<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Pre-pathological Diagnostic Criteria for Professional Burnout in Healthcare Workers</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Igor Zavgorodnii</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Iryna Perova</string-name>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Olena Litovchenko</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Tetiana Bahlaienko</string-name>
          <email>tv.bahlaienko@knmu.edu.ua</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Kharkiv National Medical University</institution>
          ,
          <addr-line>4 Nauky avenue, Kharkiv, 61022</addr-line>
          ,
          <country country="UA">Ukraine</country>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>Kharkiv National University of Radio Electronics</institution>
          ,
          <addr-line>14 Nauky avenue, Kharkiv, 61166</addr-line>
          ,
          <country country="UA">Ukraine</country>
        </aff>
      </contrib-group>
      <fpage>316</fpage>
      <lpage>324</lpage>
      <abstract>
        <p>work is devoted to identifying significant changes in the individual and psychological characteristics of medical workers with the initial level of professional burnout according the pathological states professional psychophysiological which implement preventive measures against the professional burnout development.</p>
      </abstract>
      <kwd-group>
        <kwd>criteria</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>questionnaire.</p>
      <p>28
respondents
(emergency
medical services staff: doctors, paramedics) took part in the questionnaire to determine the
level of professional burnout, as well as testing psychophysiological reactions of the
central nervous system for a comprehensive assessment of the respondents’ general
condition. The proposed mathematical approach enabled us to determine a group of
pre</p>
    </sec>
    <sec id="sec-2">
      <title>1. Introduction</title>
      <p>
        The dynamic development and intensification of production processes, the increase in the social
component of a person's environment is characterized by an additional psychological burden in
professions of the socionomic type, which activities are based on the "person-to-person" communication
[
        <xref ref-type="bibr" rid="ref1 ref2">1,2</xref>
        ]. The development of neuro-emotional stress during the intensive communication with a client,
constant provision of professional assistance, being in an emotionally charged atmosphere creates a state
of tension with changes in the psychophysiological parameters of the body. This can lead to the
development of professional stress, as a result of the internal accumulation of negative emotions by the
individual and the depletion of personal, emotional, energy resources, with the subsequent development
of the professional deformation and depersonalization of the individual, which is interpreted by scientists
      </p>
      <p>The professional burnout syndrome is included in the 11th revised version of the International
the state of public health and referrals to health care institutions" (Z00- Z99), which contains the reasons
for the population referral to health care institutions, which are not classified as diseases or medical</p>
      <p>
        2022 Copyright for this paper by its authors.
conditions (WHO, 28 May 2019). Specialists of the Center for Disease Control (CDC, Atlanta, USA)
and the Prevention National Institute for Occupational Safety (NIOS) annually review and supplement
the list of professional specialties related to the control of the chronic fatigue syndrome ("burnout") and
its prevention measures (CDC, NIOS, 2019) [
        <xref ref-type="bibr" rid="ref4 ref5">4,5</xref>
        ].
      </p>
      <p>
        In modern literature there are data characterizing the development of stress at work [
        <xref ref-type="bibr" rid="ref4 ref6">4,6</xref>
        ]. Different
categories of workers, first of all, workers in the social sphere (doctors, bank workers, teachers, law
enforcement officers) showed various manifestations of stress reactions to the professional burnout
syndrome [
        <xref ref-type="bibr" rid="ref10 ref6 ref7 ref8 ref9">6-10</xref>
        ]. The UK annual Labor Force Survey (LFS) statistical estimate shows that the total
number of work-related stress, depression or anxiety in 2020/2021 was 2,480 per 100,000 workers at a
prevalence rate. Healthcare and Social Work were identified as fields with the highest levels of stress,
depression or anxiety [
        <xref ref-type="bibr" rid="ref6">6</xref>
        ]. The frequency of professional burnout among employees of "helping", socially
significant professions is quite high, which is associated with the significant involvement of the specialist
in interpersonal communication, the constant ability to empathize and understand the problems of
another person [
        <xref ref-type="bibr" rid="ref11">11</xref>
        ]. Healthcare workers are a professional group of people who are an integral part of
any society, whose activities are aimed at providing highly qualified medical care, implementing
preventive measures in order to protect and improve the public health. The work of healthcare personnel
is associated with complex mental work, which unites people with both secondary and higher medical
education, as well as various specialties employed in the field of healthcare, which are characterized by
specific working conditions for each medical specialist separately [
        <xref ref-type="bibr" rid="ref12">12</xref>
        ].
