex46-2015.
REFERENCES
Abo-Zahhad, M., Ahmed, S.M., Elnahas, O., 2014. A
Wireless Emergency Telemedicine System for Patients
Monitoring and Diagnosis. Int. J. Telemed. Appl. 1–11.
Agreda, J.A., Gonzalez, E., 2014. Ambient intelligence
based multi-agent system for attend elderly people, in:
Computing Colombian Conference (9CCC), 2014 9th.
IEEE, pp. 115–120.
Baig, M.M., Gholamhosseini, H., Connolly, M.J., 2013. A
comprehensive survey of wearable and wireless ECG
monitoring systems for older adults. Med. Biol. Eng.
Comput. 51, 485–495. doi:10.1007/s11517-012-1021-
6.
Basilakis, J., Lovell, N.H., Redmond, S.J., Celler, B.G.,
2010. Design of a decision-support architecture for
management of remotely monitored patients. IEEE
ICT4AWE 2017 - 3rd International Conference on Information and Communication Technologies for Ageing Well and e-Health
186
Trans. Inf. Technol. Biomed. 14, 1216–1226.
doi:10.1109/TITB.2010.2055881.
Celler, B.G., Sparks, R.S., 2015. Home telemonitoring of
vital signs - Technical challenges and future directions.
IEEE J. Biomed. Health Inform. 19, 82–91.
doi:10.1109/JBHI.2014.2351413.
Chandler, D.L., 2014. How technology will be transforming
both inpatient and at-home care. IEEE Pulse 5, 16–21.
doi:10.1109/MPUL.2014.2355298.
Chauhan, R., Kaur, H., Alam, M.A., 2010. Data clustering
method for discovering clusters in spatial cancer
databases. Int. J. Comput. Appl. 0975–8887 Vol.
Cheng, P.-H., Lai, F., Lai, J.-S., 2009. A service-oriented
healthcare message alerting architecture in an asia
medical center: A case study. Int. J. Environ. Res.
Public. Health 6, 1870–1881.
doi:10.3390/ijerph6061870.
Dogali Cetin, G., Cetin, O., Bayilmis, C., 2015. A real-time
life-care monitoring framework: WarnRed hardware
and software design. Turk. J. Electr. Eng. Comput. Sci.
23, 1040–1050. doi:10.3906/elk-1304-178.
Duclos-Gosselin, L., Rigaux-Bricmont, B., Darmon, R.Y.,
2015. How Health Managers Can Use Data Mining for
Predicting Individuals’ Risks of Contracting
Nosocomial Pneumonia. Health Mark. Q. 32, 1–13.
doi:http://www.tandfonline.com/loi/whmq20.
Dudik, J.M., Kurosu, A., Coyle, J.L., Sejdić, E., 2015. A
Comparative Analysis of DBSCAN, K-Means, and
Quadratic Variation Algorithms for Automatic
Identification of Swallows from Swallowing
Accelerometry Signals. Comput. Biol. Med. 59, 10–18.
doi:10.1016/j.compbiomed.2015.01.007.
Fanucci, L., Saponara, S., Bacchillone, T., Donati, M.,
Barba, P., Sanchez-Tato, I., Carmona, C., 2013.
Sensing devices and sensor signal processing for
remote monitoring of vital signs in CHF patients. IEEE
Trans. Instrum. Meas. 62, 553–569.
doi:10.1109/TIM.2012.2218681.
Gallego Londoño, J., Lemos Duque, J.D., Hernández
Valdivieso, A.M., 2010. DISEÑO DE UN
DISPOSITIVO PORTÁTIL E INALÁMBRICO
PARA EL MONITOREO AMBULATORIO DE LA
PRESIÓN ARTERIAL NO INVASIVA. (Spanish).
Des. PORTABLE Wirel. DEVICE Ambul. Monit.
NONINVASIVE BLOOD Press. Engl. 6, 121.
