Journal of Depression and Anxiety

Journal of Depression and Anxiety
Open Access

ISSN: 2167-1044

Research Article - (2018) Volume 7, Issue 4

Cognitive Behavioural Therapy to Compute Psychological Illness of Students by Using an Expert System

Syeda Huma Jabeen1*, Muhammad Ejaz Sandhu1, Faiza Tahir1, Amadullah M1, Noshaba Noureen1, Sehrish Munawar Cheema1, Saad Salman1, Rida Ali1, Dr. Nadeem Ahmed2 and Zeeshan Maqsood2
1Faculty of Computing and Information Technology, University of Sialkot, Pakistan
2Department of Quantitative Methods, University of Management and Technology, Lahore, Pakistan
*Corresponding Author: Syeda Huma Jabeen, Faculty of Computing and Information Technology, University of Sialkot, Pakistan, Tel: 943040461989 Email:

Abstract

Objective: This research is being conducted to develop a technological solution for mentally distorted students. Though the mental health of university students is known globally as a momentous public health matter. Academicals and social stresses are playing quite a negative role in university student’s life, especially in forms of mental illness like stress, depression, and anxiety. These mental health issues are becoming a major constraint towards their studies and career.
Method: Psychologist used different scales to measure a level of mental disorder. However, to measure such a disease level, we are working on a knowledge-based expert system that will be used to compute its level among the students who are affiliated with technological studies. Mostly psychologist does psychotherapy and use other instruments to cure such a patient for which they must have to visit the psychologist. However, if the psychologist is not available especially in remote areas then the expert system can be used as reciprocal. In order to make our expert system more validate and authentic the knowledge of psychological expert will be used under the process of development.
Results: Data from 500 technological University Students are collected from one of the universities in Sialkot, Pakistan. Almost more than 200 students remained clear minded and fall under the normal state of depression and 122 students in case of Anxiety remained normal. 206 students out of 500 were responders to the abnormal stage of anxiety on a 5-point scale from an average of 4.5 points. On the basis of this, an expert system is being designed to facilitate the students.
Conclusion: As per results, 30% - 35% students were in the range of abnormality. Therefore, we are further going to develop an evaluation mechanism by using technological ways so that an expert system can replace a psychologist

Keywords: Anxiety; Depression; Expert system; Mental health; Artificial intelligence

Introduction

Mental health is very important in all phases of lifecycle and it comprises of the emotional, mental and social welfare of each individual. The performance of an individual depends upon its mental health, if the mental health is good then the performance will be high in general work life or as a student or in business. Certain stresses, social pressures or family disturbances may have impact on performance. It has been observed that student’s performance also suffers due to these stresses, therefore, there is a need to evaluate such stresses during study period and it is required to design such systems which can evaluate and provide relief to the students so that they can perform well. This study has been conducted for the technological students of a university in Sialkot, Punjab, Pakistan. Students facing peer ill-treatment such as mistreatment habitually show symbols of suffering and these practices can interoperate students’ comfort and make them further defenseless to mental health hitches [1].

Each person’s mental health may regulate in what way that person can knobs stress, relays to others, and select the option as per scenarios. It depicts how one reflects, feels, and acts [2].

Marvelous growth has thru in emerging Empirically Supported Interventions (ESIs) to indorse optimistic psychological health results for youngsters. Though acceptance, eminence employment, and accountability of these ESIs in actual arena situations continues to hold-up behindhand [3,4].

Artificial Intelligence, preliminary in the expansion of e-science [5,6], is the topmost technology newly recognized as valuable inside healthcare and is predictable to “support healthcare physicians by means of medical data that is clinically pertinent instantaneous distinction information.” [5].

By 2025, AI is likely to be affected in 60% of international hospitals and 90% of US insurance corporations. Stimuli for AI comprise enhanced patient results, condensed treatment charges, and patient core plans of treatment. AI supports treatment policies of Directly Observed Therapy (DOT) can leads towards a progress of patient results grounded on automatic patient supervision and engagement resolutions to recuperate healing program devotion [6,7].

