Jerzy Gomuła

Classification of MMPI profiles of patients with mental disorders: experiments with attribute reduction and extension

In the research presented in the paper we try to find efficient methods for classification of MMPI profiles of patients with mental disorders. Each profile is described by a set of values of thirteen attributes (scales). Patients can be classified into twenty categories concerning nosological types. It is possible to improve classification accuracy by reduction or extension of the number of attributes with relation to the original data table. We test several techniques of reduction and extension. Experiments show that the proposed attribute extension approach improves classification accuracy, especially in the case of discretized data.

A preliminary attempt to rules generation for mental disorders

Our research concerns psychometric data coming from the Minnesota Multiphasic Personality Inventory (MMPI) test. MMPI is used to count the personality-psychometric dimensions which help in diagnosis of mental diseases. In this paper, we test different algorithms for rule generation from data coming from the MMPI-WISKAD test. Especially, for each algorithm, we take into consideration a number of generated rules and their ability for classification of new cases. Results are important for selection of suitable algorithms that will create a base for a computer tool supporting diagnosis of mental diseases.

Computer-aided diagnosis of patients with mental disorders using the Copernicus system

In the paper, a general overview of the Copernicus system is given. Copernicus is a tool for computer-aided diagnosis of mental disorders based on personality inventories. The tool was designed for the Java platform. There have been selected and implemented different quantitative groups of methods useful for differential interprofile diagnosis. The aim of this paper is to present the capabilities of the software tool created for effective and specialized analysis of psychometric data to support psychologists in diagnostic decision making.

Rule-Based Classification of MMPI Data of Patients with Mental Disorders: Experiments with Basic and Extended Profiles

In the research presented in the paper, we test the classification potential for different configurations of the MMPI data used for nosological diagnosis of patients with mental disorders. Originally, each patient is described by an MMPI profile consisting of a set of values of thirteen attributes (the so-called scales). A profile can be extended by different specialized indexes defined in the professional domain literature. Each created index has been proposed by practitioners and has a proper meaning from the clinical point of view. Adding such indexes leads to the extension of the attribute space for cases to be classified. In the paper, we present results of experiments with basic (original) and extended profiles.

Rule-Based Analysis of MMPI Data Using the Copernicus System

Our research concerns psychometric data coming from the Minnesota Multiphasic Personality Inventory (MMPI) test. MMPI is used to count the personality-psychometric dimensions which help specialists in diagnosis of mental diseases. In this paper, we present a part of the Copernicus system – the tool for computer-aided diagnosis of mental diseases based on personality inventories. This part is devoted to the rule-based analysis of the MMPI data expressed in the form of the so-called profiles. The paper characterizes the knowledge base embodied in Copernicus which can be used for the rule-based analysis of the patients’ MMPI data as well as the functionality of the designed tool.

Consistency-Based Preprocessing for Classification of Data Coming from Evaluation Sheets of Subjects with ASDs

In general, the aim of our research is to adapt computational intelligence methods for computer-aided decision support in diagnosis and therapy of persons with Autism Spectrum Disorders (ASDs). In the paper, we are focusing on the data preprocessing step for cleaning a training data set for classifiers. An approach based on consistency factors is proposed.

Multiway classification of MMPI profiles

In the paper, an overview of the Copernicus system from the classification ability perspective is given. Copernicus is a tool for computer-aided diagnosis of mental disorders based on data coming from the Minnesota Multiphasic Personality Inventory (MMPI) test. MMPI is one of the most frequently used personality tests in clinical mental health as well as psychopathology (mental and behavioural disorders). A variety of classification ways useful for differential interprofile diagnosis is listed and described in this paper.

Ant based clustering of MMPI data: an experimental study

Our research concerns psychometric data coming from the Minnesota Multiphasic Personality Inventory (MMPI) test. MMPI is one of the most frequently used in clinical mental health personality tests as well as psychopathology (mental and behavioral disorders). We are developing the Copernicus system. It is a tool for the analysis of MMPI profiles of patients with mental disorders. In this system, there have been selected and implemented different quantitative groups of methods useful for differential interprofile diagnosis. In the paper, we investigate clustering of MMPI data using one of nature-inspired heuristic approaches - ant based clustering. For this approach, we test usefulness of different dissimilarity measures of MMPI profiles both standard and those defined in the professional domain literature.

Random forest feature selection for data coming from evaluation sheets of subjects with ASDs

We deal with the problem of initial analysis of data coming from evaluation sheets of subjects with Autism Spectrum Disorders (ASDs). In our research, we use an original evaluation sheet including questions about competencies grouped into 17 spheres. In the paper, we are focused on a feature selection problem. The main goal is to use appropriate data to build simpler and more accurate classifiers. The feature selection method based on random forest is used.

A Tool for Computer-Aided Diagnosis of Psychological Disorders Based on the MMPI Test: An Overview

The goal of the paper is to summarize our research, conducted for over the last five years, on computer-aided diagnosis of patients screened with the MMPI (Minnesota Multiphasic Personality Inventory) test. The MMPI test delivers psychometric data, in the form of the so-called profiles (thirteen descriptive attributes corresponding to three validity and ten clinical scales), enabling us to diagnose selected psychological disorders. The notable effect of conducted research is a new computer tool, called Copernicus, aiding diagnosis of psychological disorders based on MMPI profiles. The paper is focused on outlying the functionality of the created tool.