=Paper=
{{Paper
|id=Vol-1839/MIT2016-p09
|storemode=property
|title= Decision support system for real-time diagnosis of musculoskeletal system
|pdfUrl=https://ceur-ws.org/Vol-1839/MIT2016-p09.pdf
|volume=Vol-1839
|authors=Anastasia Grecheneva,Igor Konstantinov,Oleg Kuzichkin,Nicolay Dorofeev
}}
== Decision support system for real-time diagnosis of musculoskeletal system==
https://ceur-ws.org/Vol-1839/MIT2016-p09.pdf
Mathematical and Information Technologies, MIT-2016 — Information technologies
Decision Support System for Real-Time
Diagnosis of Musculoskeletal System
Anastasia Grecheneva, Igor Konstantinov, Oleg Kuzichkin, and
Nokolay Dorofeev
Belgorod National Research University, Belgorod,
308015, 85 Pobedy st., Russia
1155464@bsu.edu.ru
http://www.bsu.edu.ru/bsu
Abstract. A construction principle of a technical system for diagnosis
and rehabilitation of the musculoskeletal system based on accelerometer
method, together with synchronization algorithms measuring patient pa-
rameters, is considered. The optimal accuracy estimations of the techni-
cal parameters of the accelerometric goniometer system are determined;
they are the sample rates of the accelerometer signal converters, the
required sensitivity of the sensor, etc. The advantages of the proposed
approaches to the construction of rehabilitation and diagnostic systems
of the musculoskeletal system are adaptability and reliability of the di-
agnoses.
Keywords: biomechanics, information system, goniometric control, ac-
celerometer, mathematical model.
1 Introduction
Accurate diagnosis and objective assessment of the treatment efficiency of mo-
tor function disorders to date remains one of the urgent problems of modern
traumatology and orthopedics. The large number of evaluation approaches and
techniques reveal a lack of reliability of the proposed criteria for diagnosis and
assessing recovery efficiency. For example, the diversity of human movement is
characterized by a number of parameters: torque, speed, complexity of trajecto-
ries, changes in the level neuromuscular and brain activity. In existing systems,
goniometry and diagnosis of musculoskeletal system mainly take into account
only the kinematic parameters of the skeletal system, regardless of the bone
structure and neurophysiological parameters of state of the patient [1]. There-
fore, in diagnosis, the study of the central control mechanisms of purposeful
physical activity is of great importance, as well as the parameters of the skeletal
system at the structural level [2,3].
2 Statistical basis of the goniometric measurements
Formation of the goniometric criteria and selection of the optimal working pa-
rameters of the system rehabilitation is carried out on the basis of statistical
84
Mathematical and Information Technologies, MIT-2016 — Information technologies
clinical studies of patients under normal conditions and in the presence of devia-
tions. In medical diagnostics, the assessment of the angular movement indicators
and their permissible deviation from the normal ones is done according to the
joints motion estimation table “Regulations on military-medical examinations”
(approved by the Russian Federation Government Decree No 565 dated July 4,
2013) [4].
Table 1. Assessment of range of motion in the joints of the limbs
Joint Motion Norm,∘ Restriction of movement,∘
slight moderate considerable
Shoulder to flexion 180 179-135 134-100 <100
shoulder girdle abduction 180 179-135 134-100 <100
Shoulder extension 60 59-40 39-15 <15
(simple) internal rotation 90 89-45 44-20 <20
external rotation 90 89-45 44-20 <20
Elbow flexion 30 31-70 71-90 >100
(complex) abduction 180 179-150 149-120 <120
Combined elbow wrist pronation 90 89-45 44-20 <20
radial shoulder wrist supination 70 69-30 30-15 <15
flexion 105 106-145 146-160 >160
Carpal extension 115 116-150 149-165 >165
(combined) radial abduction 160 161-175 176-185 >185
ulnar abduction 140 141-155 154-180 >180
knee extension 90 91-120 121-150 >150
knee flexion 60 61-90 91-150 >150
Hip extension 140 141-160 161-170 >170
(simple) abduction 50 49-30 29-15 <15
internal rotation 35 34-25 24-15 <15
external rotation 45 44-25 24-15 <15
Knee flexion 135 134-90 89-60 <60
(complex) abduction 180 179-170 169-160 <160
Ankle flexion 130 129-120 119-100 <100
(complex) abduction 70 71-80 79-90 >90
On the basis of the data presented in Table 1, it can be said that indicator
deviations of joint angles by 1∘ are violations. Consequently, goniometric sys-
tem must meet the requirements of measurement accuracy, with the threshold
sensitivity of the measurement of mutual deviations and measuring range of mo-
tion being at least 1∘ . In this case, the measurement error must be smaller than
this threshold. It should be noted that at present mechanical goniometers are
widely used in medical diagnostics. Their accuracy does not meet the present
requirements stated by the system of this class. Low accuracy of the measured
parameters might be resulted from the design features of the device, and a high
85
Mathematical and Information Technologies, MIT-2016 — Information technologies
degree of subjectivity of diagnosis due to professional experience and the influ-
ence of the human factor [5].
