Integration of Social Media Platforms and Specialized Web
Resources for the Effective Use of High-tech Medical Information
Alyona Kozyreva
a
, Uliana Nazarenko
b
, Grigory Shovkoplias
c
, Artem Beresnev
d
,
Elizaveta Klevtsova
e
and Natalia Gusarova
f
ITMO University, 49 Kronverksky av., St. Petersburg, Russia
Keywords: Chatbot, Healthcare, High-tech Medical Information, Internet Community Structure, Publicly Accessible
Resource, Second Opinion, Spontaneous Crowdsourcing, Telegram Channels, VK Community.
Abstract: The widespread dissemination of information technologies and technological breakthroughs in recent years
have led to the fact that the flow of medical information falls on the patient. In order to interpret their high-
tech medical information, given the lack of doctors and mistrust in them, which is especially typical for
developing countries, people make attempts to use available Internet sources and collective network
intelligence, i.e. appeal to collective opinion. Thus, there is the problem of integration of social media
platforms and specialized web resources for the effective use of high-tech medical information. We consider
this problem in relation to the Russian-speaking segment of the Internet. We have experimentally studied the
structure of public medical Internet communities typical for Russia. We found that they are characterized by
self-organization. We have developed and launched a web resource for the effective use of high-tech medical
information, and to form the motivational component of the resource, we use the identified structure of already
existing network communities of medical focus. We use specialized chat bots as an effective means of
integrating the developed resource and network communities.
1 INTRODUCTION
The prevention and treatment of top 10 diseases
causing the most deaths worldwide, including
coronary artery disease, stroke, chronic obstructive
pulmonary disease, bronchus and lung cancers etc.
(Pietrangelo, 2019) increasingly relies on high
medical technologies. Incorrect or inadequate
(without taking into account the health status in
general) interpretation of such information by the
patient himself can cause him psychological and
social problems, not to mention the treatment of the
disease itself (Giardina, 2018). However, the demand
for such information from patients all over the world
is only increasing.
There are several reasons provoking this situation.
The first is the objectively existing shortage of
a
https://orcid.org/0000-0001-8117-0869
b
https://orcid.org/0000-0002-0083-8917
c
https://orcid.org/0000-0002-5483-716Х
d
https://orcid.org/0000-0002-4646-6856
e
https://orcid.org/0000-0001-8394-5605
f
https://orcid.org/0000-0002-1361-6037
specialists (Allyn, 2020), discrepancies between two
expert interpreters (Sawan, 2017), and distrust of
doctors, caused by the dissatisfaction of patients with
existing medical care. For example, only 15% of
Brazilians and 13% of Russians assess their medicine
positively (Romir, 2020).
In this situation, the importance of the so called
second opinion for patients increases dramatically
(Benbassat, 2019), especially, a second opinion from
a radiologist (re-analysis of diagnostic images,
namely CT, MRI, radiography, mammography, PET-
CT, etc.) or cardiologist (ECG, MRI of the heart, etc.).
In principle, such analysis can be carried out by
an expert doctor remotely, using modern
telecommunications, in combination with automated
interpretation of high-tech medical information by
means of artificial intelligence (Ahuja, 2019).
154
Kozyreva, A., Nazarenko, U., Shovkoplias, G., Beresnev, A., Klevtsova, E. and Gusarova, N.
Integration of Social Media Platforms and Specialized Web Resources for the Effective Use of High-tech Medical Information.
DOI: 10.5220/0010404501540162
In Proceedings of the 7th International Conference on Information and Communication Technologies for Ageing Well and e-Health (ICT4AWE 2021), pages 154-162
ISBN: 978-989-758-506-7
Copyright
c
2021 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
Besides, the high-tech medical information contained
in patient requests could serve as a good basis for the
formation of appropriate datasets, but for this it is
necessary to develop appropriate organizational and
motivational procedures.
Thus, there is the problem of integration of social
media platforms and specialized web resources for
the effective use of high-tech medical information,
which is especially relevant for medicine in
developing countries. In the article, this task is
considered in relation to the Russian-speaking
segment of the Internet.
Namely, we have experimentally studied the
structure of public medical Internet communities
typical for Russia. We found that they are
characterized by self-organization. We have
developed and launched a web resource for the
effective use of high-tech medical information, and to
form the motivational component of the resource, we
use the identified structure of already existing
network communities of medical focus. We use
specialized chat bots as an effective means of
integrating the developed resource and network
communities.
