ArticlePDF Available

Artificial neural networks in medical diagnosis

Authors:
  • Universidad de La Laguna, Tenerife, Spain

Abstract and Figures

An extensive amount of information is currently available to clinical specialists, ranging from details of clinical symptoms to various types of biochemical data and outputs of imaging devices. Each type of data provides information that must be evaluated and assigned to a particular pathology during the diagnostic process. To streamline the diagnostic process in daily routine and avoid misdiagnosis, artificial intelligence methods (especially computer aided diagnosis and artificial neural networks) can be employed. These adaptive learning algorithms can handle diverse types of medical data and integrate them into categorized outputs. In this paper, we briefly review and discuss the philosophy, capabilities, and limitations of artificial neural networks in medical diagnosis through selected examples.
Content may be subject to copyright.
Journal of
APPLIED
BIOMEDICINE
EDITORIAL
Articial neural networks in medical diagnosis
Filippo Amato
1
, Alberto López
1*
, Eladia María Peña-Méndez
2
, Petr Vaňhara
3
, Aleš Hampl
3, 4
,
Josef Havel
1, 5, 6
1
Department of Chemistry, Faculty of Science, Masaryk University, Brno, Czech Republic
2
Department of Analytical Chemistry, Nutrition and Food Science, Faculty of Chemistry, University of La Laguna,
La Laguna, Tenerife, Spain
3
Department of Histology and Embryology, Faculty of Medicine, Masaryk University, Brno, Czech Republic
4
International Clinical Research Center, St. Anne’s University Hospital, Brno, Czech Republic
5
Department of Physical Electronics, Faculty of Science, Masaryk University, Brno, Czech Republic
6
R&D Centre for low-cost plasma and nanotechnology surface modications, CEPLANT, Masaryk University,
Brno, Czech Republic
* On leave from the University of Salamanca, Spain
Received 17
th
December 2012.
Published online 7
th
January 2013.
Summary
An extensive amount of information is currently available to clinical specialists, ranging from details of
clinical symptoms to various types of biochemical data and outputs of imaging devices. Each type of data
provides information that must be evaluated and assigned to a particular pathology during the diagnostic
process. To streamline the diagnostic process in daily routine and avoid misdiagnosis, articial intelligence
methods (especially computer aided diagnosis and articial neural networks) can be employed. These
adaptive learning algorithms can handle diverse types of medical data and integrate them into categorized
outputs. In this paper, we briey review and discuss the philosophy, capabilities, and limitations of articial
neural networks in medical diagnosis through selected examples.
Key words: medical diagnosis; articial intelligence; articial neural networks; cancer; cardiovascular
diseases; diabetes
J Appl Biomed. 11: 47–58, 2013
DOI 10.2478/v10136-012-0031-x
ISSN 1214-0287
Josef Havel, Department of Chemistry, Faculty
of Science, Masaryk University, Kamenice 5/
A14, 625 00 Brno, Czech Republic
havel@chemi.muni.cz
+420 549 494 114
+420 549 492 494
© Journal of Applied Biomedicine
INTRODUCTION
Articial neural networks (ANNs) are widely used
in science and technology with applications in
various branches of chemistry, physics, and biology.
For example, ANNs are used in chemical kinetics
(Amato et al. 2012), prediction of the behavior of
industrial reactors (Molga et al. 2000), modeling
kinetics of drug release (Li et al. 2005), optimization
of electrophoretic methods (Havel et al. 1998),
classication of agricultural products such as onion
varieties (Rodrguez Galdn et al. 2010), and even
species determination (Fedor et al. 2008, Michalkova
et al. 2009, Murarikova et al. 2010). In general,
very diverse data such as classication of biological
objects, chemical kinetic data, or even clinical
parameters can be handled in essentially the same
way. Advanced computational methods, including
ANNs, utilize diverse types of input data that are
processed in the context of previous training history
48
Amato et al.: Articial neural networks in medical diagnosis
on a dened sample database to produce a clinically
relevant output, for example the probability of a
certain pathology or classication of biomedical
objects. Due to the substantial plasticity of input data,
ANNs have proven useful in the analysis of blood
and urine samples of diabetic patients (Catalogna et
al. 2012, Fernandez de Canete et al. 2012), diagnosis
of tuberculosis (Er et al. 2008, Elveren and Yumuşak
2011), leukemia classication (Dey et al. 2012),
analysis of complicated effusion samples (Barwad
et al. 2012), and image analysis of radiographs or
even living tissue (Barbosa et al. 2012, Saghiri et al.
2012).
The aim of this paper is to present the general
philosophy for the use of ANNs in diagnostic
approaches through selected examples, documenting
the enormous variability of data that can serve as
inputs for ANNs. Attention will not only be given
to the power of ANNs applications, but also to
evaluation of their limits, possible trends, and future
developments and connections to other branches of
human medicine (Fig. 1).
Fig. 1. Overview of the main applications of articial neural networks in medicine.
ARTIFICIAL NEURAL NETWORKS
An ANN is a mathematical representation of the
human neural architecture, reecting its “learning”
and “generalization” abilities. For this reason, ANNs
belong to the eld of articial intelligence. ANNs are
widely applied in research because they can model
highly non-linear systems in which the relationship
among the variables is unknown or very complex. A
review of various classes of neural networks can be
found in (Aleksander and Morton 1995, Zupan and
Gasteiger 1999).
Mathematical background
A neural network is formed by a series of “neurons”
(or “nodes”) that are organized in layers. Each neuron
in a layer is connected with each neuron in the next
layer through a weighted connection. The value of
the weight w
ij
indicates the strength of the connection
between the i-
th
neuron in a layer and the j-
th
neuron
in the next one.
The structure of a neural network is formed by
an “input” layer, one or more “hidden” layers, and
the “output” layer. The number of neurons in a
layer and the number of layers depends strongly on
the complexity of the system studied. Therefore, the
optimal network architecture must be determined.
The general scheme of a typical three-layered ANN
architecture is given in Fig. 2.
The neurons in the input layer receive the data
and transfer them to neurons in the rst hidden
layer through the weighted links. Here, the data are
mathematically processed and the result is transferred
to the neurons in the next layer. Ultimately, the neurons
in the last layer provide the network’s output. The j-
th
neuron in a hidden layer processes the incoming data
(x
i
) by: (i) calculating the weighted sum and adding a
“bias” term (θ
j
) according to Eq. 1:
(1)
n,=jθ+wx=net
jij
m
=i
ij
2,1,
1
49
Amato et al.: Articial neural networks in medical diagnosis
Fig. 2. General structure of a neural network with two hidden layers. The w
ij
is the weight of the connection between the i-
th
and the j-
th
node.
(ii) transforming the net
j
through a suitable mathe-
matical “transfer function”, and (iii) transferring the
result to neurons in the next layer. Various transfer
functions are available (Zupan and Gasteiger 1999);
however, the most commonly used is the sigmoid
one:
Network learning
The mathematical process through which the network
achieves “learning” can be principally ignored by
the nal user. In this way, the network can be viewed
as a “black box” that receives a vector with m inputs
and provides a vector with n outputs (Fig. 3). Here
we will give only a brief description of the learning
process; more details are provided for example in the
review by (Basheer and Hajmeer 2000). The network
“learns” from a series of “examples” that form
the “training database” (Fig. 4). An “example” is
formed by a vector X
im
= (x
i1
, x
i2
, …., x
im
) of inputs and
a vector Y
in
= (y
i1
, y
i2
,....., y
in
) of outputs. The objective
of the training process is to approximate the function
f between the vectors X
im
and the Y
in
:
(2)
 
x
e+
=xf
1
1
(3)
mi,ni,
Xf=Y
This is achieved by changing iteratively the values
of the connection weights (w
ij
) according to a suitable
mathematical rule called the training algorithm.