3 Information and technical support of the automated
systems of diagnostics of the musculoskeletal system
The emergence of the movement results from neuro-cerebral human activity
aimed at the implementation of any function. Therefore, the main objective of
the design of goniometric systems is to assess the effectiveness of motor actions
with respect to the application efforts of their execution [6].
For a comprehensive solution of this problem, we are to study design aspects
of an automated diagnostic system of human musculoskeletal apparatus. The
design is based on the synthesis of adequate informative physiological methods
such as goniometry (accelerometer), computed tomography, electroencephalog-
raphy (EEG) and electroneuromyography (ENMG). For this purpose, a medico-
technical basis was formed. A block diagram of an automated system of complex
real time diagnostics of the musculoskeletal system has been developed. The di-
agram consists of the circuit of accelerometric goniometer, an electroencephalo-
graph and electromyography imaging is used as well (Fig. 1).
Synchronization
Neural
С alibration network
Electroencephalogram
EEG
Neuromuscular options
Signal Mathematical
Electroneuromyography Brain parameters Measurements Model DB
Patient ENMG processing model
DB
unit Goniometer angles
Diagnosis
Acceleration knowledge base
Accelerometers
Neural Evoked potential Neural
Test
network processing network
DSS
Actuators
Diagnosis
Test Evoked Decease
methods potential DB
DB
Fig. 1. A block diagram of hardware and software of the automated system of gonio-
metric control
The above structure includes a measuring unit functioning in real time. It
consists of a chain of inverters biokinematic driving parameters of the locomotor
apparatus of man (accelerometric goniometer), recording units of psycho- and
neurophysiological parameters (EEG and ENMG), and the registration equip-
ment for bone and structural parameters (tomography).
The synchronous processing of the recorded parameters form time series,
which are visualized with various degree of detail. The time series are the basis
of a model of patient (Fig. 2). The model is processed by a neural network
86
Mathematical and Information Technologies, MIT-2016 — Information technologies
and is stored in the model data base. The model that most closely matches
the time series are instantiated. Pain thresholds and threshold of sensitivity
of the patient for generating control signals to the actuators are determined
by neural network algorithms. This is possible via a feedback method, whose
implementation is carried out by the processing unit of evoked potentials of brain
(patient’s reactions to test stimuli). Based on the processed data, operation mode
of the actuators is generated and selected from a database of test techniques.
h, cm
Biometric parameters
m, kg
αi
Accelerometric
αi , βi , γi
parameters
d
Os , mkV Dunamic model
P sd, mkV of the patient
Electroencephalography C sd, mkV
F sd, mkV
CPBm, uV
Electroneuromiography CPBs, uV
F-wave, uV
Fig. 2. The dynamic information model of patient
Thanks to neural network algorithms and decision support system (DSS)
based on databases of the time series, diseases and evoked potentials, an ap-
proximate diagnosis of patient is determined.
