aspg-3-4052-1791419321 20261008002841 American Scientific Publishing Group admin@americaspg.com American Scientific Publishing Group Fusion: Practice and Applications FPA 2770-0070 2692-4048 2026 21 2 Hayder M Hani Collage of Computer Engineering , University of Technology, Iraq, Baghdad, Iraq Ahmed Musa Dinar Collage of Computer Engineering , University of Technology, Iraq, Baghdad, Iraq 16.00 Normal false false false TR X-NONE AR-SA DefSemiHidden="false" DefQFormat="false" DefPriority="99" LatentStyleCount="371"> UnhideWhenUsed="true" QFormat="true" Name="heading 2"/> UnhideWhenUsed="true" QFormat="true" Name="heading 3"/> UnhideWhenUsed="true" QFormat="true" Name="heading 4"/> UnhideWhenUsed="true" QFormat="true" Name="heading 5"/> UnhideWhenUsed="true" QFormat="true" Name="heading 6"/> UnhideWhenUsed="true" QFormat="true" Name="heading 7"/> UnhideWhenUsed="true" QFormat="true" Name="heading 8"/> UnhideWhenUsed="true" QFormat="true" Name="heading 9"/> Name="index 1"/> Name="index 2"/> Name="index 3"/> Name="index 4"/> Name="index 5"/> Name="index 6"/> Name="index 7"/> Name="index 8"/> Name="index 9"/> UnhideWhenUsed="true" Name="toc 1"/> UnhideWhenUsed="true" Name="toc 2"/> UnhideWhenUsed="true" Name="toc 3"/> UnhideWhenUsed="true" Name="toc 4"/> UnhideWhenUsed="true" Name="toc 5"/> UnhideWhenUsed="true" Name="toc 6"/> UnhideWhenUsed="true" Name="toc 7"/> UnhideWhenUsed="true" Name="toc 8"/> UnhideWhenUsed="true" Name="toc 9"/> Name="Normal Indent"/> Name="footnote text"/> Name="annotation text"/> Name="header"/> Name="footer"/> Name="index heading"/> UnhideWhenUsed="true" QFormat="true" Name="caption"/> Name="table of figures"/> Name="envelope address"/> Name="envelope return"/> Name="footnote reference"/> Name="annotation reference"/> Name="line number"/> Name="page number"/> Name="endnote reference"/> Name="endnote text"/> Name="table of authorities"/> Name="macro"/> Name="toa heading"/> Name="List"/> Name="List Bullet"/> Name="List Number"/> Name="List 2"/> Name="List 3"/> Name="List 4"/> Name="List 5"/> Name="List Bullet 2"/> Name="List Bullet 3"/> Name="List Bullet 4"/> Name="List Bullet 5"/> Name="List Number 2"/> Name="List Number 3"/> Name="List Number 4"/> Name="List Number 5"/> Name="Closing"/> Name="Signature"/> UnhideWhenUsed="true" Name="Default Paragraph Font"/> Name="Body Text"/> Name="Body Text Indent"/> Name="List Continue"/> Name="List Continue 2"/> Name="List Continue 3"/> Name="List Continue 4"/> Name="List Continue 5"/> Name="Message Header"/> Name="Salutation"/> Name="Date"/> Name="Body Text First Indent"/> Name="Body Text First Indent 2"/> Name="Note Heading"/> Name="Body Text 2"/> Name="Body Text 3"/> Name="Body Text Indent 2"/> Name="Body Text Indent 3"/> Name="Block Text"/> Name="Hyperlink"/> Name="FollowedHyperlink"/> Name="Document Map"/> Name="Plain Text"/> Name="E-mail Signature"/> Name="HTML Top of Form"/> Name="HTML Bottom of Form"/> Name="Normal (Web)"/> Name="HTML Acronym"/> Name="HTML Address"/> Name="HTML Cite"/> Name="HTML Code"/> Name="HTML Definition"/> Name="HTML Keyboard"/> Name="HTML Preformatted"/> Name="HTML Sample"/> Name="HTML Typewriter"/> Name="HTML Variable"/> Name="Normal Table"/> Name="annotation subject"/> Name="No List"/> Name="Outline List 1"/> Name="Outline List 2"/> Name="Outline List 3"/> Name="Table Simple 1"/> Name="Table Simple 2"/> Name="Table Simple 3"/> Name="Table Classic 1"/> Name="Table Classic 2"/> Name="Table Classic 3"/> Name="Table Classic 4"/> Name="Table Colorful 1"/> Name="Table Colorful 2"/> Name="Table Colorful 3"/> Name="Table Columns 1"/> Name="Table Columns 2"/> Name="Table Columns 3"/> Name="Table Columns 4"/> Name="Table Columns 5"/> Name="Table Grid 1"/> Name="Table Grid 2"/> Name="Table Grid 3"/> Name="Table Grid 4"/> Name="Table Grid 5"/> Name="Table Grid 6"/> Name="Table Grid 7"/> Name="Table Grid 8"/> Name="Table List 1"/> Name="Table List 2"/> Name="Table List 3"/> Name="Table List 4"/> Name="Table List 5"/> Name="Table List 6"/> Name="Table List 7"/> Name="Table List 8"/> Name="Table 3D effects 1"/> Name="Table 3D effects 2"/> Name="Table 3D effects 3"/> Name="Table Contemporary"/> Name="Table Elegant"/> Name="Table Professional"/> Name="Table Subtle 1"/> Name="Table Subtle 2"/> Name="Table Web 1"/> Name="Table Web 2"/> Name="Table Web 3"/> Name="Balloon Text"/> Name="Table Theme"/> Name="List Paragraph"/> Name="Intense Quote"/> Name="Subtle