      </p>
      <p>
        The study of functions that affect professional medical activity is becoming more and more relevant
in the social aspect, which is due to the great practical interest in the effective work of physicians. In
particular, the professional burnout syndrome in emergency medical staff develops as a result of
constantly being in situations related to communication, urgent decision-making, responsibility for the
life and health of the patient, variability in work and other organizational factors [
        <xref ref-type="bibr" rid="ref13">13</xref>
        ]. The consequences
of working in conditions of high labor intensity include suffering from psycho-emotional stress, which
leads to functional changes of the central nervous system (CNS), which can initially be characterized by
headaches, sleep and memory disorders, loss of attention and concentration, etc. In the future, functional
disorders may change to persistent psychosomatic disorders and as a result of professional burnout. The
literature review sufficiently illuminates the problem of the nature of burnout of emergency medical
staff, with full consideration of the entire link of professional training: doctors, paramedics, nurses
[
        <xref ref-type="bibr" rid="ref12 ref13">12,13</xref>
        ]. All of the above confirms the relevance of a detailed study of the impact of the tension of the
labor process on psychological and physiological processes in the body and manifestations of industrial
stress among emergency medical staff, as the primary link of patient care. Recent changes in the
provision of health care in the context of the COVID-19 pandemic caused by the SARS-CoV-2 [
        <xref ref-type="bibr" rid="ref14">14</xref>
        ], the
beginning of the war on the territory of Ukraine in 2022, also raise concerns about the professional
exhaustion of emergency medical care workers. In these conditions, the risk of professional burnout
increases, it can affect the quality and safety of health care as a whole and additionally negatively affect
the economic status of the state [
        <xref ref-type="bibr" rid="ref9">9</xref>
        ].
      </p>
      <p>The general scheme of the problem is shown in Figure No. 1. It provides factors of the labor process
that are triggers for the appearance of psycho-emotional stress in medical workers, which subsequently
lead to the development of symptoms of professional burnout. It is clear from the given scheme that the
task of our research was to develop an approach that will help to identify a complex of early criteria of
pre-pathological conditions in order to prevent the occurrence of symptoms of psycho-emotional stress,
and therefore to prevent the development of professional burnout.</p>
      <p>The relevance of this issue is sufficiently highlighted in the above. The given data indicate the state
and relevance of the problem and do not offer measures and algorithms to prevent the professional
burnout development, do not determine the criteria for diagnosing pre-pathological states development
in such a socially significant category of workers as healthcare workers. Therefore, it is important to
search for adequate and informative criteria using modern mathematical tools, which allow to identify
with a high degree of objectivity the criteria for the pre-pathological states of the professional burnout
syndrome in specialists of the emergency medical care center.</p>
      <p>Therefore, the aim of this study is to develop a scheme of a mathematical approach to establishing
the information criteria for the early diagnosis of the professional burnout development in healthcare
workers, using the example of emergency medical care to form a group of pre-pathological states of
professional burnout.</p>
    </sec>
    <sec id="sec-3">
      <title>2. Materials and methods</title>
      <p>The study of emergency medical care staff was conducted in two stages. At the first stage, a medical
and psychological questionnaire was conducted as a means of diagnosing the development of
professional burnout and assessing the individual and psychological qualities of respondents using the
questionnaire "Maslach Burnout Inventory - General Survey" (MBI-GS) in order to identify the levels
of professional burnout based on the three-factor model of the professional burnout syndrome
development. The second stage is the study for a comprehensive assessment of the psychophysiological
and psychological properties and functions of the respondents’ body based on the results of performing
test tasks, namely: a simple visual-motor reaction, a choice reaction, a difference reaction, a critical