It should be noted that the above adaptive goniometric control system in-
cludes both stationary and mobile measuring systems [6]. The number of moni-
tored parameters is determined according to the severity of the patient’s pathol-
ogy. In the case of low severity injuries, it is sufficient to use of a several portable
accelerometric goniometers, guaranteeing the freedom of the patient’s move-
ment. In the presence of more serious violations in the functioning of the mus-
culoskeletal system is suspected, the accelerometric goniometers coupled with
electroneuromyograph (Fig. 3a), EEG data and computing tomography (Fig.
3b) is recommended for use.
It is shown that the dynamic activity of brain neurons relating to the imple-
mentation of tool movements, typically starts 50-150 ms. prior to the occurrence
of EMG activity and ended after a traffic stop. Thus, the joint reaches equi-
librium during the dynamic development of the motor cortex neuronal activity
phase long before the establishment of steady equilibrium level of neural activ-
ity. The maximum value of the mean frequency of neural activity in one bin of
duration is 50 ms. In dynamic phase, reactions of neurons did not correlate with
the magnitude of the equilibrium steady-articular angle (Fig. 5). At the same
time, a positive correlation was revealed between the average frequency of neural
87
Mathematical and Information Technologies, MIT-2016 — Information technologies
Fig. 3. Diagnostic biomechanics of the patient on the base of a physical exercise; a)
the patient’s posture; b) tomographic image
activity in the whole dynamic phase and magnitude response of articular angle
[7, 8].
The presented results show that the maximum level of motor neuron activ-
ity depends primarily on the joint movement velocity and the duration of the
movement. Thus, the obtained data contribute to definition of the criteria of
permissible values, characterizing limits of the patient physiological parameters
with respect to the normal ones; the limits are determined according to the de-
gree of deviation and the conditions of pathology. Neurophysiological criteria
are also formed based on statistical analysis of clinical studies of patients under
normal conditions and in the presence of deviations.
4 The DSS for setting of a diagnosis
Development of DSS based on the dynamics analysis of the recorded time series
for goniometric, kinematic and neurophysiological parameters is a complex and
multicritelial problem. The algorithms of the DSS are based on Bayes’ rule. The
rule accounts various heterogeneous types of input data expressing many kinds of
deceases of the musculoskeletal system and a large number of symptoms. Bayes’
rule in a generalized form is as follows [9]:
𝑃 (𝑆1 ∩ · · · ∩ 𝑆𝑘 | 𝑑) · 𝑃 (𝑑)
𝑃 (𝑑 | 𝑆𝑘 ∩ · · · ∩ 𝑆1 ) = , (1)
𝑃 (𝑆1 ∩ · · · ∩ 𝑆𝑘 )
where 𝑃 (𝑑) is a priori probability of the diagnosis 𝑑, and 𝑆1 . . . 𝑆𝑘 are functional
physiological parameters.
88
Mathematical and Information Technologies, MIT-2016 — Information technologies
Fig. 4. Interpretation of results. Angle support limb (40 degrees) is not greater than
(equal to) the angle of the working limb (40 degrees). The right lower limb: abduction
angle in the hip joint is 40 degrees, the average EMG gluteus 1 is 563.6 mV. The left
lower limb: abduction angle in the hip joint is 40 degrees, the average EMG gluteus
is 893.5 mV.; muscle operation mode stabilizing. The ratio of the two coefficients of
reciprocity for medium gluteus is 1.75.
Impulse/s Impulse/s Impulse/s
N=21 N=25 N=29
40º
1 2 3
ms ms ms
2000 0 2000 4000 2000 0 2000 4000 2000 0 2000 4000
Fig. 5. The activity of neurons of the motor cortex and variations of the articular angle
measured during the flexion movements at different joint velocity. Legend: 40∘ is the
variation of the articular angle. The line parallel to the y-axis is the average frequency
of pulses in the bin. The line parallel to the x-axis indicates the average frequency
of the background activity of the neuron. Vertical lines indicate the boundaries of
dynamic and stationary phases of motion. 1,2,3 are joint flexion speeds, N is number
of iterations.