Cognitive Computing is an overall awareness offering Artificial intelligence as figuring abilities which can be functional to enhance/ increase human cognition, maintain balances and imitates human knowledge in cognitive-task presentation. Cognitive computing contains AI knowledge and techniques that includes understanding of natural language, machine learning, speech and image recognition, familiar human interface [8,9].

Psychological suffering and mental illness is causing the disability of student’s world widely; an account for one-third decades lost for the reason of disability [10,11]. The concern about whether the students in colleges and universities are mentally ill or not and prevalence ratio of mental illness among them; is increasing day by day on international level [12,13]. Information extracted from precise epidemiological data is necessary for: (1) understanding the affliction and degree of mental illness among third-of-order pupils; (2) Guiding many interpolations to minimize the risk of psychological well-being and help in improvement too; (3) to observe lean curve with time [14]. Several studies have been done to identify and measure at-risk populations regarding mental health problems and explored many subgroups of tertian students that comprises of international, graduate, undergraduate students [15,16] students especially from information technology [17] and members from counselling centers of many universities [18,19].

It is important to set benchmarking in contrast to healthy population so that it would make identifiable the mental health of university students is on risk or not? Previously estimates were made containing some limitations and difficulties in benchmarking process due to non-comparable methodology used; because it would not provide a comparison with general population’s epidemiological research results. Previous methodologies that were used include questions from students: (1) Whether they have depression or not? (2) Some checklists of symptoms and screening devices to calculate psychological distress that lacks powerful comparable data with general population [15,20,21].

Participants in previous research were of different age groups, nationality differences, and employment status, marital status, living situations, financial stress and hours of employment [22].

To estimate psychological illness of a well-being surveys were conducted and participants were asked about: (1) the number of days you were incapable to do work, manage daily activities or study in last four weeks; (2) the number of days you were able to do any task, manage their daily activities or study but with cut down because of distress or of these feelings? [22]. Responses were assessed and tagged as Days out of Role (DOR) and Days Cut Back (DCB). Univariate ANOVA applied to analyze the difference between various age-groups in terms of distress.

Research Methodology

Mental wellness of educate is a domain of clear dignity. This research writes up aim to a web-based multi-stone's homo being body through cognitive restructuring program to help true identification. This can be great deal with maladaptive, harmful feelings, and emotion, thereby efficiently endeavoring to promote flexibility and recover mental health between college students. The organization should be collaborative and consistent response must be taken from the depicted object. Therapist into the organization required for a person subject field if further evaluation is essential. The system suggests a generic platform for implementation of web-based cooperative Cognitive Behavioural Grooming (CBG) a need for particular university [23].

Researchers conducted a National epidemiological survey in Canadian Universities for assessing elevated psychological distress among undergraduate or teen aged students. Probable a sample of 7800 participants from 16 universities were intervened in survey and a screening tool for mental health 12-item General Health Questionnaire (GHQ-12) was used. About 30-35% participants reported elevated psychological disorder varied with respect to region, sex, academic year, orientation, recreational. It was analyzed that ratio of undergraduate students were representing high elevated distress than general population of Canada. Question stem read as follows: “How is your mental health since last some weeks on overall and tell about complaints that you recently put”. Possible responses were: (1) usual (2) underneath than usual; (3) very fewer than usual; (4) healthier than usual [24].

By keeping the others work under consideration, we have designed Expert System named as E-Psychologist based on artificial intelligence techniques to compute level of mental illness among students of university who are related to technological fields.

There are mainly four steps to reach the final results. These four stages are followed one after another in order as shown in Figure 1. Process Diagram of Expert System. The methodology adopted in this research comprises an Action Research (AR) case study based on mentally unhealthy technological students in university.

depression-anxiety-Process-diagram

Figure 1: Process diagram of expert system.

Further explanation of way to develop this expert is as follows:

Participants

The contributors consisted of 500 students from Faculty of Computing and IT from one university in Sialkot, Punjab, Pakistan. The age of the contributors was ranged from 19 to 24 having an average age of 21-years.