Emphasis"/> Name="Intense Emphasis"/> Name="Subtle Reference"/> Name="Intense Reference"/> UnhideWhenUsed="true" Name="Bibliography"/> UnhideWhenUsed="true" QFormat="true" Name="TOC Heading"/> Name="Grid Table 1 Light Accent 1"/> Name="Grid Table 6 Colorful Accent 1"/> Name="Grid Table 7 Colorful Accent 1"/> Name="Grid Table 1 Light Accent 2"/> Name="Grid Table 6 Colorful Accent 2"/> Name="Grid Table 7 Colorful Accent 2"/> Name="Grid Table 1 Light Accent 3"/> Name="Grid Table 6 Colorful Accent 3"/> Name="Grid Table 7 Colorful Accent 3"/> Name="Grid Table 1 Light Accent 4"/> Name="Grid Table 6 Colorful Accent 4"/> Name="Grid Table 7 Colorful Accent 4"/> Name="Grid Table 1 Light Accent 5"/> Name="Grid Table 6 Colorful Accent 5"/> Name="Grid Table 7 Colorful Accent 5"/> Name="Grid Table 1 Light Accent 6"/> Name="Grid Table 6 Colorful Accent 6"/> Name="Grid Table 7 Colorful Accent 6"/> Name="List Table 1 Light Accent 1"/> Name="List Table 6 Colorful Accent 1"/> Name="List Table 7 Colorful Accent 1"/> Name="List Table 1 Light Accent 2"/> Name="List Table 6 Colorful Accent 2"/> Name="List Table 7 Colorful Accent 2"/> Name="List Table 1 Light Accent 3"/> Name="List Table 6 Colorful Accent 3"/> Name="List Table 7 Colorful Accent 3"/> Name="List Table 1 Light Accent 4"/> Name="List Table 6 Colorful Accent 4"/> Name="List Table 7 Colorful Accent 4"/> Name="List Table 1 Light Accent 5"/> Name="List Table 6 Colorful Accent 5"/> Name="List Table 7 Colorful Accent 5"/> Name="List Table 1 Light Accent 6"/> Name="List Table 6 Colorful Accent 6"/> Name="List Table 7 Colorful Accent 6"/> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin-top:0in; mso-para-margin-right:0in; mso-para-margin-bottom:8.0pt; mso-para-margin-left:0in; line-height:107%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri",sans-serif; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:Arial; mso-bidi-theme-font:minor-bidi; mso-ansi-language:TR;} Early and accurate diagnosis of Autism Spectrum Disorder (ASD) using neuroimaging has become increasingly viable with the advent of deep learning (DL) technologies. Current clinical diagnostic processes for ASD are largely subjective and time-intensive, creating an urgent need for objective diagnostic tools. This study presents a comprehensive comparison of three prominent functional Magnetic Resonance Imaging (fMRI) feature extraction methods, ALFF (Amplitude of Low-Frequency Fluctuations), fALFF (fractional ALFF), and ReHo (Regional Homogeneity), alongside structural Magnetic Resonance Imaging (sMRI) data, to evaluate their effectiveness in classifying ASD using various deep learning architectures. Preprocessed data from the ABIDE dataset were utilized, with uniform preprocessing pipelines applied, followed by feature extraction using the AAL (Automated Anatomical Labeling) atlas. Synthetic data augmentation was performed using Generative Adversarial Networks (GANs) to mitigate class imbalance. We trained and tuned multiple models, including 1-dimensional Convolutional Neural Networks (1D CNNs) with multi-head attention, Long Short-Term Memory (LSTM), and Vision Transformers (ViTs), with and without hyperparameter optimization. The findings indicate that the highest classification performance was attained using ALFF features with a hyperparameter-optimized CNN enhanced by attention mechanisms, achieving an accuracy of 0.83. Similarly, ReHo features yielded an equal accuracy of 0.83 when analyzed using a Vision Transformer (ViT) model. Across all experiments, functional neuroimaging features consistently outperformed structural features in classifying ASD. Notably, systematic hyperparameter tuning led to substantial improvements, particularly for ALFF-based models, where accuracy increased markedly from 59% to 83% using the CNN+Attention architecture. This study presents a comprehensive evaluation of feature types and model architectures across neuroimaging modalities, offering critical insights into their relative diagnostic value for ASD. The achieved accuracy of 83% using both ALFF and ReHo features marks a meaningful advancement in the field, setting realistic benchmarks for future research while adhering to stringent methodological rigor. 2026 149 158 4052 https://creativecommons.org/licenses/by/4.0/ 10.54216/FPA.210209 https://www.americaspg.com/journal/3/article/4052 https://www.americaspg.com/storage/81756577808.pdf