89
Mathematical and Information Technologies, MIT-2016 — Information technologies
This formula requires (𝑚 · 𝑛)2 + 𝑚2 + 𝑛2 calculations of probability estimates,
where 𝑚 is the number of possible diagnoses, and 𝑛 is the number of different
variations. In order to calculate the total probabilities 𝑃 (𝑆1 ∩ · · · 𝑆𝑘 ), we are to
calculate 𝑃 (𝑆1 /𝑆2 ∩ · · · ∩ 𝑆𝑘 ) · 𝑃 (𝑆2 /𝑆3 ∩ · · · ∩ 𝑆𝑘 ) · . . . · 𝑃 (𝑆𝑘 ).
Therefore, the model 𝑃 for automated diagnostic expert system will be based
on
𝑃 (𝑆 | 𝑑) · 𝑃 (𝑑)
𝑃 (𝑑 | 𝑆) = ¯ · 𝑃 (𝑑)¯. (2)
(𝑃 (𝑆 | 𝑑) · 𝑃 (𝑑) + 𝑃 (𝑆 | 𝑑)
The probability of the hypothesis 𝑑 in the presence of certain abnormalities in
the recorded data 𝑆 is calculated based on the prior probability of the hypothesis
without confirming abnormalities and the likelihood of having abnormalities in
conditions that hypothesis is correct (event 𝑑) or incorrect (event 𝑑¯ ). Therefore,
for the problem of diagnosis of diseases of the musculoskeletal system, it appears
that
𝑃𝑦𝑒𝑠 · 𝑃 (𝑑)
𝑃 (𝑑 | 𝑆) = ¯ . (3)
(𝑃𝑦𝑒𝑠 · 𝑃 (𝑑) + 𝑃𝑛𝑜𝑡 · 𝑃 (𝑑))
Let the pathology probability 𝑃 (𝑑) be equal to 𝑃 . The program generates a
condition (parameters in the presence of pathology) and calculates the proba-
bility 𝑃 (𝑑 | 𝑆) depending on the it’s implementation. The answer “Yes” (𝑃𝑦𝑒𝑠 )
confirms the above calculations, the answer “No” (𝑃𝑛𝑜 ) does it too but with
probability (1 − 𝑃𝑦𝑒𝑠 ), and (1 − 𝑃𝑛𝑜 ) instead 𝑃𝑦𝑒𝑠 and 𝑃𝑛𝑜 . Thereafter, the a pri-
ori probability 𝑃 (𝑑) is replaced with 𝑃 (𝑑 | 𝑆). The program execution is cyclic,
with the constant value 𝑃 (𝑑) refining at each iteration. The general scheme of
the diagnosis selection algorithm is shown in Fig. 6.
The diagnosis selection algorithm structure consists of several branches:
Step 1. Enter the input data – a set of biometric, goniometric, neurophysio-
logical and structural parameters; then the program retrives information on the
number of the diseases recorded having the corresponding symptoms from in the
database (𝑁 is the number of relevant deviations disease, 𝑛 is the number of the
disease in question: 0 6 𝑛 6 𝑁 ).
Step 2. Set counter of a disease incrementally from the initial state 𝑛 = 0 till
𝑛 6 𝑁.
Step 3. Traverse all the a priori probabilities 𝑃 (𝑑), relating to the input data
set and to the selected disease, to prioritize detected deviations. Deviations with
the minimal likelihood are excluded from the probability set (𝐽 is the selected
number of deviations in the set, 𝑗 the number of the current deviation 0 6 𝑗 6 𝐽).
Step 4. Set the deviation counter incrementally from the initial state 𝑗 = 0
to 𝑗 6 𝐽.
Step 5. Select the deviation with the greatest probability of presence.
Step 6. Evaluate the degree of reliability of the diagnosis according to the
interval [−5, +5] (a scale). If the value belongs to the interval, then the program
calculates the proportion of the degree of affiliation to a particular diagnosis
parameters, using the corresponding weighting coefficients.
Step 7. Poll of the counter of registered deviations. If there is no new events
of a decease then process the next unprocessed selected departure, go to step 4.