Guevara-Valdivia, M.E., Echegaray-Trelles, A.,
Hernández, J., de Jesús Cordero-Pérez, L., Valderrama
de-León, R., Santos, M.A., Huarte-Hernández, Y.,
2011. Monitoreo remoto y seguimiento del paciente con
desfibrilador automático implantable y terapia de
resincronización cardiaca. (Spanish). Remote Monit.
Follow Implant. Cardioverter Defibrillators Card.
Resynchronization Ther. Devices Engl. 81, 93–99.
Jiménez González, F.C., Vergara Villegas, O.O., Torres
Ramírez, D.E., Cruz Sánchez, V.G., Ochoa
Domínguez, H., 2014. Smart Multi-Level Tool for
Remote Patient Monitoring Based on a Wireless Sensor
Network and Mobile Augmented Reality. Sens.
14248220 14, 17212–17234.
Lamonaca, F., Polimeni, G., Barbé, K., Grimaldi, D., 2015.
Health parameters monitoring by smartphone for
quality of life improvement. Meas. J. Int. Meas.
Confed. 73, 82–94.
doi:10.1016/j.measurement.2015.04.017.
Leite, C.R.M., Sizilio, G.R.A., Neto, A.D.D., Valentim,
R.A.M., Guerreiro, A.M.G., 2011. A fuzzy model for
processing and monitoring vital signs in ICU patients.
Biomed. Eng. Online 10. doi:10.1186/1475-925X-10-
68.
Morelos Ramírez, R., Ramírez Pérez, M., Sánchez
Dorantes, G., Chavarín Rivera, C., Meléndez-Herrada,
E., 2014. El trabajador de la salud y el riesgo de
enfermedades infecciosas adquiridas. (Spanish).
Healthc. Provid. Risk Acquir. Infect. Dis. Stand.
Biosaf. Precaut. Engl. 57, 34–42.
Rajkomar, A.( 1 ), Blandford, A.( 1 ), Mayer, A.( 2 ), 2015.
Understanding safety-critical interactions with a home
medical device through Distributed Cognition. J.
Biomed. Inform. 56, 179–194.
doi:10.1016/j.jbi.2015.06.002.
Scheil-Adlung, X., 2013. Revisión de las políticas para
avanzar hacia la cobertura universal de salud en los
países de bajos ingresos: ventajas de los pisos
nacionales de protección social. Rev. Int. Segur. Soc.
66, 159–186.
Skubic, M.( 1 ), Guevara, R.D.( 1, 2 ), Rantz, M.( 3 ), 2015.
Automated health alerts using in-home sensor data for
embedded health assessment. IEEE J. Transl. Eng.
Health Med. 3. doi:10.1109/JTEHM.2015.2421499.
Tomar, D., Agarwal, S., 2013. A survey on Data Mining
approaches for Healthcare. Int. J. Bio-Sci. Bio-Technol.
5, 241–266. doi:10.14257/ijbsbt.2013.5.5.25.
Toshiba Semiconductor & Storage Products Company,
2013. Toshiba Desarrolla un Módulo Sensor de Signos
Vitales Ponible Inteligente. Bus. Wire Esp.
Tovar-Cuevas, L.M., Arrivillaga-Quintero, M., 2014.
Estado del arte de la investigación en acceso a los
servicios de salud en Colombia, 2000-2013: revisión
sistemática crítica. (Spanish). State Art Access Health
Serv. Res. Colomb. 2000-2013 Syst. Rev. Engl. 13, 11–
26.
Venkatesh, V., Bala, H., 2008. Technology acceptance
model 3 and a research agenda on interventions. Decis.
Sci. 39, 273–315.
Villar-Montini, A., 2009. Tecnología de monitoreo remoto
inalámbrico. (Spanish). Remote Wirel. Monit. Engl. 79,
75–78.
Winkley, J., Jiang, P., Jiang, W., 2012. Verity: An ambient
assisted living platform. IEEE Trans. Consum.
Electron. 58, 364–373.
doi:10.1109/TCE.2012.6227435.
Zapata Jaramillo, T., Sánchez Villavicencio, D.P., 2012.
Medición de inequidades en salud para Colombia.
Población rural - urbana.
SICOBIO - Consolidation and Analysis System of Biomedical Information
187