Diagnosis of psychiatric patients using data mining techniques

ISRCTN	ISRCTN75534069
DOI	https://doi.org/10.1186/ISRCTN75534069
Secondary identifying numbers	N/A

Submission date: 05/02/2016
Registration date: 29/02/2016
Last edited: 23/01/2019

Recruitment status: No longer recruiting
Overall study status: Completed
Condition category: Mental and Behavioural Disorders

Prospectively registered

Protocol

Statistical analysis plan

Results

Individual participant data

Plain English summary of protocol

Background and study aims
Mental illness is a general term used to describe a range of conditions that affect mood, thoughts and behavior. They are extremely common and it is thought that most people come into contact with mental illness, either directly or indirectly, at some point in their lives. Unlike many physical illnesses, they can be difficult to diagnose as the symptoms can greatly vary from person to person. Data mining is a technique which uses a procedure or formula to solve a problem (algorithm). It is possible that by collecting and analyzing data from people suffering from a mental illness and those who treat them, an algorithm could be used in order make diagnosing patients easier. The aim of this study is to look at a range of different data mining techniques in order to find the best one for helping the diagnosis of mental illness.

Who can participate?
Adults with a mental illness who are being treated at El Mamoura Hospital and the doctors looking after them.

What does the study involve?
Doctors interview a number of in-patients with mental health problems who are being treated at El Mamoura Hospital using a specially designed survey. The data from these surveys is then entered into a computer so that it can be analysed using a number of different data mining techniques. This involves putting the data into different formulas so that any patterns can be identified by the different algorithms. At the end of the study, the accuracy of these techniques are compared by seeing how well the results of the techniques matches with the patients current diagnosis.

What are the possible benefits and risks of participating?
There are no direct benefits or risks to participants taking part in this study.

Where is the study run from?
El Mamoura Hospital for Psychiatric and Addiction Treatment (Egypt)

When is the study starting and how long is it expected to run for?
September 2015 to November 2015

Who is funding the study?
Self-funding

Who is the main contact?
Ms Horeya Abou Warda

Contact information

Ms Horeya Abou Warda
Scientific

Arab Academy for Science, Technology, and Maritime Transport
Abu Quir
Alexandria
21524
Egypt

ORCID ID	0000-0003-0957-4011

Study information

Study design	Single-centre observational cross-sectional study
Primary study design	Interventional
Secondary study design	Randomised controlled trial
Study setting(s)	Hospital
Study type	Diagnostic
Participant information sheet	Not available in web format, please use the contact details below to request a patient information sheet.
Scientific title	Diagnosis of psychiatric patients using data mining techniques
Study objectives	The aim of this study is to identify the best data mining technique to facilitate psychiatric diagnosis.
Ethics approval(s)	General Secretariat of Mental Health and Addiction Treatment (GSMHAT), Ministry of Health and Population (Egypt), 17/08/2015, ref: 1860\|8/17
Health condition(s) or problem(s) studied	Mental disorders
Intervention	Clinicians interview a number of inpatients at El Mamoura Hospital that are known to have a psychiatric condition, assessing them with the help of a survey that the clinicians complete. This data is then used to test a number of different data mining techniques (i.e. Random Forest, Decision Tree, Neural Networks, K Nearest Neighbour and Support Vector Machines) to see which was the best one to use to create an automated model for a diagnosing psychiatric conditions.
Intervention type	Behavioural
Primary outcome measure	Identification of the best data mining technique is measured using the accuracy percentage of data mining classification (10-fold cross-validation) after data was collected.
Secondary outcome measures	Classification of psychotic diseases are measured using the accuracy percentage of data mining classification (10-fold cross-validation) after data was collected.
Overall study start date	01/09/2015
Completion date	30/11/2015

Eligibility

Participant type(s)	Patient
Age group	Adult
Sex	Both
Target number of participants	1,800 patients.
Key inclusion criteria	Patients: 1. Adults 2. Diagnosed with a psychiatric condition (major depression, drug-induced psychosis, schizophrenia, schizoaffective disorder, obsessive – compulsive disorder, bipolar disorder - manic episode and bipolar disorder - mixed episode) 3. Being treated as an inpatient at El Mamoura Hospital 4. Not taking any psychoactive substances Clinicians: Clinicians working at the El Mamoura Hospital for psychiatric and addiction treatment.
Key exclusion criteria	Patients: 1. Not being treated as an inpatient at El Mamoura Hospital 2. Taking psychoactive substances Clinicians: Not working at El Mamoura Hospital for psychiatric and addiction treatment
Date of first enrolment	15/05/2015
Date of final enrolment	17/08/2015

Locations

Countries of recruitment

Egypt

Study participating centre

El Mamoura Hospital for Psychiatric and Addiction Treatment

Alexandria
21912
Egypt

Sponsor information

Arab Academy for Science, Technology, and Maritime Transport (Egypt)
University/education

Abu Quir
Alexandria
21625
Egypt

Phone	+20 19838
Email	ccit@aast.edu
Website	http://www.aast.edu/en/colleges/ccit/
"ROR"	https://ror.org/0004vyj87

Funders

Funder type

Other

No information available

Results and Publications

Intention to publish date
Individual participant data (IPD) Intention to share	Yes
IPD sharing plan summary	Available on request
Publication and dissemination plan	Planned publication in an academic journal.
IPD sharing plan

Study outputs

Output type	Details	Date created	Date added	Peer reviewed?	Patient-facing?
Results article	results	18/10/2016	23/01/2019	Yes	No

Editorial Notes

23/01/2019: Publication reference added
22/01/2019: The following changes have been made to the publication record:
1. The overall trial start date has been changed from 17/08/2015 to 01/09/2015
2. The funder, Arab Academy for Science, Technology, and Maritime Transport (Egypt), has been removed
4. The plain English summary has been updated to reflect the changes above