The values of the weights are changed by using
the steepest descent method to minimize a suitable
function used as the training stopping criteria. One
of the functions most commonly used is the sum-of-
squared residuals given by Eq. 4:
(4)
where y
ij
and y
ij
*
are the actual and network’s j-
th
output
corresponding to the i-
th
input vector, respectively.
The current weight change on a given layer is
given by Eq. (5):
(5)
ij
ij
dw
dE
η=Δw
where η is a positive constant called the learning rate.
To achieve faster learning and avoid local minima, an
additional term is used and Eq. 5 becomes:
(6)
1k
ij
ij
k
ij
μΔw+
dw
dE
η=Δw
where μ is the “momentum” term and Δw
ij
k-1
is the
change of the weight w
ij
from the (k-1)-
th
learning
cycle. The learning rate controls the weight update
rate according to the new weight change and the
momentum acts as a stabilizer, being aware of the
previous weight change.
m
i
n
j
ijij
yy=E
1 1
2
*
2
1
50
Amato et al.: Articial neural networks in medical diagnosis
Fig. 3. Details of input and output items concerning ANNs-based diagnosis (ANN architecture is often hidden and it is
indicated here as a black box).
Fig. 4. Example of training database structure. Each row refers to a different patient labeled with a numerical code. The
element data
k,i
refers to the i-
th
medical data (symptom, laboratory data, etc.) of the k-
th
patient.
The function given by Eq. 4 is also used as the
criterion to optimize the network architecture because
it depends on the number of hidden layers and the
number of neurons therein. To nd the optimal
architecture, the most common approach is to plot the
value of E (Eq. 4) as a function of the number of nodes
in the hidden layer (q). An example of such a plot is
given in Fig. 5. As q increases, E decreases. However,
after an optimal value of q the improvement is rather
poor. Usually, the optimal value of q is found from
the intersection point of the two branches of the plot.
After the optimal neural network architecture is
found, the training process is performed until a proper
minimum value of E is reached. Afterward, the
network is checked with examples not previously used
in the training step. This process is called verication.
Finally, the network can be used to predict outputs for
new input vectors.
Structure of the training database
As stated above, the network must be trained using a
suitable database. The database is a table (or matrix)
of data concerning patients for whom the diagnosis
(positive or negative) about a certain disease is already
known. Each row of the matrix refers to one patient.
The rst m elements of the row are medical data and
the last n elements represent the output (diagnosis).
The term “medical data” indicates biochemical,
nuclear magnetic resonance (NMR), laboratory data,
and symptoms and other information provided by
the medical specialist (Table 1). An example of such
training matrix with one output variable (n = 1) that
may assume two possible values (positive or negative)
is given in Fig. 4.
51
Amato et al.: Articial neural networks in medical diagnosis
Table 1. Brief overview of data in clinical context used as inputs for ANN.
Input data or method Clinical context Output information Reference
Age, cholesterol
concentration, arterial
hypertension
Coronary artery disease Diagnosis (Atkov et al. 2012)
Heart sound Valve stenosis Diagnosis (Uğuz 2012)
Hematologic prole Chronic myeloid leukemia Classication of leukemia (Dey et al. 2012)
Visual information of
wireless capsule endoscopy
Small bowel tumors
Diagnosis, classication of
tumor
(Barbosa et al. 2012)
Glucose concentration –
Near-infrared spectroscopy
Diabetes Diagnosis (Arnold 1996)
Demographic and
clinicopathologic data,
surgical outcome
Hepatocellular carcinoma
Prediction of disease free
survival
(Ho et al. 2012)
Cytology of effusion uid Carcinoma Presence of malignant cells (Barwad et al. 2012)
Speech record Oral/Oropharyngeal cancer
Detection of nasalence
(hypernasality)
(de Bruijn et al. 2011)
Electroencephalographic
(EEG) recordings
Epilepsy Prediction of seizures
(Fernandez-Blanco et al.
2012)
OVERVIEW OF ANNs IN MEDICAL
DIAGNOSIS
There are several reviews concerning the application
of ANNs in medical diagnosis. The concept was rst
outlined in 1988 in the pioneering work of (Szolovits
et al. 1988) and since then many papers have been
published. The general application of ANNs in
medical diagnosis has previously been described
(Alkim et al. 2012). For example, ANNs have been
applied in the diagnosis of: (i) colorectal cancer (Spelt
et al. 2012), (ii) multiple sclerosis lesions (Mortazavi
et al. 2012a, b), (iii) colon cancer (Ahmed 2005),
(iv) pancreatic disease (Bartosch-Härlid et al. 2008),
(v) gynecological diseases (Siristatidis et al. 2010),
and (vi) early diabetes (Shankaracharya et al. 2010). In
addition, ANNs have also been applied in the analysis
Fig. 5. Example of the plot used to select the optimal number of nodes in a given hidden layer. It is indicated that too high
number of nodes might lead to overtting.
52
Amato et al.: Articial neural networks in medical diagnosis
of data and diagnostic classication of patients with
uninvestigated dyspepsia in gastroenterology (Pace
and Savarino 2007) and in the search for biomarkers
(Bradley 2012). A novel, general, fast, and adaptive
disease diagnosis system has been developed
based on learning vector quantization ANNs. This
algorithm is the rst proposed adaptive algorithm
and can be applied to completely different diseases,
as demonstrated by the 99.5% classication accuracy
achieved for both breast and thyroid cancers. Cancer,
diabetes, and cardiovascular diseases are among the
most serious and diverse diseases. The amount of
data coming from instrumental and clinical analysis
of these diseases is quite large and therefore the
development of tools to facilitate diagnosis is of great
relevance. For this reason, we will provide a brief
overview of the advances in the application of ANNs
to the eld of diagnosis for each of these diseases.
Cardiovascular diseases
Cardiovascular diseases (CVDs) are dened as all
diseases that affect the heart or blood vessels, both
arteries and veins. They are one of the most important
causes of death in several countries. According to the
National Center of Health Statistics (NCHS, www.
cdc.gov/nchs), CVD represents the leading cause of
death in the United States. CVD has therefore become
an important eld of study during the last 20 years.