90
Mathematical and Information Technologies, MIT-2016 — Information technologies
Begin
Enter the initial information:physiological
parameters, the number of diseases (N)
n=n+1, 0≤n≤N
Viewing probability P(d), with the
exception of deviations minimum
probability
j=j+1, 0≤J≤N
Selection of deviation j with max value of
a priori probability of the formation of the
corresponding issue
yes
Answer «-5»
no
yes
Answer «+5»
no
yes
Answer «0»
no
no
j=J
Selecting the max value of an array yes
of prior probabilities of diagnoses
and the formation of an advisory Values calculated probabilities
diagnosis based on the works of deviations
probabilities
Selecting the most probable
diagnosis
yes no
n=N
End
Fig. 6. The scheme of algorithm of diagnosis selection
91
Mathematical and Information Technologies, MIT-2016 — Information technologies
Step 8. Figure out new probabilities in Bayes rules. Specify the minimal and
maximal values of the probabilities for each disease based on the currently ex-
isting a priori probabilities and assumptions that the remaining evidence will
speak in favor of the hypothesis or contradict it. This step calculates the total
conditional probabilities for each deviations. The hypotheses whose the mini-
mal values are above a certain thresholds are considered as possible outcomes
(possible diseases) and are subject to further diagnosis.
Step 9. Check the counter of the registered diseases.
Step 10. Sort the outcome list according to the probabilities, display sub-
set with the maximal probability values as a recommendation for the diagnosis
mentioned symptoms to physician.
Step 11. Display the recommended diagnosis.
Medical information system, which implements this algorithm, produces a
finite set of the recommendations for doctor, emphasizing the presence of de-
viations registered with the diagnostics system sensors. Limiting the selection
decision by a finite set of possible diagnoses is to reduce the probability of setting
wrong preliminary diagnosis, eliminate human factor and increase the objectivity
of the diagnosis of diseases.
5 DSS based on fuzzy logic and artificial neural networks
In order to develop the control unit of diagnostics system of the musculoskeletal
system, we propose a method of computer support of diagnosis setting based on
fuzzy logic and artificial neural networks. The method is represented in the form
of two structural units: decision-making unit and knowledge base.
The decision (a diagnosis) is produced in two stages [10]. At the first stage
– the preliminary diagnosis–, the system determines in advance the possible
pathology and generate diagnostic recommendations based on data from the
medical records and X-ray images. Then, at the stage of the goniometric diagno-
sis, together with EMNG, the recommendations are confirmed or rejected on the
basis of the analysis of the obtained information, with immediate neural network
processing by the diagnosis system.
The advantage of fuzzy logic is the ability to describe the operation of the
system by means of fuzzy production rules (FPR) [11, 12]. The initial values of
the parameters (used in FPR) for the normal cases or the pathological ones are
determined at the beginning on the base of experts’ opinions. The values are
adjusted with neural network engines.
A distinctive feature of this set of rules is the allocation of a separate group
of so-called factors of pain diagnosed by EEG. Pain is one of the most important
factors in the diagnosis of diseases of the musculoskeletal system. On the basis
of these features we justify the choice of the rules of fuzzy productions of the
form:
IF AND ... AND
THEN AND ... AND
92
Mathematical and Information Technologies, MIT-2016 — Information technologies
Necessary linguistic input variables (𝐼𝑁𝐿𝑉 ) are selected according to a sta-
tistical base of the analysis of medical records and printed materials [13].
Membership functions (MF) 𝜇 of a strict magnitude to a fuzzy term set (cor-
responding to one of the values of the input linguistic variables) were determined
by means of the following expert evaluation techniques. Let expert 𝐸1 believe
that the specific value of 𝑥* belongs to the fuzzy term set at 𝑎1 6 𝑥* 6 𝑏1 ;
expert 𝐸2 at 𝑎2 6 𝑥* 6 𝑏2 ; . . . ; 𝐸𝑔 expert at 𝑎𝑔 6 𝑥* 6 𝑏𝑔 . Then the term of
MF 𝜇 is obtained in the form shown in Figure 7.
