An Umbrella Review of the Fusion of fMRI and AI in Autism

Diagnostics, Journal Year: 2023, Volume and Issue: 13(23), P. 3552 - 3552

Published: Nov. 28, 2023

The role of functional magnetic resonance imaging (fMRI) is assuming an increasingly central in autism diagnosis. integration Artificial Intelligence (AI) into the realm applications further contributes to its development. This study’s objective analyze emerging themes this domain through umbrella review, encompassing systematic reviews. research methodology was based on a structured process for conducting literature narrative using review PubMed and Scopus. Rigorous criteria, standard checklist, qualification were meticulously applied. findings include 20 reviews that underscore key research, particularly emphasizing significance technological integration, including pivotal roles fMRI AI. study also highlights enigmatic oxytocin. While acknowledging immense potential field, outcome does not evade significant challenges limitations. Intriguingly, there growing emphasis innovation AI, whereas aspects related healthcare processes, such as regulation, acceptance, informed consent, data security, receive comparatively less attention. Additionally, these Personalized Medicine (PM) represents promising yet relatively unexplored area within research. concludes by encouraging scholars focus critical health vital routine implementation applications.

Language: Английский

---

Prioritizing complex health levels beyond autism triage using fuzzy multi-criteria decision-making

Complex & Intelligent Systems, Journal Year: 2024, Volume and Issue: 10(5), P. 6159 - 6188

Published: June 4, 2024

Abstract This study delves into the complex prioritization process for Autism Spectrum Disorder (ASD), focusing on triaged patients at three urgency levels. Establishing a dynamic solution is challenging resolving conflicts or trade-offs among ASD criteria. research employs fuzzy multi-criteria decision making (MCDM) theory across four methodological phases. In first phase, identifies dataset, considering 19 critical medical and sociodemographic criteria The second phase introduces new Decision Matrix (DM) designed to manage effectively. third focuses extension of Fuzzy-Weighted Zero-Inconsistency (FWZIC) construct weights using Single-Valued Neutrosophic 2-tuple Linguistic (SVN2TL). fourth formulates Multi-Attributive Border Approximation Area Comparison (MABAC) method rank within each level. Results from SVN2TL-FWZIC offer significant insights, including higher values "C12 = Laughing no reason" "C16 Notice sound bell" with 0.097358 0.083832, indicating their significance in identifying potential symptoms. base prioritizing triage levels MABAC, encompassing behavioral dimensions. methodology undergoes rigorous evaluation through sensitivity analysis scenarios, confirming consistency results points. compares benchmark studies, distinct points, achieves remarkable 100% congruence these prior investigations. implications this are far-reaching, offering valuable guide clinical psychologists cases patients.

Language: Английский

---

Fuzzy Evaluation and Benchmarking Framework for Robust Machine Learning Model in Real-Time Autism Triage Applications

International Journal of Computational Intelligence Systems, Journal Year: 2024, Volume and Issue: 17(1)

Published: June 17, 2024

Abstract In the context of autism spectrum disorder (ASD) triage, robustness machine learning (ML) models is a paramount concern. Ensuring ML faces issues such as model selection, criterion importance, trade-offs, and conflicts in evaluation benchmarking models. Furthermore, development must contend with two real-time scenarios: normal tests adversarial attack cases. This study addresses this challenge by integrating three key phases that bridge domains fuzzy multicriteria decision-making (MCDM). First, utilized dataset comprises authentic information, encompassing 19 medical sociodemographic features from 1296 autistic patients who received diagnoses via intelligent triage method. These were categorized into one labels: urgent, moderate, or minor. We employ principal component analysis (PCA) algorithms to fuse large number features. Second, fused forms basis for rigorously testing eight models, considering scenarios, evaluating classifier performance using nine metrics. The third phase developed robust framework encompasses creation decision matrix (DM) 2-tuple linguistic Fermatean opinion score method (2TLFFDOSM) multiple-ML perspectives, accomplished through individual external group aggregation ranks. Our findings highlight effectiveness PCA algorithms, yielding 12 components acceptable variance. ranking, logistic regression (LR) emerged top-performing terms 2TLFFDOSM (1.3370). A comparative five benchmark studies demonstrated superior our across all six checklist comparison points.