blinking fusion frequency, memory for images, a visual-motor reaction with a score. For a
comprehensive assessment, a computer complex of psychophysiological testing NS-psychotest
(certificate of state registration No. 2017618884 dated August 10, 2017) was used. The representative
sample included 28 respondents – specialists in emergency medical care (doctor, paramedic) Fig. 2.</p>
      <p>
        The biggest part of Artificial Intelligence (AI) methods for classification tasks can be presented as
black-box-models. It means that special methods should be used for understanding reasons of such
decisions made by classification system. Explainable Artificial Intelligence [
        <xref ref-type="bibr" rid="ref16 ref17">16-17</xref>
        ] is the way to
explain and understand predictions made by black box models. This approach can be used for detecting
those psycho-physiological indicators that affect to dividing respondents by the level of professional
burnout. Explainable Artificial Intelligence has been developing rapidly in recent years and is
represented by a variety of systems. The most part of such algorithms, methods and approaches were
described in [
        <xref ref-type="bibr" rid="ref18">18</xref>
        ]. For our research we’ve been choose Eli5 explanator. Eli5 is based on Permutation
Importance method and LIME explanator [
        <xref ref-type="bibr" rid="ref19">19</xref>
        ].
      </p>
      <p>
        The main idea of Permutation Importance method is how much any score as accuracy, f1, R2 etc.
decreases when a feature is not available. On each step one feature is replaced by random noise (to
make feature column presented in dataset, but it no longer contains useful information), the estimator
re-trains and algorithm checks the score. So, it shows what features are important within a dataset, not
what is important within a concrete trained model. Eli5 in combination with lgbm-model shows high
accuracy for explaining model results for covid-19 data [
        <xref ref-type="bibr" rid="ref20">20</xref>
        ].
      </p>
    </sec>
    <sec id="sec-4">
      <title>2.1. Pipeline indicators For</title>
    </sec>
    <sec id="sec-5">
      <title>Defining Informativeness of psycho-physiological</title>
      <p>
        Mathematical approach is based on classification method (logistic regression) with explanator Eli5
[
        <xref ref-type="bibr" rid="ref15">15</xref>
        ]. Eli5 is specialized Python library for visualizing and debugging various machine learning models
based on calculating the contribution of each feature to the prediction. As it was mentioned, principal
goal is to define those psycho-physiological indicators that affect to the level of burnout. The basic view
of pipeline is presented in Fig. 3.
      </p>
      <p>It consists of MBI-GS questionnaire and professional burnout type that was calculated based on
special methodology. Classification model training combining with explanator is needed for defining
information indicators of MBI-GS questionnaire. Measured psycho-physiological indicators are fed to
second classification model as a training vector and results of calculation professional burnout level for
the same respondent – as a target vector. After models were trained, explanator analyses trained model
and explain result.</p>
    </sec>
    <sec id="sec-6">
      <title>2.1.1.Defining a level of professional burnout based on MBI questionnaire</title>
      <p>
        At first step professional burnout level have to be defines for each respondent. The methodology of
those calculating was modified in [
        <xref ref-type="bibr" rid="ref12">12</xref>
        ]. The MBI-GS questionnaire contains 16 questions describing
various work-related experiences. The method of assessing the emotional state of the respondents
involved the distribution of answers according to scales regarding the risk of developing burnout:
"emotional exhaustion" (ee), which corresponded to the answers to questions 1, 2, 3, 4, 6;
"depersonalization" (zy) – answers to questions 8, 9, 13, 14, 15; "reduction of personal achievements"
(ef) – answers to questions 5, 7, 10, 11, 12, 16. Results of three scales were defined in points and each
of them were transformed to professional burnout level:
 0 group – no burnout;
 1 group – initial manifestation of professional burnout process (it can be initial changes by one of
three scales but it cannot be a complete burning out by any of them);
 2 group – severe symptoms of professional burnout (it can be also a complete burning out by one
of scales).