Procedures

The methodology adopted in this research comprises an Action Research (AR) case study based on mentally unhealthy technological students in the university. We first contacted the Director of Computing and IT and asked approval to survey numerous classes on campus. After receiving permission, we arbitrarily selected the classes to get the data, survey forms were distributed among students and collected. The surveys were led in the classrooms and the instructor was currently present throughout all scenario. All participant students were acknowledged with same directions of instructions and were stated that their involvement in this section of data gathering was voluntary and furthermore, their identity would be kept unidentified. Each contributor done with a self-report survey.

Measurements

Requirements were being gathered through medically approved survey questionnaires (The hospital anxiety and depression scale (Zigmond AS, Snaith RP) by which we have tried to compute the level of depression and anxiety among them. The first portion of the questionnaire consisted of Anxiety questions (what they feel in daily life). The next series of depression questions (what they feel while working with technology). Items were assessed using 5-point Likert scales. Detailed Questionnaire is attached in appendix.

Psychological expert system

On the basis of data gathered by survey, an android application expert system (E-Psychologist) is designed. E-Psychologist will generate the patient’s profile and leads the patient towards different levels (level 1 for Anxiety and level 2 for Depression). Afterwards, it intelligently predicts disease of the patient via questionnaires specified by each level. Results are obtained by using rule based expert system.

SPSS validation

Furthermore, the result of application is validated by using Descriptive techniques of SPSS. This has ensured that results obtained by application are correct.

Implementation and System Design

Coding and implementation

E-Psychologist is designed by using java programming language. Rule based technique is adopted for coding this expert system. The Android Studio 3.1.1 is used for implementation. Data related to different users has been stored by using a real time database named as Firebase Database. Only a registered user can access application so its mandatory for a user to be authenticate via mobile number. For this purpose, Firebase AUTH is used as shown in Figure 2. Authentication of User Phone Number.

depression-anxiety-Authentication-user

Figure 2: Authentication of user phone number.

Designed model of an expert system

As shown in code below Design Model of Expert System depicts the complete scenario that how does this phycology based expert system works to compute level of mental illness among students of university who are related to technological fields. Analyses of gathered data is validated by conceptual model that is basically developed on the base of rule based expert system, once it is validated then it is implemented and verified by standard SPSS evaluated model by using various techniques of stats. At the end result of both conceptual and evaluated model are compared to validate the phycology expert system application that the results obtained by E-Psychologist are accurate.

Private void sign In With Phone Auth Credential (Phone Auth Credential credential) {mAuth. sign In With Credential (credential)

.add On Complete Listener (this, (task) -˃ {

If (task is successful ()){

Intent intent = New intent (signup. this, Generate Profile.closs);

Start Activity (intent);

Finish ();

Designing of android application

E-Psychologist is based on mainly four panels. Front end of this application is quite simple and easy to understand for users, even a new user can easily go through with it without any difficulty. Because it is designed by keeping the user’s level of understanding in mind.

E-Psychologist includes following panels:

• Login System

• Automated Profile Generation

• Runtime Repository Questionnaires

• Result Generation

• Working of E-Psychologist

Working of E-Psychologist is mainly consisting of three steps as following:

Step 1: As now a day’s authentication is major issue, especially when we are talking about medical application and people data. First of all, user needs to login on application and if it is new user it required to register him by using his data i.e., name, age, and gender as shown in Figure 3. Generate a user profile.

depression-anxiety-Generate-user-profile

Figure 3: Generate a user profile.

Step 2: Registration will be completed by using a phone number. Once the phone number is validated, code message will be prompt over mobile screen to complete a registration over application.

Step 3: After successful registration, there are two levels, first for anxiety and second level for depression as shown in Figure 4.

depression-anxiety-mental-disease-illness

Figure 4: Levels to compute mental disease illness.

In levels to compute mental disease illness, the user will be able to compute his level of illness by entering in level (in each level expert system will prompt a questionnaire and then generate level of illness and percentage of mental sickness of that individual).

Results and Discussion

Results are compared as they are generated by two ways:

Result generated by expert system

E-Psychologist result is being validated by make it available to different students of technology and result is compiled for both anxiety and depression as shown in Table 1. Expert System Computed Result. The table shows only data of four number of patients, it can be clearly seen that two of them are on borderline anxiety only one user is abnormal. But meanwhile, on the other side of depression, none of four is suffered of it.