̀
In LV 1 LOut
0 min(ai) max(a i) min(b i) max(b i)
i i i i
Fig. 7. The membership function of a term. 𝐼𝑁𝐿𝑉 is input linguistic variable, 𝐿𝑂𝑢𝑡 is
a linguistic output
In the figure, the horizontal axis shows the value of strict variable under
fuzzification, where 𝑖 = 1, 𝑔 is an expert number. The vertical axis displays
𝜇 fraction of all the experts who believe that its value of a 𝑥 belongs to this
linguistic value of the linguistic variable. This initially plots the MF’s, which is
obtained at 𝑥 ∈ (𝑚𝑖𝑛(𝑎𝑖 ), 𝑚𝑎𝑥(𝑎𝑖 )) ∪ (𝑚𝑖𝑛(𝑏𝑖 ), 𝑚𝑎𝑥(𝑏𝑖 )); the plot is curvilinear
and will get a linear form if the least squares method is used.
The next stage is the aggregation construction applied to the RFP’s whose
terms contain more than one sub-conditions. The conditional part of the rules
is as follows:
IF OR ... OR
AND OR ... OR
Each of the 𝑛 terms OR ... OR con-
sists of 𝑚 subconditions , where VALUEij is the 𝑗-th value of
the 𝑖-th 𝐿𝑉 in subconditions. Its number is determined by the number of input
values 𝐿𝑉 : 𝑖𝑗. Let the truth degree of subconditions with the number 𝑖𝑗 be, re-
spectively, 𝜇𝑖𝑗 . The following RFP matrix 𝑀 is formed for all the subconditions:
⎛ ⎞
𝜇11 𝜇12 · · · 𝜇1𝑚
⎜ 𝜇21 𝜇22 · · · 𝜇2𝑚 ⎟
𝑀 = ⎜ ⎟
⎝ ... ... . . . ... ⎠ . (4)
𝜇𝑛1 𝜇𝑛2 · · · 𝜇𝑛𝑚
93
Mathematical and Information Technologies, MIT-2016 — Information technologies
Using this matrix at the 3-6-th stages, we get the formula for calculating
the confidence coefficient 𝜒 – the correctness precision of the system solution, –
which is calculated for each of the possible diseases identified by DSS.
∫︁
𝑚𝑖𝑛 (︂ ∑︀𝑞𝑢𝑣 )︂
𝑘=1 𝐹𝑘 · 𝑚𝑖𝑛 𝑖 (𝑚𝑎𝑥𝑗 (𝜇𝑖𝑗 (𝑥𝑖 )))
∑︀𝑞𝑢𝑣
𝜒𝑢 = 𝑦𝑢 · 𝑚𝑎𝑥𝑣 𝜇
˜𝑢𝑣 · 𝑑𝑦𝑢 /
𝑘=1 𝐹𝑘
𝑚𝑎𝑥
∫︁𝑚𝑖𝑛 (︂ ∑︀𝑞𝑢𝑣 )︂
𝑘=1 𝐹𝑘 · 𝑚𝑖𝑛𝑖 (𝑚𝑎𝑥𝑗 (𝜇𝑖𝑗 (𝑥𝑖 )))
∑︀𝑞𝑢𝑣
/ 𝑚𝑎𝑥𝑣 𝜇 ˜𝑢𝑣 · 𝑑𝑦𝑢 , (5)
𝑘=1 𝐹𝑘
𝑚𝑎𝑥
where 𝑚𝑖𝑛 and 𝑚𝑎𝑥 are the left and right limits of the carrier interval of fuzzy
set 𝐿𝑂𝑢𝑡𝜔𝑢 under consideration; 𝐹𝑘 are the weighting coefficients of the rules,
𝑘 = 1, 𝑞𝑢𝑣 ; 𝑞𝑢𝑣 is the number of RFP, which is determined in the consequent of
the 𝑢-th term of MF 𝐿𝑂𝑢𝑡𝑣; 𝜇 ˜𝑢𝑣 (𝑦𝑢 ) is the antecedent MF 𝑣-th term of 𝑢-th
MF 𝐿𝑂𝑢𝑡.