2 BACKGROUND AND RELATED
WORKS
Internet services, on which experienced doctors
provide the patient with their reports on his high-tech
medical information, are widely represented in
medically developed countries, for example in USA
Second Opinions (from 29$), 2nd.MD ($3,000); in
India SeekMed (video communication with a
specialist is provided). These services are websites or
applications with a strictly defined business model,
which implies the legitimacy and security of the
transmitted information. Most of these services are
expensive and therefore not available to everyone.
Similar services also exist in Russia, for example:
Cardio-online; National Teleradiological Network.
But, all of them are not free and have a very narrow
specialization.
In addition to specialized Internet-services, there
is also the possibility of interpreting high-tech
medical information through social networks, like
Facebook. In Russia and in most developing countries
such as Iran, Malaysia etc., the Telegram messenger
is much more popular. A large share of the Russian-
speaking segment of the Internet is also occupied by
the social network VKontakte. In these social
networks, communities related to medicine,
cardiology, cardiac surgery, radiology, and neurology
are popular. But, although these communities are
usually administered by professional doctors, it is
impossible to guarantee the adequacy of the
interpretation of medical information here.
The issues of obtaining and interpreting medical
information in social media are studied in the
literature mainly in the aspect of crowdsourcing
(Wang, 2020; Kalantarian, 2019; Tucker, 2019).
(McCoy, 2014) defines crowdsourcing to outsource a
task to a group or community of people. (Tucker,
2019) concerns crowdsourcing activity as online
collaboration systems.
Many studies suggest crowdsourcing to perform
only separate, well-structured tasks - for example, for
pre-clinical research (Tucker, 2019), for formatting
incoming information, for improving the quality of the
extracted facts (Kalantarian, 2019). To process
information at a higher level by means of
crowdsourcing, it is proposed to involve specialists.
For example, in (Yoshida, 2016) hundreds of scientists
were recruited first to generate, and then to assess
competing health research ideas using a pre-defined set
of priority-setting criteria. At the same time, there are
examples of using crowdsourcing in artificial
intelligence projects, most often related to annotation
of medical data (Wang, 2020).
Insufficient attention is paid to the composition
and structure of the interaction of crowdsourcing
participants. As noted in the review (Créquit, 2018),
сrowd workers’ characteristics and crowdsourcing
logistics are poorly reported in the reviewed articles.
Crowd workers’ characteristics are frequently
missing: even age and gender are not reported for
about 60% of the studies.
Among the motivating factors for contribution or
collaboration in medical crowdsourcing, various
researchers distinguish recognition, curiosity, intrinsic
satisfaction, or, in some situations, financial incentives
(McCartney, 2013; Go, 2015; Chiauzzi, 2015]. The
(WHO, 2018) recommendations for underdeveloped
countries suggest such an unexpected mechanism for
motivating crowdsourcing, as the organization of
challenge contests for health. However, in general this
aspect of crowdsourcing remains outside the attention of
researchers: according to the review (Créquit, 2018), of
202 studies motivations of crowd workers were
recorded for 5 only.
The analysis performed allows us to draw the
following conclusions.
The use of communities in social networks is
convenient and accessible to all segments of the
population, but at the same time it does not guarantee
the legitimacy and reliability of the interpretation
Integration of Social Media Platforms and Specialized Web Resources for the Effective Use of High-tech Medical Information
155
obtained. In such communities, there is "spontaneous
crowdsourcing" supported by self-organization, but
its use as a motivation mechanism has not been
studied. On the other hand, existing specialized
medical services have a high level of legitimacy and
reliability, but most of them are expensive for users
and do not provide for the use of incoming
information as a source for the formation of datasets
for use in artificial intelligence.
Therefore, the authors of this article set
themselves the following tasks:
1. Explore the structure of organization and self-
organization of participants in communities on social
media platforms aimed at effective use of high-tech
medical information.
2. Conduct a problem-oriented analysis of the
needs and motivations of the participants in these
communities.
3. Based on the analysis, develop a solution that
provides integration of social media platforms and
specialized web resources for the effective use of
high-tech medical information.