Based on a bibliography search (ScienceDirect),
more than one thousand papers about the use of
ANNs in cardiovascular diseases and related topics
have been published since 2008. According to the
NCHS, coronary artery disease (CAD) is currently
the leading cause of death worldwide, therefore
early diagnosis is very important. With this aim,
Karabulut and Ibrikçi applied ANNs with the
Levenberg-Marquardt back propagation algorithm
as base classiers of the rotation forest ensemble
method (Karabulut and Ibrikçi 2012). Diagnosis of
CAD with 91.2% accuracy was achieved from data
collected non-invasively, cheaply, and easily from
the patient. Other data such as age, different kinds of
cholesterol, or arterial hypertension have been used
to diagnose CAD (Atkov et al. 2012). The model
that performed with the best accuracy (93%) was
the one that included both genetic and non-genetic
factors related to the disease. Despite these promising
results, it must be noted that for some models the
accuracy was lower than 90%. ANNs have also been
applied in other heart diseases, such as heart valve
defects (Uğuz 2012). Heart sound recordings were
obtained from 120 subjects and classied as normal,
pulmonary, and mitral stenosis heart valve diseases
via stethoscopy. Correct classication was achieved
for 95% of the different sounds. An average accuracy
of 99.8% and 99.2% was achieved by two different
models developed by (Özbay 2009) for the diagnosis
of arrhythmias. Therefore, ANNs can be applied to the
medical diagnosis of completely different diseases,
for example detection of arrhythmias or coronary
artery diseases, which are major causes of death
worldwide. Classication accuracies higher than 90%
are typically achieved, even exceeding 99% in some
cases. As a result, ANNs have substantial potential in
CVD diagnosis.
Cancer
According to the American Cancer Society (www.
cancer.org), there will be more than 1.6 million
newly diagnosed cases of cancer in the US in 2012.
A rapid and correct diagnosis is essential for the
clinical management of cancer, including selection
of the most suitable therapeutic approach. The use of
ANNs in distinguishing particular cancer types or the
prediction of cancer development emerged in the late
1990s as a promising computational-based diagnostic
tool using various inputs. Use of novel molecular
approaches, such as micro-RNA screens, broadens the
possibilities for the application of ANNs in the search
for patterns specic for a certain disease, for example
rectal cancer and its response to cytoreductive therapy
(Kheirelseid et al. 2012). The application of neural
networks trained on dened data sets was evaluated
in 1994 for breast and ovarian cancer, opening a
discussion on the suitability of particular data as
inputs for ANN analysis, for example demographic
(age), radiological (NMR), oncologic (tumor markers
CA 15-3 or CA 125), and biochemical (albumin,
cholesterol, high-density lipoprotein cholesterol,
triglyceride, apolipoproteins A1 and B) data (Wilding
et al. 1994). Later on, reasonable prediction of the
measured in vitro chemotherapeutic response based
on
1
H NMR of glioma biopsy extracts was achieved
using ANNs to obtain automatic differential diagnosis
of glioma (El-Deredy et al. 1997).
Moreover, advanced analytical methods, such as
mass spectrometry, can offer suitable information for
clinically relevant ANN analysis. This technique has
recently been applied successfully in the diagnosis
of ovarian cancer. Upon identication of the most
informative points of the mass spectrum curve, for
example by data mining, the integrated feed-forward
ANN classier showed an overall sensitivity of
98% and a specicity of 96%, surpassing standard
statistical methods such as Student’s t-test (Thakur et
al. 2011).
The applications of ANNs in radiology aim
to develop automated decision support systems,
assisting, for example, in the classication of brain
tumors by magnetic resonance (Tate et al. 2006). The
53
Amato et al.: Articial neural networks in medical diagnosis
concept and design raised by Tate was also applied
by Brougham and colleagues on lung carcinoma
(Brougham et al. 2011). In this process, the same
experimental protocol was carried out by two
different teams of researchers examining two sets
of whole-cell
1
H NMR spectra at different times.
Whole-cell
1
H NMR spectra were recorded for two
groups of human lung carcinoma cell lines, which
were grown in culture and included (i) the parent
cell line DLKP, a human squamous non-small cell
lung carcinoma; (ii) DLKP-A and (iii) DLKP-A5F,
two resistant daughter lines; and (iv) A549 a human
lung adenocarcinoma cell line. Despite demonstrated
operator-induced sources of variation in the network,
the ANN was able to classify the cell line correctly
in 100% of cases independent of the spectra selected
for training and validation. Therefore, the power of
ANNs for the classication of different types of lung
carcinoma in real clinical or laboratory situations was
demonstrated. In summary, ANNs have been shown
to use various primary data, ranging from clinical
parameters to biochemical values, and provide
increased diagnostic accuracy for various kinds of
cancers.
Diabetes
Diabetes represents a serious health problem in
developed countries, with estimated numbers
reaching 366 million diabetes cases globally in 2030
(Leon et al. 2012). The most common type of diabetes
is type II, in which the cellular response to insulin is
impaired leading to disruption of tissue homeostasis
and hyperglycemia. The standard in diabetes
diagnosis or monitoring is direct measurement
of glucose concentration in blood samples. Non-
invasive methods based on near-infrared or Raman
spectroscopy to monitor glucose levels were
developed in 1992 (Arnold 1996) and nowadays
are even available as a smartphone application.
The ANNs extrapolate glucose concentrations from
spectral curve, thus enabling convenient monitoring
of diabetes during daily activities.
Quality of life itself, including satisfaction, social
interactions, and depression, has been dened as an
integral outcome measure in management of diabetes
mellitus. Recently, Narasingarao et al developed
a prototype neural network model to measure the
quality of life in diabetic patients (Narasingarao et
al. 2009). For this particular purpose, biological or
biographical variables such as age, gender, weight, or
fasting plasma glucose were used as input data. The
results were comparable with those from traditional
statistical methods. In 1997, simulation studies on
neural predictive control of glucose concentration
in subcutaneous tissue were carried out (Trajanoski
et al. 1998). As a control strategy, Trajanoski
and colleagues combined ANNs and nonlinear
model predictive control. This approach allowed
for large noise levels and time delays. However,
administration of insulin or analogues was found
to be unsuitable for rapid control and was adequate
only for controlling slow perturbations. Later, a
new control system called a neuro-fuzzy control
system, which uses fuzzy logic principles and neural
networks employing a minimal number of input data
for correct output, was developed (Dazzi et al. 2001).
This system is especially suitable for direct control
of critically ill diabetic patients on intravenous
nutrition and reduced the need for increased blood
glucose testing and even the risk of hypoglycemia.
Blood glucose level has also been monitored using
inverse optimal neural control as a preliminary study
in patients with type I diabetes (Leon et al. 2012).
Recurrent neural networks were used to control
the level of glucose and insulin following a desired
trajectory (normal glucose absorption of a healthy
person) to avoid hypoglycemia and hyperglycemia.
ANNs have not only been applied in the tracking of
glucose, but also in the diagnosis of diabetes. A neural
network-based diagnostic system was developed by
(Chan et al. 2011) with the aim of predicting fatal
hypoglycemia episodes in type I diabetes patients.
The data were collected from a cohort of 420 patients
and included: (i) physiological parameters, (ii) rate
of change of heart rate, (iii) corrected QT interval of
electrocardiogram signal, and (iv) rate of change of
corrected QT interval. Data from 320 patients were
used to train the network and the remainder for its
verication. A sensitivity of 79.30% and specicity
of 60.53% was achieved, better values than those
obtained using other approaches such as statistical
or fuzzy regression methods. In conclusion, the use
of ANNs in the diagnosis of diabetes provides an
improvement in accuracy, sensitivity, and specicity
in comparison with other methods, thus contributing
to improved clinical management of diabetes
mellitus.
