See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/323120094

Adopting Cognitive Computing Solutions in Healthcare

Article in Journal of E-Learning and Knowledge Society · February 2018

DOI: 10.20368/1971-8829/1451

CITATIONS READS

6 2,352

2 authors:

Paolo Maresca Mauro Coccoli

University of Naples Federico II Università degli Studi di Genova
183 PUBLICATIONS 659 CITATIONS 96 PUBLICATIONS 752 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

programming cognitive computing View project

smarter university View project

All content following this page was uploaded by Paolo Maresca on 17 July 2018.

The user has requested enhancement of the downloaded file.

 Je-LKS
PEER REVIEWED PAPERS
NEW TRENDS, CHALLENGES AND PERSPECTIVES ON HEALTHCARE COGNITIVE
COMPUTING: FROM INFORMATION EXTRACTION TO HEALTHCARE ANALYTICS
Vol. 14, n.1, 2018
ISSN: 1826-6223 | eISSN: 1971-8829
Journal of e-Learning and Knowledge Society
The Italian e-Learning Association Journal

ADOPTING COGNITIVE
COMPUTING SOLUTIONS IN
HEALTHCARE

Mauro Coccoli1
Paolo Maresca2
1
University of Genoa, Italy - mauro.coccoli@unige.it
2
Federico II University of Naples, Italy - paolo.maresca@unina.it

Keywords: Cognitive Computing, E-Health, Healthcare Big Data, Artificial Intelligence

This paper discusses possible motivations to adopt cognitive computing-

based solutions in the field of healthcare and surveys some recent
experiences. From a very practical point of view, the use of cognitive
computing techniques can provide machines with human-like reasoning
capabilities, thus allowing them to face heavy uncertainties and to cope with
problems whose solution may require computing intensive tasks. Moreover,
empowered by reliable networking infrastructures and cloud environments,
cognitive computing enables effective machine-learning techniques, resulting
in the ability to find solutions on the basis of past experience, taking
advantage from both errors and successful findings. Owing to these special
features, it is perceptible that healthcare can greatly benefit from such a
powerful technology. In fact, clinical diagnoses are frequently based on
statistics and significant research advancements were accomplished through
the recursive analysis of huge quantity of unstructured data such as in the
for citations:
Coccoli M., Maresca P. (2018), Adopting Cognitive Computing Solutions in Healthcare, Journal of
e-Learning and Knowledge Society, v.14, n.1, 57-69. ISSN: 1826-6223, e-ISSN:1971-8829
DOI: 10.20368/1971-8829/1451
 Je-LKS PEER REVIEWED PAPERS - NEW TRENDS, CHALLENGES AND PERSPECTIVES ON HEALTHCARE COGNITIVE COMPUTING:
FROM INFORMATION EXTRACTION TO HEALTHCARE ANALYTICS
Vol. 14, n. 1, January 2018

case of X-ray images or computerized axial tomography scans. As another example, let us consider
the problem of DNA sequence classification with the uncountable combinations that derive from such
a complex structure.

1 Introduction
Pushed by the fast and unstoppable innovation in Information and Commu-
nication Technology (ICT), we are experiencing daily evolutions in applications
and services that we commonly use. This is due to both the wide availability of
computational resources, and the large amount of data exchanged at high speed
between a variety of heterogeneous devices and systems. Such an overwhel-
ming progress impacts on a vast class of applications and gives way to the raise
of a new wave of advanced services. In particular, the technological framework
enabling this new wave can be depicted by the following keywords: Cloud,
Semantic Web, Big Data, and Cognitive Computing. In this paper, we focus
on cognitive computing that, in turn, relies on the cloud infrastructure and
profitably implements semantic Web techniques to analyse big data, making
them meaningful and transforming them in valuable information. In fact, cogni-
tive computing systems owe their success to the capability of fast-processing
huge amounts of data through the novel and sophisticated machine learning
algorithms they are based on. In the present situation, healthcare is one of the
pioneer fields in which cognitive computing is being applied extensively. We
will present an overview of challenging research topics and we will showcase
some of the results already achieved.
The remainder of the paper is organized as follows. In Section 2 a general
introduction to cognitive computing is presented. Section 3 covers issues on
programming such systems, while Section 4 takes into account recent cognitive
computing applications in the specific field of healthcare. Conclusions follow
in Section 5.