3 METHODS AND MATERIALS
The paper analyzes the largest and most popular
medical communities of the VKontakte (VK) social
network on cardiology and the channels of the
Telegram messenger on cardiology and radiology -
More than Holter Monitoring (https://vk.com/holter),
Medic: ECG (https://vk.com/medic_ecg), ECG
electrocardiography (https://t.me/medecg), THE
SYNAPSUS Cardiology (https://t.me/cardiologlove),
Glowing Radiologist (https://t.me/radiologyMMA)
and Radiology Chat (https://t.me/radiologyMMAchat).
The detailed information about the community is
given in the Appendix.
The main audience of the communities was
determined after additional analysis of each
community and detailed review of the participants. A
detailed analysis of the content published in the
communities showed that all communities have an
administrator, who is a doctor, and community
members can be roughly divided into three groups -
patients, students, and professional doctors.
In our community research, we used only publicly
available information. We used the Popsters
analytics service and R programming language. We
used the free developed environment RStudio, as well
as the vkR package, which provides access to the VK
API. To visualize the “friendship network” and build
an interactive graph, we used the tkrplot library.
For each VK community, we calculated 22
variables characterizing community users (Variables
№1-9: Active Users, Population, Clear population,
Members, Share Active Members, Connected Users,
Connected Users Share, Isolates, Isolates Share), the
“friendship networks” (Variables №10-19: Edges,
Connected Components, Vertex Giant Component,
Density, Modularity (from 0 to 1), Clusters, Mean
Geodesics, Diameter, Mean Degree) and the
structural indicators of community users (Variables
№19-22: Female Share, Writer Share, Liker Share,
Passive Share). The detailed description of the
variables is presented in Appendix.
We used the k-means (MacQueen, 1967) and the
LDA (Blei, 2003) algorithms to identify the
community keywords.
To analyze messages in Telegram channels, we
wrote a data grabber in the Python language using the
telethon library. We used the analytics service and the
grabber to see the total number of posts of a certain
text length and certain content.
To view the dynamics of the visibility of the site
Cardio-online, we used the resource Be1.ru.
4 RESULTS
Statistics on the number of subscribers to
Communities №1 and 2 hosted on the VK platform
are shown in Figure 1. It shows that the number of
subscribers is growing in both communities for the
last month.
Figure 1: Statistics on the number of subscribers of
Community №1 and №2.
Figure 2: Community №1 and №2 statistics on the total
number of views for all publications.
ICT4AWE 2021 - 7th International Conference on Information and Communication Technologies for Ageing Well and e-Health
156
Figure 2 shows Community №1 and №2 statistics
for the total number of views for all posts. We can see
a slight increase in each community in the total
number of views for all publications. The growth has
been most noticeable since the beginning of 2020,
especially for Сommunity №2.
Similar data for the communities
№3 and №4
hosted on the Telegram platform are shown in Figure
3 showing active growth in subscribers. Figures 4, 5
display the total number of views in Telegram
channels for all publications on different dates in the
analyzed period. The Figures illustrate a significant
increase of views in the period from 2019 to 2020 for
all Telegram channels. The most pronounced growth
has been observed since the beginning of 2020.
Figure 3: Statistics on the number of subscribers of
Communities №3 and №4.
Figure 4: Statistics on the total number of views for all
publications of Communities №3 and №4.
Figure 5: Statistics on the total number of views for all
publications of Communities №5 and №6.
Figure 6: Dynamics of website «Almazov National Medical
Research Centre of the Ministry of Health of the Russian
Federation» visibility for 12 months.
For comparison, Figure 6 illustrates dynamics of
website «Almazov National Medical Research Centre
of the Ministry of Health of the Russian Federation»
visibility. The orange line shows the number of
requests and the blue line shows the number of
effective impressions. We can see that the number of
requests (the frequency of entering keywords from
site visibility in Yandex search per month, according
to Wordstat) for the last month is 179. The number of
site snippets shown to users in Yandex search results,
according to the site visibility data is 1,816. The
number of requests to the site decreases, the activity
on the site is low. A summary of the obtained user
indicators in Communities №1 and №2 under
consideration is presented in Table 1.
Table 1: The indicators of community users.
Variable
Com.
№1
Com.
