Modified Cuckoo Search-Cascade Forest (MCS-CF) for Attention Deficit Hyperactivity Disorder (ADHD) Diagnosis

• Published in: NeuroQuantology Vol. 18; no. 7; pp. 83 - 94
• Main Authors: Padmavathy, Dr.T.V., Vinothkumar, Dr.M., Vimal Kumar, Dr.M.N.
• Format: Journal Article
• Language: English
• Published: Bornova Izmir NeuroQuantology 2020
• Subjects: Accuracy; Artificial neural networks; Attention deficit hyperactivity disorder; Children; Classification; Decision trees; Feature extraction; Feature selection; Image classification; Machine learning; Magnetic resonance imaging; Preprocessing; Search algorithms
• DOI: 10.14704/nq.2020.18.7.NQ20196
• ISSN: 1303-5150

Attention deficit hyperactivity disorder (ADHD) is a disease state of the mind which is frequently observed in young children. Different machine learning approaches, which include Deep Neural Networks (DNNs) and it helps in ADHD classification. The following have been recently proposed: ADHD to examine employing functional Magnetic Resonance Imaging (fMRI) information and gcForest to differentiate between ADHD and normal theme, cascade forest is employed to make use of the concatenated feature vector samples in the form of input for classification. But, classification accuracy takes large time consuming. In order to deal with this problem, Modified Cuckoo Search- Cascade Forest (MCS-CF) based feature selection algorithm is suggested which helps in the accuracy improvement of the classifier used in ADHD.