2 Cognitive Computing
From a practical point of view, we regard cognitive computing as the revamp
of precedent well-funded theories that hardly found practical applications at
the time of their formulation, due to the lack of computing power. This is the
case, for example, of Artificial Intelligence (AI) and neural networks, which are
characterized by high complexity and by the need of executing huge numbers
of parallel operations in strict time frames. Traditional AI techniques rely on
the model of expert systems and exploit statistics and complex mathematical
model, thus require that a significant amount of operations per second is exe-
cuted on large dimension data sets for their training. In this respect, cognitive

58
 Mauro Coccoli, Paolo Maresca - Adopting Cognitive Computing Solutions in Healthcare

computing can be considered as the revenge of AI since, nowadays, we have

computing architectures suited to face large dimensional problems on large
data sets. This continuous number crunching allows performing analytics and
deriving new knowledge, which results in the ability to anticipate solutions in
a heterogeneous class of problems. As an example, let us consider the Inter-
net search engines. They employ such a kind of AI, to the aim of giving back
information that is relevant to the users, based both on individuals’ data and
on the application of patterns. This involves issues such as, e.g., language
contextualization, classification, clustering, entity extraction, and more. As
another example, let us consider the popular sites of electronic commerce and
the recommendation systems they adopt. When one is seeking an item, or im-
mediately after a purchase, the algorithm driving the recommendation system
exploits cognitive computing techniques to provide focused shopping advices
related to the products one is browsing, has marked as favourite, has added to
his/her wish list, also relying on users’ preferences and on their purchase history
(Pazzani & Billsus, 2007).
To conclude, we observe that cognitive computing systems should be re-
garded as a “more human” AI. In fact, they mimic human reasoning methodo-
logies, showing special abilities in dealing with uncertainties and in solving
problems that typically entail computationally heavy processes. Moreover, they
expose learning capabilities, thus their knowledge base is continuously-growing
and, accordingly, their reasoning ability is continuously-enhancing. Besides,
cognitive computing plays an important role in improving both man-to-machine
and machine-to-machine communications, and in fostering the development of
new human-computer interactions models, based on the Natural User Interface
(NUI) paradigm such as, e.g., conversional systems (Nishida et al., 2014).
Moreover, through the effective and reliable simulation of the reasoning pro-
cesses, machines can be trained to learn from experts’ behavior in approaching
problems and from their problem-solving techniques, to become, in turn, able
to train newbies and to teach humans new concepts and/or new procedures.
Such intelligent systems could be used, e.g., in training and customization, or
other activities that requires data analysis in order to improve both processes
and products (Earley, 2015).
Furthermore, cognitive computing is thought to be the corner-stone for the
future enlargement of the Internet of Things (IoT) scope (Zhang et al., 2012)
and, consequently, on the relevant interconnect technologies (Orii et al., 2016).
In fact, the expected near-future scenario is forecasting people and things to
interact naturally (Coccoli & Torre, 2014), striving spoken language, while
producing and consuming data performing their actions. Therefore, we need
advanced analytics to gather information and extract data and vice versa, to
the aim of realizing novel somehow-intelligent systems that are able to react

59
 Je-LKS PEER REVIEWED PAPERS - NEW TRENDS, CHALLENGES AND PERSPECTIVES ON HEALTHCARE COGNITIVE COMPUTING:
FROM INFORMATION EXTRACTION TO HEALTHCARE ANALYTICS
Vol. 14, n. 1, January 2018