FUNDAMENTAL STEPS IN ANNs-BASED
MEDICAL DIAGNOSIS
The workow of ANN analysis arising from the
outlined clinical situations is shown in Fig. 6 which
provides a brief overview of the fundamental steps that
should be followed to apply ANNs for the purposes of
medical diagnosis with sufcient condence.
54
Amato et al.: Articial neural networks in medical diagnosis
For the reasons discussed above, the network
receives patient’s data to predict the diagnosis of a
certain disease. After the target disease is established,
the next step is to properly select the features (e.g.,
symptoms, laboratory, and instrumental data) that
provide the information needed to discriminate the
different health conditions of the patient. This can be
done in various ways. Tools used in chemometrics
allow the elimination of factors that provide only
redundant information or those that contribute only
to the noise. Therefore, careful selection of suitable
features must be carried out in the rst stage. In
the next step, the database is built, validated and
“cleaned” of outliers. After training and verication,
the network can be used in practice to predict
the diagnosis. Finally, the predicted diagnosis is
evaluated by a clinical specialist. The major steps can
be summarized as:
Features selection
Building the database
Data cleaning and preprocessing
Data homoscedasticity
Training and verication of database using ANN
Network type and architecture
Training algorithm
Verication
Robustness of ANN-based approaches
Testing in medical practice
Fig. 6. Diagram of fundamental steps in ANNs-based medical diagnosis. Building of the database and “learning” represents
the left half (green) and its application for the diagnosis is the right part (blue).
The individual steps listed above will be shortly
commented and some details given.
Features selection
Correct diagnosis of any disease is based on
various, and usually incoherent, data (features): for
example, clinicopathologic evaluation, laboratory
and instrumental data, subjective anamnesis of the
patient, and considerations of the clinician. Clinicians
are trained to extract the relevant information from
each type of data to identify possible diagnoses.
In articial neural network application such data
are called “features”. Features can be symptoms,
biochemical analysis data and/or whichever other
relevant information helping in diagnosis. Therefore,
the experience of the professional is closely related to
the nal diagnosis. The ability of ANNs to learn from
examples makes them very exible and powerful tools
to accelerate medical diagnosis. Some types of neural
networks are suitable for solving perceptual problems
while others are more adapted for data modeling and
functional approximation (Dayhoff and Deleo 2001).
Regardless of the features selected, those chosen
for training the neural network should be “robust”
indicators for a given clinical situation or pathology.
In general, feature selection relies upon previous
clinical experience. Features that bring insufcient,
redundant, non-specic, or noisy information about
the investigated problem should be avoided. The
selection/extraction of suitable features among all
55
Amato et al.: Articial neural networks in medical diagnosis
available ones is usually carried out using various
approaches. The most important and best-known tools
for variable selection are powerful mathematical means
of data mining such as principal components analysis,
genetic algorithm (Yan et al. 2008), or ANNs (Verikas
and Bacauskiene 2002).
Building the database
The neural network is trained using a suitable database
of “example” cases. An “example” is provided by one
patient whose values for the selected features have been
collected and evaluated. The quality of training and the
resultant generalization, and therefore the prediction
ability of the network, strongly depend on the database
used for the training. The database should contain a
sufcient number of reliable “examples” (for which
the diagnosis is known) to allow the network to learn
by extracting the structure hidden in the dataset and
then use this “knowledge” to “generalize” the rule to
new cases. In addition, clinical laboratory data should
be in a form that is readily transferable to programs for
computer-aided diagnosis (Strike et al. 1986).
Data cleaning and preprocessing
Data in the training database must be preprocessed
before evaluation by the neural network. Several
approaches are available for this purpose. Data are
normally scaled to lie within the interval [0, 1] because
the most commonly used transference function is the so-
called logistic one. In addition, it has been demonstrated
that cases for which some data are missing should be
removed from the database to improve the classication
performance of the network (Gannous and Elhaddad
2011). A decrease in the classication performance
of the network is observed for imbalanced databases
(those with a different number of cases for each class)
(Mazurowski et al. 2008).
Data homoscedasticity
Once the suitable features, database, data preprocessing
method, training algorithm, and network architecture
have been identied, data concerning “new” patients
who are not included in the training database can be
evaluated by the trained network. The question asked is
whether the new data belong to the same population as
those in the database (homoscedasticity). Failure at this
step might lead the network to misclassify the new data.
This problem can be solved by the use of an additional
parameter that indicates the population to which a
certain sample belongs.
Training and verication of database using ANN
Network type and architecture
Although multilayer feed-forward neural networks
are most often used, there are a large variety of other
networks including bayesian, stochastic, recurrent, or
fuzzy. The optimal neural network architecture must be
selected in the rst stage. This is usually done testing
networks with different number of hidden layers and
nodes therein. The optimal architecture is that for which
the minimum value of E (Eq. 4) for both training and
verication is obtained.
Training algorithm
Various training algorithms are available. However,
the most commonly used is back propagation (Zupan
and Gasteiger 1999; Ahmed 2005). As discussed
in “Network learning” section, backpropagation
algorithm requires the use of two training parameters:
(i) learning rate and (ii) momentum. Usually, high
values of such parameters lead to unstable learning, and
therefore poor generalization ability of the network.
The optimal values of the training parameters depend
upon the complexity of the studied system. In general,
the value of momentum is lower than that of learning
rate. In addition, the sum of their values should be
approximately equal to one.
Verication
ANNs-based medical diagnosis should be veried by
means of a dataset different from that one used for
training.
Robustness of ANN-based approaches
It is well known that ANNs are able to tolerate a certain
level of noise in the data and consequently they typically
provide sufcient prediction accuracy. However,
this noise might sometimes cause misleading results,
especially when modeling very complex systems
such as the health condition of a human body. Such
noise would not only impact the normal uncertainty
of the measured data but might also impact secondary
factors, for example the coexistence of more than one
disease. Crossed effects cannot be predicted unless they
have been considered during building of the training
database. Any factor that inuences the symptoms of
the disease under study must be taken into account by
including such cases in the database. Only in this way
can the network correctly classify the patient. Of course,
one way to avoid this is to combine the experience of
the clinical specialist with the discriminative power of
ANN-based approaches.
Testing in medical practice
As the nal step in ANN-aided diagnosis should be
testing in medical practice. For each new patient the
network’s outcome is to be carefully examined by
a clinician. Medical data of patients for which the
predicted diagnosis is correct can be eventually included
in the training database.
56
Amato et al.: Articial neural networks in medical diagnosis
However, wide and extensive evaluation of ANN-
aided diagnosis applications in clinical setting is
necessary even throughout different institutions. Veried
ANN-aided medical diagnosis support applications
in clinical setting are necessary condition for further
expansion in medicine.
CONCLUSIONS
ANNs represent a powerful tool to help physicians
perform diagnosis and other enforcements. In this
regard, ANNs have several advantages including:
(i) The ability to process large amount of data
(ii) Reduced likelihood of overlooking relevant
information
(iii) Reduction of diagnosis time
ANNs have proven suitable for satisfactory diagnosis
of various diseases. In addition, their use makes the
diagnosis more reliable and therefore increases patient
satisfaction. However, despite their wide application in
modern diagnosis, they must be considered only as a
tool to facilitate the nal decision of a clinician, who
is ultimately responsible for critical evaluation of the
ANN output. Methods of summarizing and elaborating
on informative and intelligent data are continuously
improving and can contribute greatly to effective,
precise, and swift medical diagnosis.