      </p>
      <p>Scatter plot of respondents in the space of three principal components is presented in Fig.4.</p>
      <p>It is easy to see that density of group without professional burnout (0 group) is higher than density
of group of high professional burnout level (2 group). It can be explained by the fact that professional
burnout process can be occurred on one of three scales (“emotional exhaustion” (ee),
“cynicism/depersonalization” (zy) and “reduction of personal achievements” (ef)). The most interesting
for diagnostic is 1 group – the group of probands with initial manifestation of professional burnout
process.</p>
      <p>MBI-GS
questionnaire</p>
      <sec id="sec-6-1">
        <title>Professional burnout type</title>
      </sec>
      <sec id="sec-6-2">
        <title>Psychophysiological measurements</title>
      </sec>
      <sec id="sec-6-3">
        <title>Classification model training #1</title>
      </sec>
      <sec id="sec-6-4">
        <title>Classification model training #2</title>
      </sec>
      <sec id="sec-6-5">
        <title>Explanator Eli5</title>
        <p>(information indicators of</p>
      </sec>
      <sec id="sec-6-6">
        <title>MBI-GS questionnaire)</title>
      </sec>
      <sec id="sec-6-7">
        <title>Explanator Eli5 (information indicators of psycho-physiological measurements)</title>
        <p>Eli5 explainer helps to determine the information indices that demonstrate the development of
depersonalization phenomena according to the answers to questions 13, 14, 15 (MBI_14_zy,
MBI_13_zy, MBI_15_zy) (Fig. 5).</p>
        <p>These indicators are most informative ones for the presented division into groups. The accuracy of
the logistic regression model is 0,96  0,13 (based on 10-fold cross-validation).</p>
      </sec>
    </sec>
    <sec id="sec-7">
      <title>2.1.2. Dependence of psycho-physiological indicators</title>
      <p>professional burnout
on
the
level
of</p>
      <p>The next part of research is based on defining the dependencies between psycho-physiological
indicators and level of professional burnout. Psycho-physiological indicators consist of six subgroups:
Simple visual-motor reaction (SVMR), Choice reaction (CR), Discrimination reaction (DR),
Visualmotor reaction with counting (VMRC), Critical flicker fusion frequency (CFFF) and Memory for
images (MI). Psycho-physiological indicators for each of subgroups are presented in Table 1 and Table
2. It should be noted that the number of psycho-physiological indicators is close to the number of
respondents. That’s why each type of psycho-physiological indicators was processed separately. Results
presented in Fig. 6. The most important features have higher value of weight.</p>
      <p>Visual-motor reaction with</p>
      <p>counting (VMRC)</p>
      <p>Average reaction time
Root mean square deviation of</p>
      <p>the reaction time
Number of correct reactions
Number of mistakes made
Number of passes made
Number of false reactions
Number of premature clicks
Attention span and working
memory</p>
      <p>Critical flicker fusion frequency</p>
      <p>(CFFF)</p>
      <p>Average frequency with
increasing flicker frequency</p>
      <p>Average frequency with
decreasing flicker frequency</p>
      <p>Memory for images</p>
      <p>(MI)
Number of correct answers</p>
      <p>Number of mistakes</p>
      <p>Memory capacity</p>
    </sec>
    <sec id="sec-8">
      <title>3. Results</title>
      <p>According to the results of the medical and psychological survey, a group of workers with
prepathological conditions was identified, for which significant criteria questions from the MBI
questionnaire were determined. The analysis of the obtained data proved that the workers from the
prepathology group feel exhausted at the end of the working day (emotional exhaustion), strive to fulfill
only their direct duties, have manifestations of cynicism (depersonalization), doubts about the
significance of their work, but at the same time have confidence that the work is performed effectively
(reduced personal accomplishment).</p>
      <p>During the study of psychophysiological changes in the body using the "simple visual-motor
reaction" method among the emergency medical staff, it was established that the significant criteria
indicators are an increase in the "visual-motor reaction rate" indicator; an increase in the "performance
evaluation by reaction rate" indicator; an increase in the "functional level of the system" with a
simultaneous decrease in the "level of functional capabilities" indicators.</p>
      <p>According to the "visual-motor reaction with counting" method, the criterion-significant indicators
are an increase in the "number of errors", "attention span and working memory" and decrease in "root
mean square deviation of the sensorimotor reaction rate". According to the method of "difference
reaction", all indicators had the same low level of informativeness. According to the "choice reaction"