№2
Units
1
Active
Users
1,213 1,199 Users
2 Population 19,030 36,869 Users
3
Clear
Population
18,392 35,936 Users
4 Members 18,228 35,936 Users
5
Share
Active
Members
7% 3%
% of
group
members
6
Connected
Users
11,225 14,260 Users
7
Connected
Users Share
63% 40%
% of
group
members
8 Isolates 7,167 21,676 Users
9
Isolates
Share
39% 60%
% of
group
members
Based on the processed data, it can be concluded
that users showing any activity on the pages of
Community №1 and Community №2 constitute a
small part of all users of the group, only 7% and 3%.
At the same time, users who are on each other's
friends list make up more than half of all members of
Community №1 (63%) and less than half of all
members of Community2 (40%). There are also
quite a few isolates in the communities.
A summary of the obtained structural indicators
of the “friendship network” is presented in Table 2.
Integration of Social Media Platforms and Specialized Web Resources for the Effective Use of High-tech Medical Information
157
Table 2: Structural indicators of the “friendship network”.
Variable Com. №1 Com. №2
10 Edges 33,993
33,682
11
Connected
Components
228
187
12
Vertex Giant
Component
95%
96%
13 Density 0.0002
0.0005
14 Modularity 0.0216
0.0295
15 Clusters 264
234
16
Mean
Geodesics
6.6
7.1
17 Diameter 18
26
18 Mean Degree 6.18 4.72
The general structure of the of the network of
active users of Communities №1 and №2 is shown in
Figure 7. The vertices represent users, next to the
black circle, the user ID is indicated with blue
numbers. An arc link is established between those
group members who have each other in the friend list.
Figure 7: General structure of the network of active users of
Communities №1 and №2.
Figure 8: Fragments of the general structure of the network
of active users of Communities №1 and №2, where isolates
are visible.
Figure 9: General structure of the network of active users,
where isolates are visible.
Figure 10: “Friendship Network” Components.
From the constructed graphs, we can see in more
detail the structure of communities. From Figures 8-
10, it is obvious that not all participants in
communities have close ties - many participants are
not connected with each other practically at all. There
are many subgraphs in networks – clusters (Figure 6).
A summary of the community users’ structural
indicators is presented in Table 3 (Community names
have been translated from Russian). For a better
understanding of the specificity of these
communities’ profile and users’ interests, keywords
were highlighted (Table 4). Keywords have been
translated from the native language.
Table 3: Composition of community users.
Variable
Com.
№1
Com.
№2
19
Female
Share
73.1% 70.2%
20
Writer
Share
0.4% 0.1%
21 Liker Share 6.2% 3.2%
22
Passive
Share
95.4% 96.9%
Table 4: Keywords of Communities №1 and №2.
Ke
y
words of Com. №1 Ke
y
words of Com. №2
monitoring
Holter monitoring
Halter cost
analysis
Holter examination
Halter at home
ECG
patient
risk
disease
help
year
heart
female
tests
medic ECG
subscribe
answer
follow the news
participate in public life
get to know ECG
vascular
cardiologist
artery
stenosis
coronary
literature
aorta
From Table 4, we can see that Community1
aims to provide high-tech health information
assessment and advice on Holter monitoring, and
ICT4AWE 2021 - 7th International Conference on Information and Communication Technologies for Ageing Well and e-Health
158
Community №2 aims to educate and deepen
knowledge in cardiology.
Figures 11, 12 show the total number of posts for
Communities №3-6.
Left to Right Values: small (less
than 160 symbols), average (from 160 to 1000) and
large (more than 1000) length.
Figures 13, 14 display the
total number
of
publications with a certain content. Left to Right
Values: text, photos, videos, links.
Figure 11: Statistics of the total number of posts in
Communities №3 and №4 of publications with a certain text
length.
Figure 12: Statistics of the total number of posts in
Communities №5 and №6 of publications with a certain text
length.
Figure 13: Statistics of the of publications with a certain
content in Communities №3 and №4 of publications with a
certain text length.
Figure 14: Statistics of the of publications with a certain
content in Communities №5 and №6 of publications with a
certain text length.
5 DISCUSSION
Since the subject of the medical Telegram channels
and VK communities under consideration is rather
narrow and specific, the number of subscribers is not
large. However, from the Figures 1, 3 we can see that
the number of subscribers is growing significantly,
especially from the second half of 2020. This is
mostly pronounced for the Telegram channels.
Figures 2, 4, 5 illustrate the increase of the total
number of views for all publications. This indicates
an increase in the level of people's interest in these
medical topics.