ACKNOWLEDGEMENTS
Support from Ministry of Education, Youth and Sports
of the Czech Republic (Projects MSM0021622411,
MSM0021627501, MSM0021622430, CZ.1.05/1.1.00/
02.0123 and HistoPARK CZ.1.07/2.3.00/20.0185),
and the Czech Science Foundation (Projects No.
104/08/0229, 202/07/1669) is acknowledged. This
research was also supported by CEPLANT, the project
R&D center for low-cost plasma and nanotechnology
surface modications (CZ.1.05/2.1.00/03.0086 funding
from the European Regional Development Fund). E. M.
Peña-Méndez thanks the University of La Laguna for
partial support.
REFERENCES
Ahmed F. Articial neural networks for diagnosis and
survival prediction in colon cancer. Mol Cancer. 4:
29, 2005.
Aleksander I, Morton H. An introduction to neural
computing. Int Thomson Comput Press, London
1995.
Alkim E, Gürbüz E, Kiliç E. A fast and adaptive
automated disease diagnosis method with an
innovative neural network model. Neur Networks.
33: 88–96, 2012.
Amato F, González-Hernández J, Havel J. Articial
neural networks combined with experimental design:
a “soft” approach for chemical kinetics. Talanta. 93:
72–78, 2012.
Arnold M. Non-invasive glucose monitoring. Curr Opin
Biotech. 7: 46–49, 1996.
Atkov O, Gorokhova S, Sboev A, Generozov E,
Muraseyeva E, Moroshkina S and Cherniy N.
Coronary heart disease diagnosis by articial neural
networks including genetic polymorphisms and
clinical parameters. J Cardiol. 59: 190–194, 2012.
Barbosa D, Roupar D, Ramos J, Tavares A and Lima
C. Automatic small bowel tumor diagnosis by using
multi-scale wavelet-based analysis in wireless
capsule endoscopy images. Biomed Eng Online. 11:
3, 2012.
Bartosch-Härlid A, Andersson B, Aho U, Nilsson J,
Andersson R. Articial neural networks in pancreatic
disease. Br J Surg. 95: 817–826, 2008.
Barwad A, Dey P, Susheilia S. Articial neural network
in diagnosis of metastatic carcinoma in effusion
cytology. Cytometry B Clyn Cytom. 82: 107–111,
2012.
Basheer I, Hajmeer M. Articial neural networks:
fundamentals, computing, design, and application. J
Microbiol Meth. 43: 3–31, 2000.
Bradley B. Finding biomarkers is getting easier.
Ecotoxicology. 21: 631–636, 2012.
Brougham D, Ivanova G, Gottschalk M, Collins D,
Eustace A, O’Connor R, Havel J. Articial neural
networks for classication in metabolomic studies
of whole cells using 1H nuclear magnetic resonance.
J Biomed Biotechnol. 2011: 158094, 2011.
Catalogna M, Cohen E, Fishman S, Halpern Z, Nevo
U, Ben-Jacob E. Articial neural networks based
controller for glucose monitoring during clamp test.
PloS One. 7: e44587, 2012.
Chan K, Ling S, Dillon T, Nguyen H. Diagnosis of
hypoglycemic episodes using a neural network
based rule discovery system. Expert Syst Appl. 38:
9799–9808, 2011.
Dayhoff J, Deleo J. Articial Neur Networks: Opening
the Black Box. Cancer. 91: 1615–1635, 2001.
Dazzi D, Taddei F, Gavarini A, Uggeri E, Negro R,
Pezzarossa A. The control of blood glucose in the
critical diabetic patient: a neuro-fuzzy method. J
Diabet Complicat. 15: 80–87, 2001.
57
Amato et al.: Articial neural networks in medical diagnosis
de Bruijn M, ten Bosch L, Kuik D, Langendijk J,
Leemans C, Verdonck-de Leeuw I. Articial neural
network analysis to assess hypernasality in patients
treated for oral or oropharyngeal cancer. Logoped
Phoniatr Vocol. 36: 168–174, 2011.
Dey P, Lamba A, Kumari S, Marwaha N. Application
of an articial neural network in the prognosis of
chronic myeloid leukemia. Anal Quant Cytol Histol.
33: 335–339, 2012.
El-Deredy W, Ashmore S, Branston N, Darling J,
Williams S, Thomas D. Pretreatment prediction of
the chemotherapeutic response of human glioma
cell cultures using nuclear magnetic resonance
spectroscopy and articial neural networks Cancer
Res. 57: 4196-4199, 1997.
Elveren E, Yumuşak N. Tuberculosis disease diagnosis
using articial neural network trained with genetic
algorithm. J Med Syst. 35: 329–332, 2011.
Er O, Temurtas F, Tanrıkulu A. Tuberculosis Disease
Diagnosis Using Articial Neural Networks. J Med
Syst. 34: 299–302, 2008.
Fedor P, Malenovsky I, Vanhara J, Sierka W, Havel J.
Thrips (Thysanoptera) identication using articial
neural networks. Bull Entomol Res. 98: 437–447,
2008.
Fernandez de Canete J, Gonzalez-Perez S, Ramos-Diaz
JC. Articial neural networks for closed loop control
of in silico and ad hoc type 1 diabetes. Comput Meth
Progr Biomed. 106: 55–66, 2012.
Fernandez-Blanco E, Rivero D, Rabunal J, Dorado J,
Pazos A, Munteanu C. Automatic seizure detection
based on star graph topological indices. J Neurosci
Methods. 209: 410–419, 2012.
Gannous AS, Elhaddad YR. Improving an Articial
Neural Network Model to Predict Thyroid Bending
Protein Diagnosis Using Preprocessing Techniques.
WASET. 50: 124–128, 2011.
Havel J, Peña E, Rojas-Hernández A, Doucet J, Panaye
A. Neural networks for optimization of high-
performance capillary zone electrophoresis methods.
J Chromatogr A. 793: 317–329, 1998.
Ho W-H, Lee K-T, Chen H-Y, Ho T-W, Chiu H-C.
Disease-free survival after hepatic resection in
hepatocellular carcinoma patients: a prediction
approach using articial neural network. PloS One.
7: e29179, 2012.
Karabulut E, Ibrikçi T. Effective diagnosis of coronary
artery disease using the rotation forest ensemble
method. J Med Syst. 36: 3011–3018, 2012.
Kheirelseid E, Miller N, Chang K, Curran C, Hennessey
E, Sheehan M, Newell J, Lemetre C, Balls G, Kerin
M. miRNA expressions in rectal cancer as predictors
of response to neoadjuvant chemoradiation therapy.
Int J Colorectal Dis. 2012.
Leon BS, Alanis AY, Sanchez E, Ornelas-Tellez F, Ruiz-
Velazquez E. Inverse optimal neural control of blood
glucose level for type 1 diabetes mellitus patients. J
Franklin I. 349: 1851–1870, 2012.
Li Y, Rauth AM, Wu XY. Prediction of kinetics of
doxorubicin release from sulfopropyl dextran
ion-exchange microspheres using articial neural
networks. Eur J Pharm Sci. 24: 401–410, 2005.