method, only a decrease in the "number of clicks" indicator is informative.</p>
      <p>Indicators of the "critical frequency of flicker fusion" method are sufficiently informative compared
to the indicators of the "memory for images" method.</p>
      <p>The general characteristic of the comparison of methods and the dynamics of changes in indicators
shows that the establishment of the most significant changes from the CNS was determined using the
"simple visual-motor reaction" method. Therefore, the decrease in the efficiency of visual-motor
reactions is provoked by the formation of inertia of the nervous processes due to the decrease in the
sensorimotor reaction rate, which is probably related to the branching of neural connections between
the frontal parts of the cerebral cortex and the motor areas of the cortex, subcortical structures
responsible for the motor reaction formation. These changes can be considered as the instability of the
visual-motor response to external stimuli against the background of the suppression of the functional
stability of the brain structures work, their reduced strength, balance, which is caused by a high level of
work intensity. The above-mentioned changes are the triggers for declining performance levels,
developing fatigue, and reducing sensory memory capacity.</p>
    </sec>
    <sec id="sec-9">
      <title>4. Conclusions</title>
      <p>1. The establishment of early diagnostic criteria for the prenosological state of the professional
burnout development in healthcare workers (on the example of the emergency care staff) should be
considered a necessary stage in the preventive measures development.</p>
      <p>2. According to the results of selection of the cluster of the pre-pathological state group based on the
MBI-GS questionnaire using a mathematical model based on artificial intelligence, it was established
that the phenomena of emotional exhaustion and state of reduced personal accomplishment were
observed in the group of pre-pathology, while the most informative indicators were the development of
depersonalization.</p>
      <p>3. The use of an original mathematical approach made it possible to establish objective
psychophysiological indicators by which it is possible to determine the pre-pathological state from the CNS
functioning, namely: a simple visual-motor reaction (according to the criteria of "visual-motor reaction
rate", "performance evaluation by reaction rate", "functional level of the system", "level of functional
capabilities"), visual-motor reaction with counting (according to the criteria "number of errors", "root
mean square deviation of the sensorimotor reaction rate").</p>
    </sec>
    <sec id="sec-10">
      <title>5. Acknowledgements</title>
      <p>The study was conducted as a part of the scientific research work "Justification of criteria for
prepathological states of professional burnout in healthcare workers", state registration number
0121U110914, financed by the Ministry of Health of Ukraine at the expense of the state budget.</p>
    </sec>
    <sec id="sec-11">
      <title>6. References</title>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [1]
          <string-name>
            <given-names>Y.</given-names>
            <surname>Malhi</surname>
          </string-name>
          ,
          <article-title>The Concept of the Anthropocene, Annual Review of Environment and Resources (</article-title>
          <year>2017</year>
          )
          <fpage>77</fpage>
          -
          <lpage>104</lpage>
          . doi.org/10.1146/annurev-environ-
          <volume>102016</volume>
          -060854
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [2]
          <string-name>
            <given-names>G.</given-names>
            <surname>Weßel</surname>
          </string-name>
          , E. Altendorf,
          <string-name>
            <given-names>M.</given-names>
            <surname>Schwalm</surname>
          </string-name>
          et al.,
          <article-title>Self-determined nudging: a system concept for humanmachine interaction</article-title>
          .
          <source>Cogn Tech Work</source>
          <volume>21</volume>
          (
          <year>2019</year>
          ). doi.org/10.1007/s10111-019-00556-5
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [3]
          <string-name>
            <given-names>M.</given-names>
            <surname>Katana</surname>
          </string-name>
          ,
          <string-name>
            <given-names>C.</given-names>
            <surname>Röcke</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S. M.</given-names>
            <surname>Spain</surname>
          </string-name>
          ,
          <string-name>
            <given-names>M.</given-names>
            <surname>Allemand</surname>
          </string-name>
          , Emotion Regulation,
          <article-title>Subjective Well-Being, and Perceived Stress in Daily Life of Geriatric Nurses</article-title>
          , Frontiers in Psychology.