The obtained indicators of users of Community
1 (Table 1) indicate a small percentage of active users
(7%) and many isolates (7,167). From the structural
indicators of the “friendship network” (Table 2) and
the general network structure, we can see that not all
users are linked with each other. It shows the diversity
of the audience, which suggests an individual
approach to every person. From the obtained
structural indicators of users of Community № 1
(Table 3), it can be seen that a very small part of the
group creates content and a huge part of the group
does not show any activity. This allows us to say that
most of the community members are not aimed at
active communication among most participants.
Having additionally analyzed the participants and
their actions in the communities, we made sure that
users enter these communities mainly to receive an
assessment of high-tech medical information (Table
4), which is provided by a limited circle of people
(doctors). At the same time, medical students and
doctors join these communities to gain experience
and deepen their knowledge (Figure 10).
Table 4 shows the thematic profile of
communities, their interests and needs. In
combination with the analysis of the group structure
(Figures 7, 9), this once again confirms that people
have little horizontal connection with each other.
Figure 10 shows the most common network
clusters are formed by a group of students from one
medical university and a group of medical workers.
However, there are several leaders with great activity
they are often contacted, this can be seen from the
variables and pictures.
Figures 11, 12 show that most publications in
Communities 3 and №4 have medium and large
length. This is due to the patients' description of their
symptoms and the first medical opinion, as well as
detailed answers from doctors and medical students.
A more detailed analysis of the posts shows that
people write their medical data directly. This means
that the need for additional interpretation of their
Integration of Social Media Platforms and Specialized Web Resources for the Effective Use of High-tech Medical Information
159
medical data is so great that it often outweighs the
requirements for privacy and legitimating. This
circumstance should be taken into account when
building a web resource.
Figures 13, 14 show that users send their data in
various graphical formats to get an assessment of
high-tech medical information. At the same time, the
radiology Community №5 is characterized by short
messages. This is because videos and links to useful
materials are more often published on the main
channel. In Community 6 there are more long
messages.
Summing up the results of the analysis, we can
state that in order to obtain an effective assessment
of high-tech medical information, residents of Russia
address primarily in public Internet spaces. There
emerge and become popular communities working in
the mode of "spontaneous crowdsourcing", and they
have a pronounced self-organization. Users are
divided into groups - patients, students and doctors -
with their interests that do not contradict each other
and are fully satisfied there. This allows such
communities to provide the necessary informational
services to their participiants for free.
However, the medical information discussed in
such communities is transmitted and disseminated in
an open manner, without regard to privacy and
legitimating requirements. In addition, despite the
obvious value of this information, it is practically not
used as a source for the formation of datasets for
subsequent use in electronic medicine through
artificial intelligence.
It points to the relevance of creating a publicly
available solution for evaluating high-tech
information, combining the accessibility and ease of
communication inherent in social networks,
providing a legitimate second opinion for patients
and at the same time allowing the use of the
information provided to form datasets. The authors
see such a solution as a combination of a specialized
network resource and a set of chat bots that integrate
it with network platforms. The results presented
above made it possible to formulate the requirements
for the proposed solution:
1. VK communities and Telegram channels are
not legitimate enough. The chatbot solves this
problem because it integrates with a high-tech
service.
2. People need to be confident that a service can
be trusted. Therefore, when designing a chat bot, it is
necessary to ensure the concealment of personal data.
3. Users need the ability to provide data in various
graphic formats for convenience.
4. The opportunity to exchange information in a
simple dialogue form is necessary. It is necessary to
design different dialog scenarios, different levels of
description for different clusters of users, considering
their peculiarities. Therefore, it is necessary to foresee
several chatbots targeted at specific user groups.
6 PROPOSED SOLUTION
A use-case diagram of the developed specialized web
resource is shown in Figure 15. The resource works
on the principle of teleradiology. Its datasets are
formed thanks to the attending physicians and their
patients, who are ready to provide their anonymized
medical information for forming datasets. The main
actors of this resource are the patient, the expert
doctor, the trainee, and the owner of the resource:
each of them gets a certain benefit from the use. It
also has a high level of privacy and security: this
inspires trust among users. The resource allows using
high-tech tools: it served as the motivation for
creating a bot. The chat-bot is served as an integration
tool. Through its implementation in social networks,
it engages people in using of a high-tech tool and
brings them together, thereby meeting their needs.
The chatbot is easy to use and does not require any
special knowledge.