Mazurowski M, Habas P, Zurada J, Lo J, Baker J,
Tourassi G. Training neural network classiers for
medical decision making: the effects of imbalanced
datasets on classication performance. Neural
networks. 21: 427–436, 2008.
Michalkova V, Valigurova A, Dindo M, Vanhara J.
Larval morphology and anatomy of the parasitoid
Exorista larvarum (Diptera: Tachinidae), with
an emphasis on cephalopharyngeal skeleton and
digestive tract. J Parasitol. 95: 544–554, 2009.
Molga E, van Woezik B, Westerterp K. Neural networks
for modelling of chemical reaction systems with
complex kinetics: oxidation of 2-octanol with nitric
acid. Chem Eng Process. 39: 323–334, 2000.
Mortazavi D, Kouzani A, Soltanian-Zadeh H.
Segmentation of multiple sclerosis lesions in MR
images: a review. Neuroradiology. 54: 299–320,
2012a.
Mortazavi D, Kouzani AZ, Soltanian-Zadeh H.
Segmentation of multiple sclerosis lesions in MR
images: a review. Neuroradiology. 54: 299–320,
2012b.
Murarikova N, Vanhara J, Tothova A, Havel J. Polyphasic
approach applying articial neural networks,
molecular analysis and postabdomen morphology to
West Palaearctic Tachina spp. (Diptera, Tachinidae).
Bull Entomol Res. 101: 165–175, 2010.
Narasingarao M, Manda R, Sridhar G, Madhu K, Rao A.
A clinical decision support system using multilayer
perceptron neural network to assess well being in
diabetes. J Assoc Physicians India. 57: 127–133,
2009.
Özbay Y. A new approach to detection of ECG
arrhythmias: Complex discrete wavelet transform
based complex valued articial neural network. J
Med Syst. 33: 435–445, 2009.
Pace F, Savarino V. The use of articial neural network
in gastroenterology: the experience of the rst 10
years. Eur J Gastroenterol Hepatol. 19: 1043–1045,
2007.
Rodrguez Galdn B, Peña-Méndez E, Havel J,
Rodrguez Rodrguez E, Daz Romero C. Cluster
Analysis and Articial Neural Networks Multivariate
Classication of Onion Varieties. J Agric Food
Chem: 11435–11440, 2010.
Saghiri M, Asgar K, Boukani K, Lot M, Aghili H,
Delvarani A, Karamifar K, Saghiri A, Mehrvarzfar P,
58
Amato et al.: Articial neural networks in medical diagnosis
Garcia-Godoy F. A new approach for locating the
minor apical foramen using an articial neural
network. Int Endod J. 45: 257–265, 2012.
Shankaracharya, Odedra D, Samanta S, Vidyarthi
A. Computational intelligence in early diabetes
diagnosis: a review. Rev Diabet Stud. 7: 252–262,
2010.
Siristatidis C, Chrelias C, Pouliakis A, Katsimanis
E, Kassanos D. Articial neural networks in
gyneacological diseases: Current and potential
future applications. Med Sci Monit. 16: 231–236,
2010.
Spelt L, Andersson B, Nilsson J, Andersson R. Prognostic
models for outcome following liver resection for
colorectal cancer metastases: A systematic review.
Eur J Surg Oncol. 38: 16–24, 2012.
Strike P, Michaeloudis A, Green AJ. Standardizing
clinical laboratory data for the development of
transferable computer-based diagnostic programs.
Clin Chem. 32: 22–29, 1986.
Szolovits P, Patil RS, Schwartz W. Articial Intelligence
in Medical Diagnosis. Ann Intern Med. 108: 80–87,
1988.
Tate A, Underwood J, Acosta D, Julià-Sapé M, Majs
C, Moreno-Torres A, Howe F, van der Graaf M,
Lefournier V, Murphy M, Loosemore A, Ladroue
C et al. Development of a decision support system
for diagnosis and grading of brain tumours using in
vivo magnetic resonance single voxel spectra. NMR
Biomed. 19: 411–434, 2006.
Thakur A, Mishra V, Jain S. Feed forward articial
neural network: tool for early detection of ovarian
cancer. Sci Pharm. 79: 493–505, 2011.
Trajanoski Z, Regittnig W, Wach P. Simulation studies
on neural predictive control of glucose using the
subcutaneous route. Comput Meth Progr Biomed.
56: 133–139, 1998.
Uğuz H. A biomedical system based on articial neural
network and principal component analysis for
diagnosis of the heart valve diseases. J Med Syst.
36: 61–72, 2012.
Verikas A, Bacauskiene M. Feature selection with
neural networks. Pattern Recogn Lett. 23: 1323–
1335, 2002.
Wilding P, Morgan M, Grygotis A, Shoffner M, Rosato
E. Application of backpropagation neural networks
to diagnosis of breast and ovarian cancer. Cancer
Lett. 77: 145–153, 1994.
Yan H, Zheng J, Jiang Y, Peng C, Xiao S. Selecting
critical clinical features for heart diseases diagnosis
with a real-coded genetic algorithm. Appl Soft
Comput. 8: 1105–1111, 2008.
Zupan J, Gasteiger J. Neural networks in chemistry and
drug design. Wiley VCH, Weinheim, 380 p. 1999.
... ANN is used in medicine in fields like diagnostic systems, image analysis, biochemical analysis, and development of drugs, etc. It is used to deal with heart and cancerrelated problems., analyze samples of urine and blood, glucose levels tracking, find levels of ion in body fluids, detecting tuberculosis, treating AIDS and cancer, etc. [6]. After mapping, there will be a single neuron called the winning neuron in some types of ANN. ...
Article
Full-text available
Character recognition (CR) from an image of text is challenging research in pattern recognition and image processing. New CR systems use artificial neural network (ANN) methods embedded in commercially available software. However, with the rising cost of software, a research revolution in CR is becoming limited. In this work, a CR system is developed using open-source and free software, SciLab. It is the most desirable choice than other compensated software. CR experiments have been done using ANN. The topologies of the neural network varied to recognize ten numerals. The neural network is applied to classify the character with the online backpropagation algorithm by changing the weights for each input online. The results reveal a lower error and the system’s accuracy of 99.92%. With standard backpropagation (batch version) while varying weights after a particular batch. An error is comparatively more, and the system’s output accuracy of 99.62% for the same topology. The application of pre-processing techniques to the given images with topology optimization. The image recognition accuracy is increased by 100%. The system provided optimum results with a topology of 135–100-10. So, the online backpropagation algorithm is more accurate than the standard batch version and should be adopted. Other CR research models can be developed with the SciLab Toolboxes at no cost and with maximum system accuracy.
... Thus, the vast computing resources available at the cloud layer can be brought to bear on the masses of health data that are collected over time [122]. Application-specific machine learning and AI models can be trained and implemented at the cloud level, and also deployed as instances at the fog layer to generate real-time health insights [95,94,123]. ...