Patient ID Anxiety % of Anxiety Depression % of Depression
P1 Borderline Abnormal 47.62% Borderline Abnormal 47.62%
P2 Normal 28.57% Normal 14.29%
P3 Abnormal (Case) 71.43% Abnormal (case) 33.33%
P4 Borderline Abnormal 42.86% Abnormal (case) 57.14%

Table 1: Expert system computed results.

Result generated by SPSS

Data from 500 technological University Students are collected from one of the universities in Sialkot, Punjab, Pakistan. Table 2 explains descriptive analysis of the data explains that most of our participants that are selected by the sample are at the normal stage of anxiety. Standard Deviation explains that on average, the values in the dataset differ from the mean by + 0.79312 in case of anxiety and in case of depression it differs from the mean by + 0.82985. The overall variation in anxiety is 0.629 and in depression is 0.689.

Statistics Anxiety Depression
Mode 3.0 1.0
Standard Deviation 0.79312 0.82985
Variance 0.629 0.689

Table 2: Descriptive table of disease.

The frequency table mentioned above in Table 3. Frequency table of disease and graph is illustrating Anxiety and Depression-level (Graph 1). The Data that was being collected by 500 University Students for computing their level of Anxiety and Depression (Normal, Borderline and Abnormal) and among these students (collected data) 24.4% are experiencing anxiety at the normal stage and 34.4% are at borderline and 41.2% are suffering from the abnormal stage. And in case of depression among 500 students 44.8% are experiencing Depression at the normal stage and 30.6% are at borderline and 24.6% are suffering from the abnormal stage.

Level of Illness Anxiety Depression
Frequency Percentage Frequency Percentage
Normal 224 44.8 122 24.4
Borderline 153 30.6 172 34.4
Abnormal 123 24.6 206 41.2
Total 500 100 500 100

Table 3: Frequency table of disease.

depression-anxiety-frequency-disease

Graph 1: Histogam of between frequency and disease.

Conclusion and Future Work

The determination of this research was to observe the consequence of hypothetical stress and learned resourcefulness on overall academic outcomes and performance. It was hypothesised that academic stress would be linked with a low level of academic routine. Results presented authority for this supposition by justifying a important negative effect of academic pressure/stress on academic performance.

The results clearly show that around 75% of students are experience in anxiety or at border line and around 55% of students having depression problem or at border line. So that’s mean the performance of around 60% students are suffering due to psychological disorder.

The results clearly show that around 30% - 35% students falling under the range of abnormality. Therefore, it was required that a system should be designed to evaluate technological based assessment and relief mechanism to increase a performance of the students. There is no real time Stress and Anxiety release mechanism so, in order to make it possible we are working over it.

In future, we are further going to develop a cure mechanism by using technological ways i.e., psychological games, videos, automatic appointment scheduling and meditation therapies to decrease the level of mental illness that will make the brain less stressed so that they can accomplish their work in better way.