in real time to unpredictable external stimuli with unknown origins. In this

respect, we observe that the more important benefits that can derive from co-
gnitive computing will not reside in the cognitive systems themselves, but in
the coupling of cognitive systems with the surrounding environment. Then a
novel era for engineering will rise, in which design will be driven by desired
behaviours rather than by design constraints: what will make the machines
rather than how they will be made (Holtel, 2014).
In conclusion, owing to the prospect resources and capabilities available
within a working environment based on the use of cognitive computing, we
can envisage the inception of a new generation of semi-autonomous systems
committed to improve the quality of life, addressing critical societal issues
(Mohideen & Evans, 2015), helping people in facing a variety of small compli-
cations as well as awkward jobs. Such new systems will be mainly based on the
imitation of human attitudes and reasoning (Mohda et al., 2011). Systems based
on cognitive computing can successfully accomplish difficult tasks such as,
e.g., classification, natural language processing, and data mining, thus are able
to perform advanced activities such as, e.g., sentiment analysis, relationships
extraction from unstructured corpus, image recognition, speech-to-text and
text-to-speech conversions. Another prospected advantage is the growing con-
fidence in humans that machines can provide reliable answers, i.e., within an
acceptable range of trust, in delicate areas, such as, e.g., medicine, education
or economics (Coccoli et al., 2016).

3 Programming Cognitive Computing Systems

As already mentioned, cognitive computing systems owe their powerful
characteristics to the recent enhancement of traditional AI pushed by the avai-
lability of new technologies, to the rapid development of new machine learning
techniques, and to the large availability of data coming from a heterogeneous
set of sources and devices. These reflect in the possibility of implementing
effective model-based reasoning capabilities, which make cognitive systems
able to perform complex tasks such as, e.g., discovery, reasoning, and multi-
modal understanding in a variety of domains (Banavar, 2015) such as, e.g.,
healthcare, insurance, and education (Coccoli et al., 2017). As a consequence,
cognitive computing and the relevant technology are going to play a key role
in engineering systems of the future (Noor, 2015).

3.1 The Cognitive Computing Consortium

Given such premises and the forecast on the future cognitive computing
development and achievements, it is necessary to set up an open working en-

60
 Mauro Coccoli, Paolo Maresca - Adopting Cognitive Computing Solutions in Healthcare

vironment where researchers and IT professionals can find non-proprietary de-

finitions that can be used as benchmark. To this aim, a cross-disciplinary group
of experts from industry, academia and the analysts’ community, founded the
Cognitive Computing Consortium1. Constituents come from a mix of research
centres, companies and institutions such as Synthesis and NextEra Research
(founders) with Pivotal, Basis Technology, HP, IBM, BA Insight, Customer-
Matrix, SAS, Interactions, Bebaio, Microsoft, and universities such as, UCSF
and the Babson College. It is worthwhile noticing that most of the sponsors are
companies involved in big data analysis.

3.2 Programming Cognitive Computing Systems

To the aim of spreading the adoption of cognitive computing, specific
platforms and tools exist, enabling programmers to develop suited, effective
and reliable systems tailored to solve their problems. In recent years, many big
players in the IT scenario delivered their own cognitive computing kit and this
is driving both market and technology in the direction of making such systems
affordable and widely available. Among these, in the following we mention the
most significant examples that are, in alphabetical order:

(i) Enterprise Cognitive Systems by Enterra. Formerly known as Cognitive

Reasoning Platform (CRP), the enterprise cognitive systems framework is de-
fined by its developers as “an artificial intelligence platform that allows orga-
nizations to capitalize on the power and potential of big data through advanced
analytics and actionable insights that fundamentally inform organizations about
the business, customers, and value chains in which they operate”. They also
claim it can easily combine “[..] the efficiency and accuracy of computational
computing with the analytic and predictive abilities of human reasoning. [..]
can receive extraordinary volumes of data from any source, structured and un-
structured, understand the nature of the data, learn from the relationships and
connections it discovers, make decisions, and take actions to achieve defined
outcomes”;

(ii) Deep Learning Technology Center. It is the structure owned by Micro-

soft where to work with the Cognitive Toolkit, which was formerly known as
the Computational Networks Toolkit (CNTK), made available in open source
for anyone to use in their own work on GitHub. It is depicted by its developers
as “A [..] commercial-grade toolkit that trains deep learning algorithms to learn
like the human brain”. This tool allows creating deep learning networks for
different activities and “[..] empowers you to harness the intelligence within
1
http://cognitivecomputingconsortium.com

