Banca de QUALIFICAÇÃO: MARCILIO DE OLIVEIRA MEIRA

Uma banca de QUALIFICAÇÃO de DOUTORADO foi cadastrada pelo programa.
STUDENT : MARCILIO DE OLIVEIRA MEIRA
DATE: 31/07/2024
TIME: 10:00
LOCAL: Remoto
TITLE:

ADHD DIAGNOSIS: A EFFICIENT METHOD BASED ON FUNCTIONAL NETWORKS AND MACHINE LEARNING

KEY WORDS:

ADHD; diagnosis; SPECT; magnetic resonance imaging; machine learning.

PAGES: 100
BIG AREA: Ciências Exatas e da Terra
AREA: Ciência da Computação
SUBÁREA: Metodologia e Técnicas da Computação
SPECIALTY: Sistemas de Informação
SUMMARY:

Attention-deficit/hyperactivity disorder (ADHD) is a neurobiological disorder in which the patient suffers from a persistent pattern of inattention and/or hyperactivity/impulsivity that causes harm in various settings, in school, personal and professional life. It is also considered one of the most heterogeneous disorders that can co-occur with other disorders and has three possible subtypes/presentations: predominantly inattentive, predominantly hyperactive-impulsive and the combined subtype. As with other neuropsychiatric disorders, diagnosis is based on methods whose assessment criteria are subjective and inconsistent, raising doubts about their reliability. This problem also extends to research on this disorder, as the databases created based on these methods can be riddled with noise when reporting the diagnosis. In addition, patients and their families face a lengthy investigation process before a diagnosis is made. However, there are already indications that brain changes detected in rs-fMRI (resting state functional magnetic resonance imaging) and SPECT (single-photon emission computed tomography) images of affected individuals could help in the search for new diagnostic methods. At the same time, altered functional connectivity in functional networks has often been observed in ADHD. Under the premise of transferring the knowledge about the disorder into clinical practice and creating more flexible and reliable diagnostic possibilities, the aim of this work is to propose an efficient method for the diagnosis of ADHD. One of its distinguishing features is the unprecedented combination of SPECT and rs-fMRI brain imaging modalities. To this end, the approach utilizes data from ROIs/networks of executive control and the limbic system, functional connectivity, machine learning methods and statistical techniques. It is therefore expected that the method will become an ally in clinical practice, complementing and supporting diagnoses with conventional methods, with future reports informing on the predicted value compared to the reference value. The simplicity of the method allows the addition of other imaging techniques, increasing the reliability of a possible automatic diagnosis. It is believed that it can also be extended to the study of other neuropsychiatric disorders by simply adapting it to specific ROIs/networks and an image database.

COMMITTEE MEMBERS:
Presidente - 1350250 - ANNE MAGALY DE PAULA CANUTO
Interno - 2177445 - BRUNO MOTTA DE CARVALHO
Externo à Instituição - MARCELO DAMASCENO DE MELO - IFRN