It should be noted that the weights of the rules vary depending on the oc-
currence of new facts and results of fuzzy inferences at the previous stages. To
resolve this uncertainty in form of an adjustment of RFP weights, an inference
system is represented as a hybrid, i.e., fuzzy neural network (Fig. 8). Its struc-
ture is identical to the multilayer network, but its layers correspond to the stages
of fuzzy inference, which has continuously carry out the following procedures:
- Input layer performs fuzzification function based on the specified input
membership functions;
- Output layer implements the defuzzification function;
- Hidden layers reflect the totality of the RFP and the output stages: aggre-
gation, activation and accumulation.
Inputs layer 1 layer 2 layer 3 layer 4 layer 5 layer 6 layer 7 layer 8 Outputs
q
∑ F
k-1 k
max
~
̀1 1
min ~
̀11 ~
̀
~
̀umax1 ~
̀umaxvmax
1vmax
min
q
~
̀umax1
∑F
k-1 k
max def χ
~
̀ 1vout
~ min
̀umaxvmax
def χ umax
Fig. 8. The structure of hybrid neural network
94
Mathematical and Information Technologies, MIT-2016 — Information technologies
Neurons 𝑚𝑖𝑛, 𝑚𝑎𝑥 and 𝑠𝑢𝑚 indicated in Fig. 8 act as appropriate mathemat-
ical functions. Neurons, marked with “×” is transmitted to the output product
of the input signals. Symbol “≡” marks neurons establishing
⋀︀ a correspondence
between 𝐿𝑉 and an intermediate 𝐿𝑂𝑢𝑡, and symbols “ ” are neurons realizing
operation fuzzification for each term of INLV. Node “𝑎/𝑏” denotes division of the
input value by the sum of the weights of active rules. Neuron “𝑑𝑒𝑓 ” implements
the function of defuzzification, with applying the method of gravity center.
The fuzzy rule selection engine is represented as INLV inputs having “0” (a
rule is selected) and “1” (a rule is not selected) logic levels weights multiplied
by the corresponding membership functions 𝜇𝑖𝑗 (𝑥𝑖 ). Here index 𝑖 ∈ 1, 𝑛 is the
number of 𝐼𝑁𝐿𝑉 and index 𝑗 ∈ 1, 𝑚 is the number of its term.
A neural network is trained by the algorithm of error back-propagation mod-
ified for use in the fuzzy neural networks. The layers of neurons with specified
parameters are represented by one layer with a complex activation function,
fuzzy artificial neural network (ANN) is identical a three-layer ANN with one
hidden layer. Thus, network training is reduced to a three-layer perceptron learn-
ing. It is worth noting that the fuzzy neural network is used only in the case
of changes of the DSS structure (change aggregate RFP, input or output LV),
and in the case of the appearance of new evidence proving or disproving the
previously known data in the literature or medical practice.
6 The results of research
This section is devoted to the results of the benchmark tests of the calculated
model; the results are obtained with an installed bodily machinery of the human
skeleton. During the investigation, we used the method of mechanical goniom-
etry in collaboration with an orthopedic doctor. The method of accelerometer
goniometry usage is proposed in [5].
Table 2. Average values of measurement errors for accelerometric and mechanical
goniometers
Motion pattern Error of an accelerometric Error of a mechanical
goniometer goniometer
Bending ±0,02 ±1,00
Abduction ±0,02 ±1,00
Extension ±0,02 ±1,00
Internal rotation ±0,02 ±1,00
External rotation ±0,02 ±1,00
Tremor imitation ±0,02 ±3,50
The results in Table 1 confirmed that the application of the accelerometic
goniometer improves the diagnosis accuracy in average 1∘ 44′ compared to me-
chanical goniometer.
95
Mathematical and Information Technologies, MIT-2016 — Information technologies
In addition, the diagnosis was carried on 20 patients who underwent reha-
bilitation after complicated shoulder injury and wrist. It should be noted that
in all cases the diagnosis was a medical opinion on the normal rates. However,
based on the neural network processing of the recorded data of the goniometric
control, the electroneuromyographic and the tomographic control, a dial of a
estimation of a motion in the joints from the Table 1, it was found that in 14
cases the medical diagnosis coincided with the diagnosis of DSS, in 4 cases were
diagnosed of a functional deviation of the work wrist in a small extent, and in
2 cases were more prominent deviation of the combining elbow-shoulder joint in
small extent.