61
 Je-LKS PEER REVIEWED PAPERS - NEW TRENDS, CHALLENGES AND PERSPECTIVES ON HEALTHCARE COGNITIVE COMPUTING:
FROM INFORMATION EXTRACTION TO HEALTHCARE ANALYTICS
Vol. 14, n. 1, January 2018

massive datasets through deep learning by providing uncompromised scaling,

speed and accuracy with commercial-grade quality and compatibility with the
programming languages and algorithms you already use”;

(iii) DeepMind. It is the platform offered by Google that, in 2014, acquired

the namesake UK-based AI company aimed to solving artificial intelligen-
ce problems. Their claim is “Solve intelligence. Use it to make the world a
better place”. Then, to show the effectiveness of their work, in 2015, Google
announced the creation of a specific AI that learns by itself and is able to win
video games. Indeed, they “were able to create a single program that taught
itself how to play and win at 49 completely different Atari titles, with just raw
pixels as input. And in a global first, the AlphaGo program took on the world’s
best player at Go - one of the most complex and intuitive games ever devised,
with more positions than there are atoms in the universe - and won”;

(iv) IDOL (Intelligent Data Operating Layer). It is the software layer offered
by Hewlett-Packard, whose tagline is “Unified machine learning platform for
enterprise search and big data analytics - text analytics, speech analytics, image
analytics and video analytics”. Delivered by HP, which acquired Autonomy in
2011, within their big data software platform, IDOL is offering many services
and solutions for, e.g., data analysis and IoT. They claim that the “IDOL Na-
tural Language Question Answering empowers organizations to tap into the
full potential of big data by breaking down the barriers between machines and
humans. It effectively unleashes the power of machine learning by enabling
natural language based human-centric exchanges in delivering the contextually
relevant information”;

(v) Watson, by IBM. It promises to “go beyond artificial intelligence”. It is

a technology platform using natural language processing and machine learning
to reveal insights from large amounts of unstructured data. IBM claims that
Watson “can understand all form of data, interact naturally with people, and
learn and reason, at scale. [..] you can analyse and interpret all of your data, in-
cluding unstructured text, images, audio and video [..] you can utilize machine
learning to grow the subject matter expertise in your apps and systems [..] you
can provide personalized recommendations by understanding a user’s persona-
lity, tone, and emotion [..] you can create chat bots that can engage in dialog”.

4 Cognitive Computing Applications in Healthcare

Following the digitization process of medical records occurred in the recent
years, we notice that, as other application fields, healthcare too is suffering from

62
 Mauro Coccoli, Paolo Maresca - Adopting Cognitive Computing Solutions in Healthcare

an explosion of information. On one hand, a huge amount of data reveals new

opportunities for the advancement of research and for the effective treatment of
diseases. On the other hand, such an information overload is hard to manage for
both physicians and care providers. The unique reasoning ability of cognitive
systems can perform detailed analysis and comparisons exploiting all the data
available, thus becoming an effective companion.
As researchers, we can envisage a variety of applications in which cognitive
computing can contribute to evolve healthcare but, in the current situation, its
most promising feature appears to be the ability in managing huge quantity of
information. In fact, cognitive system can easily overtake present solutions for
big data management and decision support and these class of problems is very
common in medicine, especially for prevention and diagnosis based on statistics
analysis and visual pattern recognition. Most impressive and eye-catching re-
sults have been achieved in cancer diagnosis and results in this field have been
proudly announced by the technology provider of the above-presented Deep
Mind, Google, and Watson. As already stated, recognizing and classifying ima-
ges is another peculiar ability of cognitive systems (Teo et al., 2012) and this
feature is a strategic asset in the prevention of cancer pathologies, especially
for what concerns breast cancer, lung cancer and prostate cancer (Strickland,
2013). In fact, a cognitive system that “sees” is a valuable support, relieving the
doctor from the task of analysing many hundreds of thousands of documents
about the same pathology in a little time-frame as well as for providing the
semantic interpretation of diagnostic images (Ogiela et al., 2006). However, it
is worth pointing out at this point how the application of the cognitive system
in the medical field in particular finds a series of barriers and resistances by
physicians and nurses, mainly due to the lack of basic computer skills. Never-
theless, there is evidence that where cognitive systems are used for healing, a
50% improvement in results is observed, while hospitalization costs are reduced
by half (Gatenbein, 2014). In addition, there is also a reduction in diagnostic
errors, especially in carcinogenic pathologies. This last observation is changing
the focus of operators, because in the United States the first cause of death is
due to diagnostic errors.
To clarify the situation, we propose a literature review of cognitive compu-
ting solutions and related technologies applied to the healthcare. We will not
enter in details of specific solutions and methodologies adopted, yet we will
remain at an abstraction level where the benefits are highlighted and foreseen
for the future development of novel decision support systems and autonomous
services as well, to the aim of improving individuals’ quality of life and health.
In the following, with no ambition to be exhaustive, we report some recent
works that illustrate the above-cited ability. It is worthwhile noticing that many
articles in the clinical literature refer to systems based on IBM Watson and we