Article
Situated at the intersection of technology and medicine, the Internet of Things (IoT) holds the promise of addressing some of healthcare's most pressing challenges, from medical error, to chronic drug shortages, to overburdened hospital systems, to dealing with the COVID-19 pandemic. However, despite considerable recent technological advances, the pace of successful implementation of promising IoT healthcare initiatives has been slow. To inspire more productive collaboration, we present here a simple—but surprisingly underrated—problem-oriented approach to developing healthcare technologies. To further assist in this effort, we reviewed the various commercial, regulatory, social/cultural, and technological factors in the development of the IoT. We propose that fog computing—a technological paradigm wherein the burden of computing is shifted from a centralized cloud server closer to the data source—offers the greatest promise for building a robust and scalable healthcare IoT ecosystem. To this end, we explore the key enabling technologies that underpin the fog architecture, from the sensing layer all the way up to the cloud. It is our hope that ongoing advances in sensing, communications, cryptography, storage, machine learning, and artificial intelligence will be leveraged in meaningful ways to generate unprecedented medical intelligence and thus drive improvements in the health of many people.
... Networks trained with different biomarkers as input and the respective medical diagnosis as output could be utilised particularly for the diagnosis of rare and thus often overlooked diseases. The applicability of such diagnosis models has been evaluated for acute nephritis, abnormal cardiac behaviour, carcinoma, valve stenosis and other diseases [115,116]. ...
Article
Full-text available
Neural networks for deep-learning applications, also called artificial neural networks, are important tools in science and industry. While their widespread use was limited because of inadequate hardware in the past, their popularity increased dramatically starting in the early 2000s when it became possible to train increasingly large and complex networks. Today, deep learning is widely used in biomedicine from image analysis to diagnostics. This also includes special topics, such as forensics. In this review, we discuss the latest networks and how they work, with a focus on the analysis of biomedical data, particularly biomarkers in bioimage data. We provide a summary on numerous technical aspects, such as activation functions and frameworks. We also present a data analysis of publications about neural networks to provide a quantitative insight into the use of network types and the number of journals per year to determine the usage in different scientific field.
... It's found that ANN to be often used for prediction, classification, and clinical diagnosis in area of telemedicine, organizational behavior, and cardiovascular. Amato et al. [26] describes ANN is a powerful framework to assist physicians and other enforcement. ANN has proven suitable for various diseases and their use makes the diagnosis analysis more reliable and consequently increases patient satisfaction. ...
Chapter
Full-text available
A current boom in the modeling intelligence in algorithm to solve complex applications , this intelligence could be achieved through natural and biological intelligence, resulted a technology known as intelligent systems, these algorithms use soft computing tools. AI aim to make the machines and computers smarter, that make a computer to mimic like human brain in specific applications. AI algorithms are a blend of many research areas, such as biology, sociology, philosophy, and computer science. The purpose of AI is not to substitute human beings, instead offer us a more prevailing tool to support in our work, provide more computing ability, permitting them to exhibit more intelligent behavior. 6.2 Computational Intelligence CI is a fragment of AI, which deals with study of adaptive mechanisms to enable intelligent behaviour in complex changing environment Figure 6.1. Shows all paradigm. Three main columns of CI are Fuzzy Systems, Neural Networks, and Evolutionary Computation, individual components have certain weakness and these could be improved by combining them tougher and they are referred as hybrid CI. Computers learn specific tasks from diverse forms of data or experimental observation , the ability to make computers learn and adapt is usually referred to as CI. It is considered to have the ability of computational adaptation, high computational speed, and fault tolerance. Computational adaptation means the ability of a system to adapt the changes happens in its input and output instances. 6.2.1 Difference between AI and CI AI deals with study of intelligent behavior exhibited by machines, mimicking natural intelligence similar to humans, AI aims to develop an intelligent machine which can think, act, and take decision similar to human. CI is the study of adapting and building intelligent behaviors based on a changing complex environment. CI is to recognize the computational model which make intelligent behavior of artificial and c06.indd 101 c06.indd 101
Preprint
The design of a neural network is usually carried out by defining the number of layers, the number of neurons per layer, their connections or synapses, and the activation function that they will execute. The training process tries to optimize the weights assigned to those connections, together with the biases of the neurons, to better fit the training data. However, the definition of the activation functions is, in general, determined in the design process and not modified during the training, meaning that their behavior is unrelated to the training data set. In this paper we propose the definition and utilization of an implicit, parametric, non-linear activation function that adapts its shape during the training process. This fact increases the space of parameters to optimize within the network, but it allows a greater flexibility and generalizes the concept of neural networks. Furthermore, it simplifies the architectural design since the same activation function definition can be employed in each neuron, letting the training process to optimize their parameters and, thus, their behavior. Our proposed activation function comes from the definition of the consensus variable from the optimization of a linear underdetermined problem with an $L_p^q$ regularization term, via the Alternating Direction Method of Multipliers (ADMM). We define the neural networks using this type of activation functions as $pq-$networks. Preliminary results show that the use of these neural networks with this type of adaptive activation functions reduces the error in regression and classification examples, compared to equivalent regular feedforward neural networks with fixed activation functions.
Chapter
People who can use critical and creative thinking to solve problems as a group are in high demand today and tomorrow. The way knowledge is acquired, constructed, and communicated has undergone radical change as a result of technological advancements. It's debatable whether education can produce critical and creative thinkers who can meet the demands of today's social and economic world and those of the future. Computers and smart devices, on the other hand, put students' learning at risk by undermining the authority of teachers in the classroom. This has led to the use of terms like guide, facilitator, and coach in place of the word teacher. Schools are well-known for being children's learning environments. However, it's unclear how much they actually learn or how much of it is aided by modern technology. In an era where people are constantly exposed to technology at work, school, and elsewhere, smart devices and technological tools have advanced far too quickly. Education research and pedagogical approaches that incorporate education technologies have progressed faster than the advancements in the everyday technological devices that we use. Thus, utilizing technologies in education has the potential to ensure innovation in educational activities. The goal of this research is to demonstrate that educational innovation must be handled with care. If you'd like to create innovative learning environments, you'll need to review previous studies on innovation as a pre-requisite and revise your strategies for successfully adapting technology to the field of education. To summarize, innovation is critical in reshaping and reconstructing learning environments, curricula, the teacher's role, and teacher training.
Chapter
Digital technology has had a profound and long-lasting impact on organizations. Digitalization is reshaping organizations, the workplace, and processes in the same way that movable type printing did in the 1800s, posing new problems for leaders to solve. Scholars in the social sciences have been working to unravel the complexities of this complex phenomenon, but their findings have been dispersed and fragmented across different fields, with no clear picture emerging. As a result of this gap in the literature, and in order to promote greater clarity and alignment in the academic debate, this paper examines the contributions made by studies on leadership and digitalization, identifying common themes and findings across various social science disciplines, such as management and psychology. In addition to defining key terms and concepts, it also highlights the most important theories and conclusions reached by academics. As a result, it distinguishes between categories that group papers according to the macro level of analysis (e-leadership and organization), the micro level of analysis (leadership skills in the digital age, and practices for leading virtual teams), and the macro level of analysis (ethical and social movements). Researchers found that leaders are crucial to the development of digital culture because they need to build relationships with numerous and dispersed stakeholders while also focusing on enabling collaborative processes in complex settings while also attending to pressing ethical concerns. A major contribution of this study is that it offers an extensive and systematic review of the digital transformation debate, as well as identifying important future research opportunities to advance knowledge in this field.