          <volume>10</volume>
          .05 (
          <year>2019</year>
          )
          <article-title>: doi</article-title>
          .org/10.3389/fpsyg.
          <year>2019</year>
          .01097
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          <article-title>[4] National Institute for Occupational Safety and Health. Centers for disease control</article-title>
          .
          <source>Total worker health workforce development program</source>
          ,
          <year>2020</year>
          . URL: https://www.cdc.gov/niosh/twh/default.html
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          [5] Bulletin of the World Health Organization.
          <year>2019</year>
          .
          <volume>585</volume>
          -
          <fpage>586</fpage>
          . URL: http://dx.doi.org/10.2471/BLT.19.020919
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          [6]
          <string-name>
            <given-names>P. J.</given-names>
            <surname>Bridgeman</surname>
          </string-name>
          ,
          <string-name>
            <given-names>M. B.</given-names>
            <surname>Bridgeman</surname>
          </string-name>
          ,
          <string-name>
            <given-names>J.</given-names>
            <surname>Barone</surname>
          </string-name>
          ,
          <article-title>Burnout syndrome among healthcare professionals</article-title>
          ,
          <source>American Journal of Health-System Pharmacy</source>
          (
          <year>2018</year>
          ).
          <fpage>147</fpage>
          -
          <lpage>152</lpage>
          . doi.org/10.2146/ajhp170460
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          <article-title>[7] Work-related stress</article-title>
          ,
          <source>anxiety or depression statistics in Great Britain</source>
          ,
          <year>2021</year>
          . Health and Executive.
          <year>2021</year>
          . 8-
          <fpage>10</fpage>
          . URL: https://www.hse.gov.uk/statistics/causdis/stress.pdf
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          <article-title>[8] Workplace stress: a collective challenge</article-title>
          .
          <source>International Labour Organization</source>
          ,
          <year>2016</year>
          . URL:https://www.ilo.org/wcmsp5/groups/public/---ed_protect/---protrav/--- safework/documents/publication/wcms_466547.pdf
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          [9]
          <string-name>
            <given-names>M.P.</given-names>
            <surname>Salyers</surname>
          </string-name>
          ,
          <string-name>
            <given-names>K.A.</given-names>
            <surname>Bonfils</surname>
          </string-name>
          ,
          <string-name>
            <given-names>L.</given-names>
            <surname>Luther</surname>
          </string-name>
          et al.,
          <article-title>The Relationship Between Professional Burnout and Quality and Safety in Healthcare: A Meta-Analysis</article-title>
          ,
          <source>Journal of General Internal Medicine</source>
          .
          <volume>32</volume>
          (
          <year>2017</year>
          )
          <fpage>475</fpage>
          -
          <lpage>482</lpage>
          . doi.org/10.1007/s11606-016-3886-9
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          [10]
          <string-name>
            <given-names>B.</given-names>
            <surname>Thielmann</surname>
          </string-name>
          , I. Zavgorodnii,
          <string-name>
            <given-names>K.</given-names>
            <surname>Zub</surname>
          </string-name>
          ,
          <string-name>
            <surname>I. Böckelmann,</surname>
          </string-name>
          <article-title>The perception of stress, behavior in stressful situations and mental health of bank employees within a German-Ukrainian comparative study</article-title>
          ,
          <source>International Journal of Occupational Medicine and Environmental Health</source>
          .
          <volume>35</volume>
          (
          <year>2021</year>
          )
          <fpage>81</fpage>
          -
          <lpage>94</lpage>
          . doi.org/10.13075/ijomeh.