63
 Je-LKS PEER REVIEWED PAPERS - NEW TRENDS, CHALLENGES AND PERSPECTIVES ON HEALTHCARE COGNITIVE COMPUTING:
FROM INFORMATION EXTRACTION TO HEALTHCARE ANALYTICS
Vol. 14, n. 1, January 2018

reserved a specific section for those.

4.1 Decision Support

Decision making and decision support can benefit from cognitive com-
puting capabilities. One example of this is reported in the paper “Temporal
Modeling in Clinical Artificial Intelligence, Decision-Making, and Cognitive
Computing: Empirical Exploration of Practical Challenges”, Bennett and Doub
(2014) describe a decision system based on Markov Decision Processes and
the use of neural networks, implementing the so-called temporal modelling
approach, that allows capturing certain aspects of human cognition. In this re-
spect, the computing system should resemble the same process, hence cognitive
computing solution re expected to improve performances of the above-cited
approach. Other methods, such as, e.g., Interactive Metric Learning (IML) are
described in the paper “IBM’s Health Analytics and Clinical Decision Support”,
by Kohn et al. (2014). Another core aspect in healthcare involving decision
support systems is related to the management of patient records. In their paper
“Cognitive computing for electronic medical records”, Devarakonda and Mehta
(2016) describe the problems due to the overload of information in Electronic
Medical Record (EMR) systems and the inability to make sense of this informa-
tion to provide the best care for their patients. They identify the solution in the
use of cognitive systems designed to perform advanced analysis on the patient
record data. These may also require the understanding of natural language
questions about the patient record content, helping physicians to automatically
identify urgent abnormalities, and provide precise causes for such abnormali-
ties. In such a cognitive computing view, the EMR is transformed in an active
entity that helps making decisions, leveraging the large amount of knowledge
within the medical sciences, drug information, and medical ontologies.

4.2 Big Data and Analytics

In the editorial “Big Data and Analytics”, by Tan et al., the authors put in
evidence how the digitization process occurred in health and in the management
of patient data, as well as the rapid adoption of health information systems have
led to the generation of huge volumes of primary and secondary data within
the health care industry, that cannot be processed and managed by traditional
data processing tools and that have to be duly managed, in order to make them
a valuable asset for both improving the therapy effectiveness and advancing
research to enhance prevention and health outcomes, also for reducing costs
(Tan et al., 2015). In this respect, they introduce health analytics as “the sy-
stematic use of health data and related business insights developed through

64
 Mauro Coccoli, Paolo Maresca - Adopting Cognitive Computing Solutions in Healthcare

applied analytical disciplines (e.g. statistical, contextual, quantitative, predic-

tive, cognitive, other models) to drive fact-based decision making for planning,
management, measurement and learning” (Cortada et al., 2012). Consequently,
health analytics require a variety of statistical techniques borrowed from mo-
delling, machine learning, data mining, to analyse current and historical facts
to make predictions about unknown events. Then, Chen et al., in their paper
“IBM Watson: how cognitive computing can be applied to big data challenges,”
extend the scope of cognitive computing to the entire life sciences field (Chen
et al., 2016), pondering on issues that life sciences researchers have to cope, de-
tailing how cognitive technologies can help finding new solutions to aggregate
big data for a better understanding of the latent information they may contain.

