A Data-driven Approach for the Classification and Severity Estimation of Behavioral Disorders
Presenter Type
UNO Graduate Student (Doctoral)
Major/Field of Study
Computer Science
Other
Information Technology
Author ORCID Identifier
0000-0002-4986-5027
Advisor Information
Dr. Hesham H. Ali
Location
MBSC302 - G (Doctoral)
Presentation Type
Oral Presentation
Start Date
24-3-2023 9:00 AM
End Date
24-3-2023 10:15 AM
Abstract
Behavioral disorders such as depression and schizophrenia are challenging to diagnose due to their wide range of symptoms, and assessing their severity is also difficult. Current diagnostic methods primarily rely on patient self-reporting or clinician observations, which have limitations. Therefore, there is an urgent need for more objective and analytical methods to assist clinicians in developing more effective approaches to studying these disorders. In our study, we propose a new data-driven approach that uses publicly available databases to differentiate between multiple behavioral disorders and a novel index to assess disorder severity. We analyze mobility data from 72 participants with multiple behavioral disorders and construct a graph-based correlation network model to identify different groups of people suffering from specific disorders, such as depression and schizophrenia. We use graph properties to develop a new index which can discriminate between different levels of severity among patients. Our results demonstrate that the proposed approach succeeds in achieving both goals related to the classification and severity estimation, and that the obtained analysis provides useful information for healthcare providers to better diagnose behavioral conditions and propose data-driven treatment options.
Scheduling
9:15-10:30 a.m., 10:45 a.m.-Noon
A Data-driven Approach for the Classification and Severity Estimation of Behavioral Disorders
MBSC302 - G (Doctoral)
Behavioral disorders such as depression and schizophrenia are challenging to diagnose due to their wide range of symptoms, and assessing their severity is also difficult. Current diagnostic methods primarily rely on patient self-reporting or clinician observations, which have limitations. Therefore, there is an urgent need for more objective and analytical methods to assist clinicians in developing more effective approaches to studying these disorders. In our study, we propose a new data-driven approach that uses publicly available databases to differentiate between multiple behavioral disorders and a novel index to assess disorder severity. We analyze mobility data from 72 participants with multiple behavioral disorders and construct a graph-based correlation network model to identify different groups of people suffering from specific disorders, such as depression and schizophrenia. We use graph properties to develop a new index which can discriminate between different levels of severity among patients. Our results demonstrate that the proposed approach succeeds in achieving both goals related to the classification and severity estimation, and that the obtained analysis provides useful information for healthcare providers to better diagnose behavioral conditions and propose data-driven treatment options.