References

  1. Fontaine NM, Hanscombe KB, Berg MT, McCrory EJ, Viding E (2018) Trajectories of callous-unemotional traits in childhood predict different forms of peer victimization in adolescence. J Clin Child Adolesc Psychol 47: 458-466
  2. Salle TL, George HP, McCoach DB, Polk T, Evanovich LL (2018) An examination of school climate, victimization, and mental health problems among middle school students self-identifying with emotional and behavioral disorders. Behavioral Disord 43: 383-392.
  3. Livet M, Yannayon M, Sheppard K, Kocher K, Upright J, et al. (2018) Exploring provider use of a digital implementation support system for school mental health: A pilot study. Adm Policy Ment Health 45: 362-380.
  4. McHugh KR, Barlow DH (2010) The dissemination and implementation of evidence-based psychological treatments: A review of current efforts. Am Psychol 65: 73-84.
  5. Das R (2016) Five technologies that will disrupt healthcare by 2020. Pharma & Healthcare.
  6. Adlaf EM, Gliksman L, Demers A, Newton-Taylor B (2001) The prevalence of elevated psychological distress among Canadian undergraduates: Findings from the 1998 Canadian Campus Survey. J Am Coll Health 50: 67-72.
  7. Raghupathi W, Raghupathi V (2014) Big data analytics in healthcare: promise and potential. Health Inf Sci Syst 2: 3.
  8. Marshall T, Champagne-Langabeer T, Castelli D (2017) Cognitive computing and eScience in health and life science research: artificial intelligence and obesity intervention programs. Health Inf Sci Syst 5: 13.
  9. Patel NM, Michelini VV, Snell JM, Balu S, Hoyle AP, et al. (2018) Enhancing next-generation sequencing-guided cancer care through cognitive computing. The oncologist 23: 179-185.
  10. Begg S, Vos T, Barker B, Stevenson C, Stanley L, et al. (2007) The burden of disease and injury in Australia 2003. PHE 82 Canberra: AIHW.
  11. World Health Organization (2008) The global burden of disease: 2004 update. WHO Press, Geneva.
  12. Shuchman M (2007) Falling through the cracks–Virginia Tech and the restructuring of college mental health services. N Engl J Med 357: 105-110.
  13. McGinnis RS, McGinnis EW, Hruschak J, Lopez-Duran NL, Fitzgerald K, et al. (2018) Wearable sensors and machine learning diagnose anxiety and depression in young children. IEEE EMBS International Conference on Biomedical & Health Informatics: 410-413.
  14. Khawaja NG, Dempsey J (2007) Psychological distress in international university students: An Australian study. Aust J Guid Counsell 17: 13-27.
  15. Adlaf EM, Gliksman L, Demers A, Newton-Taylor B (2001) The prevalence of elevated psychological distress among Canadian undergraduates: Findings from the 1998 Canadian campus survey. J Am Coll Health 50: 67–72.
  16. Nerdrum P, Rustøen T, Rønnestad MH (2006) Student psychological distress: A psychometric study of 1750 Norwegian 1st-year undergraduate students. Scand. J. Educ. Res 50: 95-109.
  17. Du J, Li Y, Niu J, Zhang H (2008) Correlative study on mental health, coping styles and posttraumatic stress response among medical students. Chinese Journal of Clinical Psychology 16: 183-185.
  18. Connell J, Barkham M, Mellor-Clark J (2007) CORE-OM mental health norms of students attending university counselling services benchmarked against an age-matched primary care sample. Br J Guid Counc 35: 41-57.
  19. Vardhini P, Geethanjali B, Veezhinathan M (2017) Effect of relaxation stimuli on mental task performance. 2017 Fourth International Conference In Signal Processing, Communication and Networking (ICSCN): 1-5.
  20. Eisenberg D, Gollust SE, Golberstein E, Hefner JL (2007) Prevalence and correlates of depression, anxiety, and suicidality among university students. Am J Orthopsychiatry 77: 534-542.
  21. Chiao C, Chiu CH (2016) The relationships between ICT use and life quality among children with social phobia. 2016 IEEE International Conference on In Teaching, Assessment, and Learning for Engineering (TALE): 26-31.
  22. Stallman HM (2010) Psychological distress in university students: A comparison with general population data. Australian Psychologist 45: 249-257.
  23. Roy R, Guha R, Bhattacharya SD, Mukhopadhyay J (2017) Building a web based cognitive restructuring program for promoting resilience in a college campus. 2017 9th International Conference on In Communication Systems and Networks (COMSNETS). pp: 520-524.
  24. Adlaf EM, Gliksman L, Demers A, Newton-Taylor B (2001) The prevalence of elevated psychological distress among Canadian undergraduates: Findings from the 1998 Canadian Campus Survey. J Am Coll Health 50: 67-72.
Citation: Jabeen SH, Sandhu ME, Tahir F, Amadullah M, Noureen N, et al. (2018) Cognitive Behavioural Therapy to Compute Psychological Illness of Students by Using an Expert System. J Depress Anxiety 7: 315.

Copyright: © 2018 Jabeen SH, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Top