          <year>1896</year>
          .01739
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          [11]
          <string-name>
            <given-names>B.W.</given-names>
            <surname>Dubale</surname>
          </string-name>
          ,
          <string-name>
            <given-names>L.E.</given-names>
            <surname>Friedman</surname>
          </string-name>
          ,
          <string-name>
            <given-names>Z.</given-names>
            <surname>Chemali</surname>
          </string-name>
          , et al.,
          <article-title>Systematic review of burnout among healthcare providers in sub-Saharan Africa</article-title>
          .
          <source>BMC Public Health</source>
          .
          <volume>19</volume>
          (
          <year>2019</year>
          )
          <article-title>1247</article-title>
          . doi.org/10.1186/s12889- 019-7566-7
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          [12]
          <string-name>
            <given-names>I.</given-names>
            <surname>Zavgorodnii</surname>
          </string-name>
          ,
          <string-name>
            <given-names>O.</given-names>
            <surname>Lalymenko</surname>
          </string-name>
          ,
          <string-name>
            <surname>I. Perova</surname>
          </string-name>
          ,
          <string-name>
            <given-names>P.</given-names>
            <surname>Zhernova</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Kiriak</surname>
          </string-name>
          et al.,
          <source>Early Revealing of Professional Burnout Predictors in Emergency Care Workers of Lecture Notes in Computational Intelligence and Decision Making</source>
          . Springer International Publishing. Cham.
          <year>2021</year>
          . doi.org/10.1007/978-3-
          <fpage>030</fpage>
          -82014-5_
          <fpage>31</fpage>
        </mixed-citation>
      </ref>
      <ref id="ref13">
        <mixed-citation>
          [13]
          <string-name>
            <given-names>O. S.</given-names>
            <surname>Lalymenko</surname>
          </string-name>
          ,
          <string-name>
            <given-names>І. V.</given-names>
            <surname>Zavhorodnii</surname>
          </string-name>
          ,
          <string-name>
            <given-names>V. A.</given-names>
            <surname>Kapustnyk</surname>
          </string-name>
          ,
          <string-name>
            <given-names>I.</given-names>
            <surname>Boeckelmann</surname>
          </string-name>
          ,
          <string-name>
            <given-names>V. F.</given-names>
            <surname>Zabashta</surname>
          </string-name>
          ,
          <string-name>
            <given-names>M. O.</given-names>
            <surname>Stytsenko</surname>
          </string-name>
          <article-title>"Medical-psychological aspects of professional deformation of personality development among emergency medical staff"</article-title>
          ,
          <source>Zaporozhye Medical Journal</source>
          , (
          <year>2022</year>
          )
          <fpage>61</fpage>
          -
          <lpage>69</lpage>
          . doi.org/10.14739/
          <fpage>2310</fpage>
          -
          <lpage>1210</lpage>
          .
          <year>2022</year>
          .
          <volume>1</volume>
          .
          <fpage>239108</fpage>
        </mixed-citation>
      </ref>
      <ref id="ref14">
        <mixed-citation>
          [14]
          <string-name>
            <given-names>A. H.</given-names>
            <surname>Al-Rohaimi</surname>
          </string-name>
          ,
          <string-name>
            <surname>F</surname>
          </string-name>
          . Al Otaibi,
          <article-title>Novel SARS-CoV-2 outbreak and COVID19 disease; a systemic review on the global pandemic</article-title>
          ,
          <source>Genes &amp; Diseases</source>
          (
          <year>2020</year>
          )
          <fpage>491</fpage>
          -
          <lpage>501</lpage>
          . doi.org/10.1016/j.gendis.
          <year>2020</year>
          .
          <volume>06</volume>
          .004
        </mixed-citation>
      </ref>
      <ref id="ref15">
        <mixed-citation>
          [15]
          <string-name>
            <given-names>Eli5</given-names>
            <surname>Documentation</surname>
          </string-name>
          <string-name>
            <surname>URL</surname>
          </string-name>
          : https://eli5.readthedocs.io/en/latest/index.html.