7 Conclusion
Data processing algorithms and approaches to designing a system of diagnostics
of the musculoskeletal system are presented in the article. The obtained results
allow one to
- define the evaluation criteria of the “current state” of the musculoskeletal
system;
- determine the severity of biomechanical disorders with a high degree of
confidence;
- predict the biomechanical disorders of the musculoskeletal system;
- have the possibility of a science-based rehabilitation prognosis;
- create and optimize individual training programs that promote the advance-
ment of technical training of athletes and prevent the diseases.
The results show that the diagnosis of the functional state of the muscu-
loskeletal system based on the proposed system is an informative method of
detecting violations. This technique is recommended to be used as a supplement
to conventional methods of examination of the musculoskeletal system condition,
as well as a stand alone technique.
Acknowledgments. This work was supported by RFBR grant
16-38-00704 mol a.
References
1. Vitenzon, A.S., Petrushanskaya, K.A.: The concept of the use of artificial correction
movements in orthopedics, traumatology and prosthetics. In: Journal of Trauma-
tology and Orthopedics. Priorov. - 2003. - No 4. pp. 54-58.
2. Hecht, B.M., Kasatkina, L.F., Samoylov, M.I. et al.: Electromyography in the diag-
nosis of neuromuscular diseases. - Taganrog: TSURE, 1997. - 370 p.
3. Drew, T., Kalaska, J., Krouchev, N.: Muscle synergies during locomotion in the cat:
a model for motor cortex control // J Physiol. - 2008.- Vol. 586, No 5. - P. 1239-1245.
4. Resolution of the Russian Government dated on July 4, 2013 N 565 ”On approval
of the military-medical examination” http://docs.cntd.ru/document/499031795.
96
Mathematical and Information Technologies, MIT-2016 — Information technologies
5. Grecheneva, A.V., Kuzichkin, O.R., Dorofeev, N.V., Konstantinov, I.S.: The use of
the accelerometer in the goniometric measurement system. In: Information systems
and technologies, ISSN 2072-8964, No 4 (90) 2015 July-August, pp. 5-10.
6. Grecheneva, A.V., Dorofeev, N.V., Kuzichkin, O.R.: The use of the accelerometer
in the goniometric measurement systems. In: Mechanical engineering and life safety,
ISSN 2222-5285, No 1, 2015. pp 55-58.
7. Sarnadskiy, V.N., Fomichev, H.G., Monitoring of spinal deformity by computer op-
tical topography. - Benefit For Physicians Health Ministry. -Novosibirsk, NIITO
2001.
8. Zokirhodzhaev, M.A.: Diagnostic criteria and methods of rehabilitation therapy flat
feet in children. In: The doctor-graduate student, No 5.2 (54), 2012. - P. 234-239.
9. Kobrinsky, B.A.: The logic of the argument in the decision-making in medicine. In:
STI ser.2. - 2001. - No 9. - P. 1-8.
10. Podolny, M.A., Taperova, L.N.: Designing medical diagnostic system based on the
model of fuzzy inference. In: Eighth Nat. Conf. by Art. Intelligence with int. uch .:
mat. Conf. V.2. M .: FIZMATLIT, 2002. - P. 641-646.
11. Babkin, E.A., Kozyrev, O.R., Kurkina, I.V.:The principles and algorithms of artifi-
cial intelligence, Nizhny Novgorod: Nizhegorod. state. tehn. University Press, 2006.
132 p.
12. Chernov, A.V., Shtanko, S.I., Rogozin, M.A., Podvigin, S.N.: The use of mathe-
matical analysis for the diagnosis of borderline mental disorders in medical students.
In: Doctor-graduate student, No 1 (50), 2012. P. 43-49.
13. Choporov, O.N., Razinkin, K.A.: Optimization model of choice of the initial plan
of control actions for modeling information systems. In: Control Systems and Infor-
mation Technology, 2011. Vol. 46, N 4.1. P. 185-187.
97