Thesis
Full-text available
Die Synopsis setzt sich auseinander mit dem Einsatz von Künstlicher Intelligenz (Maschinelles Lernen) im Kontext biomechanischer Daten. Potentiale der Methoden werden herausgearbeitet und ausgewählte praxisrelevante Limitationen anhand von fünf Publikationen adressiert. Unter anderem können durch Verwendung von Ensemble Feature Selection, Explainable Artificial Intelligence und Metric Learning sowie die Entwicklung eines pathologieunabhängigen Klassifikators vielversprechende Perspektiven aufgezeigt werden.
Article
Full-text available
Insulin resistance (IR) is one of the most widespread health problems in modern times. The gold standard for quantification of IR is the hyperinsulinemic-euglycemic glucose clamp technique. During the test, a regulated glucose infusion is delivered intravenously to maintain a constant blood glucose concentration. Current control algorithms for regulating this glucose infusion are based on feedback control. These models require frequent sampling of blood, and can only partly capture the complexity associated with regulation of glucose. Here we present an improved clamp control algorithm which is motivated by the stochastic nature of glucose kinetics, while using the minimal need in blood samples required for evaluation of IR. A glucose pump control algorithm, based on artificial neural networks model was developed. The system was trained with a data base collected from 62 rat model experiments, using a back-propagation Levenberg-Marquardt optimization. Genetic algorithm was used to optimize network topology and learning features. The predictive value of the proposed algorithm during the temporal period of interest was significantly improved relative to a feedback control applied at an equivalent low sampling interval. Robustness to noise analysis demonstrates the applicability of the algorithm in realistic situations.
Article
Artificial Neural Networks (ANN) is a popular artificial intelligence model used to acquire knowledge from datasets in different domains by applying learning techniques which work as estimators between the available inputs and the desired outputs. Using this aspect of ANN, the diagnosis of a disease can in time be predicted from the operating data with accuracy. But, the accumulated operating data used in ANN training may contain corrupt and noisy data records. Therefore, to enhance the reliability of the trained ANN, a data preprocessing technique is necessary for preparing the training and testing data set. In this study, several data preprocessing steps are applied to a Thyroid data set before starting the training process to build an expert medical diagnosis system. Several experiments were done and the results show good improvement in predicting the diagnosis of protein bending in the thyroid.
Article
Artificial neural networks (ANN) offer attractive possibilities for providing non-linear modeling of response surfaces and optimization in capillary zone electrophoresis (CZE) when the underlying mechanisms are very complex or not well known or understood, in comparison with (non)-linear regression methods. The application of ANN in optimization of CZE methods has been examined and a new method, based on the combination of experimental design and ANN methods, which offers considerable effectiveness, has been developed.
Conference Paper
This paper deals with blood glucose level control. Inverse optimal trajectory tracking for discrete time non-linear positive systems is applied. The scheme is developed for MIMO (multi-input, multi-output) affine systems. The control law calculates the subcutaneous insulin delivery rate in order to prevent hyperglycemia and hypoglycemia levels. A neural model is obtained from an on-line neural identifier, which uses a recurrent neural network, trained with the extended Kalman filter (EKF); this neural model has an affine form, which permits the applicability of inverse optimal control scheme. The proposed algorithm is tuned to follow a desired trajectory; this trajectory reproduces the glucose absorption of a healthy person. Simulation results illustrate the applicability of the control law in biological processes.
Article
Hypoglycemia or low blood glucose is dangerous and can result in unconsciousness, seizures and even death for Type 1 diabetes mellitus (T1DM) patients. Based on the T1DM patients’ physiological parameters, corrected QT interval of the electrocardiogram (ECG) signal, change of heart rate, and the change of corrected QT interval, we have developed a neural network based rule discovery system with hybridizing the approaches of neural networks and genetic algorithm to identify the presences of hypoglycemic episodes for TIDM patients. The proposed neural network based rule discovery system is built and is validated by using the real T1DM patients’ data sets collected from Department of Health, Government of Western Australia. Experimental results show that the proposed neural network based rule discovery system can achieve more accurate results on both trained and unseen T1DM patients’ data sets compared with those developed based on the commonly used classification methods for medical diagnosis, statistical regression, fuzzy regression and genetic programming. Apart from the achievement of these better results, the proposed neural network based rule discovery system can provide explicit information in the form of production rules which compensate for the deficiency of traditional neural network method which do not provide a clear understanding of how they work in prediction as they are in an implicit black-box structure. This explicit information provided by the product rules can convince medical doctors to use the neural networks to perform diagnosis of hypoglycemia on T1DM patients.
Article
In this paper, inverse optimal neural control for trajectory tracking is applied to glycemic control of type 1 diabetes mellitus (T1DM) patients. The proposed control law calculates the adequate insulin delivery rate in order to prevent hyperglycemia and hypoglycemia levels in T1DM patients. Two models are used: (1) a nonlinear compartmental model in order to obtain type 1 diabetes mellitus virtual patient behavior, and (2) a neural model obtained from an on-line neural identifier, which uses a recurrent neural network, trained with the extended Kalman filter (EKF); the last one allows the applicability of an inverse optimal neural controller. The proposed algorithm is tuned to track a desired trajectory; this trajectory reproduces the glucose absorption of a healthy person. The applicability of the proposed control scheme is illustrated via simulations.
Article
Application of neural networks to model the conversion rates of a heterogeneous oxidation reaction has been investigated — oxidation of 2-octanol with nitric acid has been considered as a case study. Due to a more complex and unknown kinetics of the investigated reaction the proposed approach based on application of neural networks is an efficient and accurate tool to solve modelling problems. The elaborated hybrid model as well as the modelling procedure have been described in detail. Learning data used to train the networks have been extracted from the experimental results obtained in an extensive investigation programme performed with a RC1 Mettler-Toledo reaction calorimeter. The influence of different operating conditions on the accuracy and flexibility of the obtained results has been investigated and discussed. It has been found that with the proposed approach the behaviour of a semi-batch reactor, i.e. its concentration and heat flow time profiles, can be predicted successfully within a singular series of experiments; however, the generalisation of the neural network approach to all series of experiments was impossible.
Article
Introduction: Neoadjuvant chemoradiation therapy has been shown to improve the outcome in patients with rectal cancer and is generally accepted as standard care; however, only selected patients would benefit from this treatment. We aimed to identify predictors of response to neoadjuvant chemoradiation therapy in colorectal cancer using formalin-fixed paraffin-embedded (FFPE) tissues as source of genetic materials and microarray analysis as investigation tool. Methods: After optimization of RNA extraction methods from FFPE, microarray analysis was carried out on total RNA extracted from 12 pre-treatment FFPE rectal tissues using Megaplex pool A. Microarray data were analysed using an artificial neural network algorithm. Statistical analysis and correlation with clinicopathological data was performed using SPSS software. Results: A distinct miRNA expression signature predictive of response to neoadjuvant CRT in 12 FFPE pre-treatment rectal cancer tissue samples was identified. These signatures consisted of three miRNA transcripts (miR-16, miR-590-5p and miR-153) to predict complete vs. incomplete response and two miRNA transcript (miR-519c-3p and miR-561) to predict good vs. poor response with a median accuracy of 100 %. Conclusion: Using microarray analysis of pretreatment FFPE rectal cancer tissues, we identified for the first time a group of miRNA predictors of response to neoadjuvant CRT. This, indeed, can lead to a significant improvement in patient selection criteria and personalized rectal cancer management.