        </mixed-citation>
      </ref>
      <ref id="ref16">
        <mixed-citation>
          [16]
          <string-name>
            <given-names>A.</given-names>
            <surname>Torcianti</surname>
          </string-name>
          and
          <string-name>
            <given-names>S.</given-names>
            <surname>Matzka</surname>
          </string-name>
          ,
          <article-title>"Explainable Artificial Intelligence for Predictive Maintenance Applications using a Local Surrogate Model,"</article-title>
          <source>2021 4th International Conference on Artificial Intelligence for Industries (AI4I)</source>
          ,
          <year>2021</year>
          , pp.
          <fpage>86</fpage>
          -
          <lpage>88</lpage>
          , doi: 10.1109/AI4I51902.
          <year>2021</year>
          .
          <volume>00029</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref17">
        <mixed-citation>
          [17]
          <string-name>
            <given-names>A.</given-names>
            <surname>Rawal</surname>
          </string-name>
          ,
          <string-name>
            <given-names>J.</given-names>
            <surname>Mccoy</surname>
          </string-name>
          ,
          <string-name>
            <given-names>D. B.</given-names>
            <surname>Rawat</surname>
          </string-name>
          ,
          <string-name>
            <given-names>B.</given-names>
            <surname>Sadler</surname>
          </string-name>
          and
          <string-name>
            <given-names>R.</given-names>
            <surname>Amant</surname>
          </string-name>
          ,
          <article-title>"</article-title>
          <source>Recent Advances in Trustworthy Explainable Artificial Intelligence: Status, Challenges and Perspectives," in IEEE Transactions on Artificial Intelligence</source>
          , doi: 10.1109/TAI.
          <year>2021</year>
          .
          <volume>3133846</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref18">
        <mixed-citation>
          [18]
          <string-name>
            <given-names>S.</given-names>
            <surname>Sarp</surname>
          </string-name>
          ,
          <string-name>
            <given-names>M.</given-names>
            <surname>Kuzlu</surname>
          </string-name>
          ,
          <string-name>
            <given-names>U.</given-names>
            <surname>Cali</surname>
          </string-name>
          ,
          <string-name>
            <given-names>O.</given-names>
            <surname>Elma</surname>
          </string-name>
          and
          <string-name>
            <given-names>O.</given-names>
            <surname>Guler</surname>
          </string-name>
          ,
          <article-title>"An Interpretable Solar Photovoltaic Power Generation Forecasting Approach Using An Explainable Artificial Intelligence Tool,"</article-title>
          <source>2021 IEEE Power &amp; Energy Society Innovative Smart Grid Technologies Conference (ISGT)</source>
          ,
          <year>2021</year>
          , pp.
          <fpage>1</fpage>
          -
          <lpage>5</lpage>
          , doi: 10.1109/ISGT49243.
          <year>2021</year>
          .
          <volume>9372263</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref19">
        <mixed-citation>
          [19]
          <string-name>
            <given-names>M. T.</given-names>
            <surname>Ribeiro</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S.</given-names>
            <surname>Singh</surname>
          </string-name>
          , and
          <string-name>
            <given-names>C.</given-names>
            <surname>Guestrin</surname>
          </string-name>
          ,
          <article-title>"Why should I trust you?': Explaining the predictions of any classifier"</article-title>
          <source>in Proc. 22nd ACM SIGKDD Int. Conf. Knowl. Discovery Data Mining</source>
          ,
          <year>2016</year>
          , pp.
          <fpage>1135</fpage>
          -
          <lpage>1144</lpage>
          , doi: http://dx.doi.org/10.1145/2939672.2939778
        </mixed-citation>
      </ref>
      <ref id="ref20">
        <mixed-citation>
          [20]
          <string-name>
            <given-names>G.</given-names>
            <surname>Marvin and M. G. R. Alam</surname>
          </string-name>
          ,
          <article-title>"Cardiotocogram Biomedical Signal Classification and Interpretation for Fetal Health Evaluation,"</article-title>
          <source>2021 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)</source>
          ,
          <year>2021</year>
          , pp.
          <fpage>1</fpage>
          -
          <lpage>6</lpage>
          , doi: 10.1109/CSDE53843.
          <year>2021</year>
          .
          <volume>9718415</volume>
          .
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>