4.3 Watson in Healthcare

In the paper “Paging Dr. Watson: IBM’s Watson Supercomputer Now Being
Used in Healthcare”, Lee (2014) outlines how the supercomputer has moved on
to practical applications and why it was “taught” to understand the complexities
of healthcare, putting emphasis on using the term taught rather than program-
med. After a brief history of cognitive computing in history, practical uses for
Watson in are duly listed, such as, cancer research, supply chain management,
and consumer empowerment to help create better outcomes in healthcare. Si-
gnificant case studies are presented, including the work done with the Memorial
Sloan-Kettering Cancer Center (MSKCC) and with the MD Anderson Cancer
Center (University of Texas), which both experienced that the large amount of
data collected from their patients and stored in heterogeneous systems were
essentially useless due to the inability of merging the results achieved by the
oncology team with the clinical trial data. Another application of Watson in
oncology is reported in the paper “Envisioning Watson as a Rapid-Learning
System for oncology” (Malin, 2013), which also, emphasizes on the unprece-
dented reasoning ability of the cognitive computing system, using machine
learning to determine how to weigh clinical factors in patients, to the aim of
identifying the more suited treatment options and give a decision support to
physicians. In practice, Watson was trained to do this, similar to a medical
school student, which learns by observing more experienced physicians.
Besides, we observe that these successful applications are leading to future
developments involving the same technology. In fact, we highlight that the
New York Genome Centre (NYGC) and IBM are collaborating (Ratner, 2015)
to analyse genetic data to speed up the process of treatment for patients with
brain cancer (Douglass & Kearns, 2017). The IBM Watson cognitive com-
puting system will be trained to analyse genomic data from a small group of
patients with glioblastoma diagnosis, one of the most aggressive and malignant

65
 Je-LKS PEER REVIEWED PAPERS - NEW TRENDS, CHALLENGES AND PERSPECTIVES ON HEALTHCARE COGNITIVE COMPUTING:
FROM INFORMATION EXTRACTION TO HEALTHCARE ANALYTICS
Vol. 14, n. 1, January 2018

brain tumours. Its cognitive abilities will be used to analyse gene sequence
variations between normal brain tumour biopsies, medical information, and
clinical records to help physicians locate a variety of treatments and tailor the
type of cure for specific cancer. Then, applying cognitive computing power
accelerates the ability to address personalized cure for fatal diseases such as
cancer. Another interesting development filed is enhancing medicine in de-
veloping countries where 70% of new cases of brain cancer occur. In India,
there is only one oncologist for about 1,600 patients. A cognitive system such
specifically designed for Oncology (Manipal, 2017) can help with large num-
bers of patients. The Manipal Hospital in India is one of the private hospital
chains that treat 200,000 patients a year. A physician with Watson’s help takes
only 20 seconds to collect information about a patient. This is a big difference
because it allows very fast to give patients a cure. In addition to the speed we
have much more precision: mistakes in formulating a diagnosis are reduced.
Currently, IBM Watson for Oncology is used also in China, Thailand, Finland.
Important results in a different field are reported in Barrow (2017). Barrow
Neurological analysed 1,500 genes by discovering that five of them had never
been connected to SLA. Moreover, IBM actively collaborates with the New
York Collaborative Care Centre to develop a health management platform
(Douglas & Adigun, 2017, CNYCC, 2017). UNC Lineberger is another com-
prehensive cancer centre, which adopted cognitive solutions to accelerate DNA
analysis and inform personalized treatment options for patients (UNC, 2017).