Patient
TraneeExpertDoctor
Po rtal Ou ner
Medi c i ne User
Cre at in gr eport
Gettingarating
Editinganamnesis
Medicineuser
registration
Interactingwith
potf olio
Interactingwithreports
sets
Patientregistration
Creatin garequest
Interrac tingwithreport
Viewingreport
<<ex te nd>>
<<ex te nd>>
<<ex te nd>>
Editin gtheyownreport
<<ex te nd>>
Viewingportfolio
<<ex te nd>>
Clearingportfol io
<<ex te nd>>
Uploaddata
<<ex te nd>>
Filteringreports
Creatingsetofreports
Viewingseto freports
<<ex te nd>>
<<ex te nd>>
Downlodingdataset
Discussionoft hereport
<<ex te nd>>
Figure 15: Use Case diagram of the specialized web
resource.
Figure 16 shows a use case diagram describing the
integration of the specialized web resources with
social media platforms. The interaction of social
platforms users with the service is realized through
the mechanism of chat-bots. Our chat-bots
(TelegramBot-MCP and VKBot-MCP) are developed
individually for each platform, since the mechanism
of interaction between a chat-bot and a social
platform depends on the API offered by the platform.
The business logic of interaction between the user
and the bot is implemented on the side of the bot's
ICT4AWE 2021 - 7th International Conference on Information and Communication Technologies for Ageing Well and e-Health
160
software module. Social platforms provide users with
a conversational interface to a chatbot in a familiar
environment. Through the chatbot, the user can send
a request for a second opinion to the Platform and
receive answer as soon as it is ready.
The interaction of chat bots and the main part of
the platform is carried out through a special API,
which allows to scale the system and transparently
connect other social platforms. Currently the
specialized web resource for the effective use of high-
tech medical information «ITMO University's
Medical Crowdsourcing Platform» (ITMO_MCP)
has been launched on the platform of the ITMO
University (St. Petersburg, Russia).
Figure 16: Use Case diagram.
7 CONCLUSION
The article discusses the possibility of creating a
public resource for the effective use of high-tech
medical information. The tasks set by the authors of
the article have been successfully fulfilled. We have
investigated the structure of organization and self-
organization of community members in social
networks aimed at obtaining and effectively using
high-tech medical information. We have also
analyzed public medical Internet spaces (VK medical
communities in cardiology and Telegram medical
channels in cardiology and radiology), with the help
of which the motivations and needs of their
participants were identified.
Based on the results of the analysis of medical
communities, we proposed a solution that provides
the integration of social media platforms and
specialized web resources for the effective use of
high-tech medical information.
ACKNOWLEDGEMENTS
This work was supported Grant of the President of the
Russian Federation for state support of young Russian
scientists - candidates of science, MK-5723.2021.1.6.
REFERENCES
Ahuja A.S. The impact of artificial intelligence in medicine
on the future role of the physician. PeerJ. 2019; 7:
e7702. 10.7717/peerj.7702 https://www.ncbi.nlm.nih.
gov/pmc/articles/PMC6779111/
Allyn J. International Radiology Societies Tackle
Radiologist Shortage. Feb. 20, 2020.
https://www.rsna.org/en/news/2020/February/Internati
onal-Radiology-Societies-And-Shortage
Benbassat J. Obtaining a second opinion is a neglected
source of health care inequalities. 2019.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6335
699/
Blei D.M., Ng A.Y., Jordan M.I. Latent Dirichlet allocation
// Journal of Machine Learning Research. 2003. V. 3,
№ 4—5. pp. 993—1022.
Caruso Brown A.E., Arthur J.D., Mutrie L.H. and Lantos
J.D. Seeking a Second Opinion on Social Media.
Pediatrics. November 2019, 144 (5) e20190817;
DOI:https://doi.org/10.1542/peds.2019-0817
Chiauzzi E., Eichler G., Wicks P. Crowdsourcing
Advancements in Health Care Research. December
2016. DOI: 10.1016/B978-0-12-802115-6.00016-1
Clinical radiology UK workforce census report 2018.
https://www.rcr.ac.uk/publication/clinical-radiology-uk-
workforce-census-report-2018 Conceptual clarification
// Soc.Networks. 1978/79. V. 1. P. 215-239.