Conclusions
From this overview, we can argue that cognitive computing in healthcare
is a hot and promising topic. Both academics and industry are making big
efforts to improve the performances of current systems and to propose novel
solutions based on the profitable exploitation of big data. However, we put in
evidence that most of the reported experiences are from United States where
the healthcare system is organized in a peculiar manner, which is quite different
from the majority of other Countries.
Furthermore, unfortunately, there is still a lack in infrastructural settings, the
availability of open big data, and in general the minimum requirements for the
hardware to effectively run such systems are still high, despite they can rely on
modern and sophisticated cloud-based architectures. Nevertheless, cloud com-
puting is expected to uphold its rapid growth in the very next future so that we
can forecast the wide availability of affordable services for many applications.
This will be one of the main pillars to base the diffusion of cognitive systems
on, and will ease the penetration of such a novel variety of systems that will
foster new services and will bring disruption in many settled paradigms. The

66
 Mauro Coccoli, Paolo Maresca - Adopting Cognitive Computing Solutions in Healthcare

cognitive healthcare will have a strong impact on cloud evolution. In fact, there
will be a need to create an optimized cloud for all cognitive data - a hybrid and
secure cloud. It is worth considering that in addition to the cloud, we also need
to redesign the data architecture due to the heterogeneity of medical data. This
is because in medicine 90% of data is image and 80% of medical data is not
available on the Web, also due to security and privacy issues. Finally, cognitive
health care will have a very strong impact on industry.

REFERENCES
Banavar, G.S. (2015), Watson and the era of cognitive computing, Proceedings of the
2015 IEEE International Conference on Pervasive Computing and Communications
(PerCom), St. Louis, MO, 95-95.
Barrow Neurological Institute (2017), Barrow, IBM Watson health ID new genes
involved in ALS, available at the URL: https://www.barrowneuro.org/press-releases/
barrow-ibm-watson-health-new-genes-als, last visited, Nov., 2017.
Bennett, C.C and Doub, T.W. (2014), Temporal modeling in clinical artificial
intelligence decision-making and cognitive computing: Empirical exploration of
practical challenges, in Proceedings of the 3rd SIAM Workshop on Data Mining
for Medicine and Healthcare (DMMH).
Chen, Y., Argentinis, E. and Weber, G. (2016), IBM Watson: how cognitive computing
can be applied to big data challenges, in Life Sciences Research, Clinical
Therapeutics, 38(4), 688-701.
CNYCC, The Central New York Care Collaborative (2017), Central New York Care
Collaborative Announces Partnership with IBM Watson Health on Regional
Population Health Management System, available at the URL: https://cnycares.
org/news/ibm-watson-health-regional-phm, last visited, Nov., 2017.
Coccoli, M., Maresca, P. and Stanganelli, L. (2016), Cognitive computing in education,
in Journal of e-Learning and Knowledge Society, 12(2), 55-69.
Coccoli, M., Maresca, P. and Stanganelli, L. (2017), The role of big data and cognitive
computing in the learning process, in Journal of Visual Languages & Computing,
38(1), 97-103.
Coccoli, M. and Torre, I. (2014), Interacting with annotated objects in a semantic
web of things application, in Journal of Visual Languages & Computing, 25(6),
1012-1020.
Cortada, J.W., Gordon, D. and Lenihan, B. (2012), The value of analytics in healthcare:
From insights to outcomes, IBM Global Business Services, Life Sciences and
Healthcare, Executive Report.
Devarakonda M.V. and Mehta, N. (2016), Cognitive computing for electronic medical
records, in Healthcare Information Management Systems, Part of the series Health
Informatics, 555-577.

67
 Je-LKS PEER REVIEWED PAPERS - NEW TRENDS, CHALLENGES AND PERSPECTIVES ON HEALTHCARE COGNITIVE COMPUTING:
FROM INFORMATION EXTRACTION TO HEALTHCARE ANALYTICS
Vol. 14, n. 1, January 2018