Créquit P., Mansouri G., Benchoufi M., Vivot A., Ravaud
P. Mapping of crowdsourcing in health: Systematic
review. Journal of Medical Internet Research, 2018,
/10.2196/jmir.9330/. hal-02268395
Freeman L. C. The Development of Social Network
Analysis: A Study in the Sociology of Science.
Empirical Press, Vancouver, BC (2004)
Yoshida S., Rudan I., and Cousens S. Setting health
research priorities using the CHNRI method: VI.
Quantitative properties of human collective opinion. J
Glob Health. 2016 Jun; 6(1): 010503.
Giardina T.D, Baldwin J., Nystrom D.T., Sittig D.F., Singh
H. Patient perceptions of receiving test results via
online portals: a mixed-methods study. J Am Med
Inform Assoc. 2018 Apr 1;25(4):440-446. doi:
10.1093/jamia/ocx140.
Go H.M.V., Caro J.D.L., Pabico J.C.B., Tee M.L.
Crowdsourcing for healthcare resource allocation. 2015
6th International Conference on Information,
Intelligence, Systems and Applications (IISA), Corfu,
2015, pp. 1-6, doi: 10.1109/IISA.2015.7388091.
Kalantarian H, Jedoui K, Washington P, Tariq Q, Dunlap
K, Schwartz J, et al. Labeling images with facial
Integration of Social Media Platforms and Specialized Web Resources for the Effective Use of High-tech Medical Information
161
emotion and the potential for pediatric healthcare. Artif
Intell Med. 2019;98:77–86.
Krinsky J., Crossley N. (2014) Social Movements and
Social Networks: Introduction, Social Movement
Studies, 13:1, 1-21, DOI: 10.1080/14742837.2013.
862787
McCartney P. Crowdsourcing in Healthcare. MCN, The
American Journal of Maternal/Child Nursing:
November/December 2013. V. 38. Is. 6. P. 392. doi:
10.1097/NMC.0b013e3182a41571
McCoy A, Wright A, Rogith D, Fathiamini S, Ottenbacher
AJ, Sittig D. Development of a clinician reputation
metric to identify appropriate problem-medication pairs
in a crowdsourced knowledge base. J Biomed Inform.
2014; 48:66–72.
MacQueen J. (1967). Some methods for classification and
analysis of multivariate observations. In Proc. 5th
Berkeley Symp. on Math. Statistics and Probability, pp.
281—297.
Orenstein B.W. Virtually Indispensable. Radiology Today.
Vol. 19 No. 12 P. 18
Pietrangelo A. Weatherspoon D., Ph.D., R.N., CRNA. The
Top 10 Deadliest Diseases. July 10, 2019.
https://www.healthline.com/health/top-10-deadliest-
diseases
Romir/ GlobalNR International Analytics: Citizens
Assessed the General Condition of National Health
Systems (2020).
https://romir.ru/studies/mejdunarodnaya-analitika-
romirglobalnr-grajdane-ocenili-obshchee-sostoyanie-
nacionalnyh-sistem-zdravoohraneniya
Sawan P, Rebeiz K, Schoder H, Batlevi C, Moskowitz A,
Ulaner GA, Dunphy M, Mannelli L. Specialized
second-opinion radiology review of PET/CT
examinations for patients with diffuse large B-cell
lymphoma impacts patient care and management.
Medicine (Baltimore). 2017; 96:e9411.
Tucker J.D., Day S., Tang W., Bayus B. Peer J. 2019; 7:
e6762. Crowdsourcing in medical research: concepts
and applications. Published online 2019 Apr 12. doi:
10.7717/peerj.6762
Wang Ch., Han L., Stein G. et al. Crowdsourcing in health
and medical research: a systematic review. Infect Dis
Poverty 9, 8 (2020). https://doi.org/10.1186/s40249-
020-0622-9
Weinfurtner RJ, Niell B, Mekhail Y, Aguila E, Kamat L.
Specialized second opinion interpretations of breast
imaging: impact on additional workup and
management. Clin Breast Cancer. 2018;18:e1031–6
WHO. World Health Organization. Crowdsourcing
in health and health research: a practical
guide TDR/STRA/18.4. 2018. https://apps.who.int/iris/
bitstream/handle/10665/273039/TDR-STRA-18.4-
eng.pdf
ICT4AWE 2021 - 7th International Conference on Information and Communication Technologies for Ageing Well and e-Health
162