Douglass, C. and Adigun, B.J. (2017), Central New York Care Collaborative (CNYCC)
chooses IBM Watson care manager to improve health across the Region, available at
the URL: http://www-03.ibm.com/press/us/en/pressrelease/51646.wss, last visited,
Nov., 2017.
Douglass, C. and Kearns, K. (2017), IBM and New York Genome Center to create
comprehensive, open cancer data repository to tap cognitive insights from Watson,
available at the URL: https://www-03.ibm.com/press/us/en/pressrelease/49232.
wss, last visited, Nov., 2017.
Earley, S. (2015), Executive roundtable series: machine learning and cognitive
computing, in IT Professional, 17(4), 56-60.
Gatenbein, R.E. (2014), Watson, come here! The role of intelligent system in Healthcare,
in Proceedings of the World Automation Congress, 1-4.
Holtel, S. (2014), More the end of information overflow: how IBM Watson turn upside
down our view on information appliances, Proceedings of the 2014 IEEE World
Forum on Internet of Things (WF-I.T), 187-188.
Kohn, M.S., Sun, J., Knoop, S., Shabo, A., Carmeli, B., Sow, D., Syed-Mahmood, T.
and Rapp, W. (2014), IBM’s health analytics and clinical decision support, in IMIA
Year book, 154-162.
Lee, H. (2014), Paging Dr. Watson: IBM’s Watson supercomputer now being used in
healthcare, in Journal of AHIMA, 85(5), 44-47.
Malin, J.L. (2013), Envisioning Watson as a rapid-learning system for oncology, in
Journal of Oncology Practice, 9(3), 155-157.
Manipal Hospitals, (2017), Manipal Hospitals is the first healthcare provider in India
to adopt Watson for oncology for upgradation of cancer care, available at the URL:
https://www.manipalhospitals.com/blog/ibm-watson-for-oncology, last visited,
Nov., 2017.
Modha, D.S., Ananthanarayanan, R., Esser, S.K., Ndirango, A., Sherbondy, A.J. and
Singh, R. (2011), Cognitive computing, Communications of the ACM, 54(8), 62-71.
Mohideen, R. and Evans, R. (2015), Shaping Our Technological Futures, in IEEE
Technology and Society Magazine, 34(4), 83-86.
Nishida, T., Nakazawa, A., Ohmoto, Y. and Mohammad, Y. (2014), History of
conversational system development, in Conversational Informatics: A Data-
Intensive Approach with Emphasis on Nonverbal Communication, 43-62, Springer,
Tokyo.
Noor, A.K. (2015), Potential of cognitive computing and cognitive systems, Open Eng.,
5, 75–88.
Ogiela, L., Tadeusiewicz, R. and Ogiela, M.R. (2006), Cognitive computing in
intelligent medical pattern recognition systems, in Intelligent Control and
Automation, Springer Berlin Heidelberg, 851-856.
Orii, Y., Horibe, A., Matsumoto, K., Aoki, T., Sueoka, K., Kohara, S., Okamoto,
K., Yamamichi, S., Hosokawa, K. and Mori, H. (2016), Advanced interconnect
technologies in the era of cognitive computing, Proceedings of the Pan Pacific
Microelectronics Symposium (Pan Pacific), 1-6.

68
 Mauro Coccoli, Paolo Maresca - Adopting Cognitive Computing Solutions in Healthcare

Pazzani, M.J. and Billsus, D. (2007), Content-based recommendation systems, in The

Adaptive Web, P. Brusilowsky, A. Kobsa and W. Nejdl (Eds.), LNCS 4321, 325-
341, Springer-Verlag, Berlin Heidelberg.
Ratner, M. (2015), IBM’s Watson group signs up genomics partners, in Nature
Biotechnology, 33(1), 10-11.
Strickland, E. (2013), Watson goes to Med School, IEEE Spectrum, 50(1), 42-45.
Tan, S.S.-L., Gao, G. and Koch, S. (2015), Big data and analytics in healthcare,
Methods of Information in Medicine, 54(6), 546-547.
Teo, C.L., Yang, Y., Daumè, H., Fermuller, C. and Aloimonos, Y. (2012), Towards a
Watson that sees: language-guided action recognition for robots, in Proceedings
of the International Conference on Robotics and Automation (ICRA), 374-381.
UNC Lineberger Comprehensive Cancer Center (2017), Partnering with IBM and
Watson to accelerate DNA analysis and inform personalized treatment, available
at the URL: https://unclineberger.org/news/partnering-with-ibm-and-watson-to-
accelerate-dna-analysis-and-inform-personalized-treatment, last visited, Nov., 2017.
Zhang, M., Zhao, H., Zheng, R., Wu, Q., and Wei, W. (2012), Cognitive Internet of
Things: concepts and application example, in International Journal of Computer
Science, 9(6-3), 151-158.

69

View publication stats