Language: Английский

---

Hybrid Diagnosis Models for Autism Patients Based on Medical and Sociodemographic Features Using Machine Learning and Multicriteria Decision-Making (MCDM) Techniques: An Evaluation and Benchmarking Framework

Computational and Mathematical Methods in Medicine, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 19

Published: Nov. 16, 2022

Background and Contexts. Autism spectrum disorder (ASD) is difficult to diagnose, prompting researchers increase their efforts find the best diagnosis by introducing machine learning (ML). Recently, several available challenges issues have been highlighted for of ASD. High consideration must be taken into feature selection (FS) approaches classification process simultaneously using medical tests sociodemographic characteristic features in autism diagnostic. The constructed ML models neglected importance a training evaluation dataset, especially since some different contributions processing data possess more relevancies information than others. However, role physician’s experience towards remains limited. In addition, presence many criteria, criteria trade-offs, categorize benchmarking concerning intersection between FS methods given under complex multicriteria decision-making (MCDM) problems. To date, no study has presented an framework hybrid classify patients’ emergency levels considering solutions. Method. three-phase integrated MCDM develop evaluate benchmark best. Firstly, new ASD-dataset-combined identified preprocessed. Secondly, developing three techniques five algorithms introduces 15 models. selected from each technique are weighted before feeding fuzzy-weighted zero-inconsistency (FWZIC) method based on four psychiatry experts. Thirdly, (i) formulate dynamic decision matrix all developed seven metrics, including accuracy, precision, F1 score, recall, test time, train AUC. (ii) fuzzy opinion score (FDOSM) used metrics. Results. Results reveal that obtained size others number features; sets were 39, 38, 41 out 48 features. Each set its weights FWIZC. Considered mostly within techniques. first “ReF-decision tree,” “IG-decision “Chi2-decision with values 0.15714, 0.17539, 0.29444. model (ReF-decision tree) 0.4190, 0.0030, 0.9946, 0.9902, 0.9951 C1=train C2=test C3=AUC, C4=CA, C5=F1 C6=precision, C7=recall, respectively. would beneficial advancing, accelerating, selecting tools therapy can identify severity as light, medium, or intense

Language: Английский

---

Towards Trustworthy Myopia Detection: Integration Methodology of Deep Learning Approach, XAI Visualization, and User Interface System

Applied Data Science and Analysis, Journal Year: 2023, Volume and Issue: unknown, P. 1 - 15

Published: Feb. 23, 2023

Myopia, a prevalent vision disorder with potential complications if untreated, requires early and accurate detection for effective treatment. However, traditional diagnostic methods often lack trustworthiness explainability, leading to biases mistrust. This study presents four-phase methodology develop robust myopia system. In the initial phase, dataset containing training testing images is located, preprocessed, balanced. Subsequently, two models are deployed: pre-trained VGG16 model renowned image classification tasks, sequential CNN convolution layers. Performance evaluation metrics such as accuracy, recall, F1-Score, sensitivity, logloss utilized assess models' effectiveness. The third phase integrates trustworthiness, transparency through application of Explainable Artificial Intelligence (XAI) techniques. Specifically, Local Interpretable Model-Agnostic Explanations (LIME) employed provide insights into decision-making process deep learning model, offering explanations myopic or normal. final user interface implemented XAI bringing together aforementioned phases. outcomes this contribute advancement objective explainable in field detection. Notably, achieves an impressive accuracy 96%, highlighting its efficacy diagnosing myopia. LIME results valuable interpretations cases. proposed enhances transparency, interpretability, trust process.

Language: Английский

---

Explainable Artificial Intelligence Multimodal of Autism Triage Levels Using Fuzzy Approach-Based Multi-criteria Decision-Making and LIME

International Journal of Fuzzy Systems, Journal Year: 2023, Volume and Issue: 26(1), P. 274 - 303

Published: Nov. 17, 2023

Language: Английский

---

Fuzzy Decision‐Making Framework for Evaluating Hybrid Detection Models of Trauma Patients

Expert Systems, Journal Year: 2025, Volume and Issue: 42(3)

Published: Feb. 13, 2025

ABSTRACT This study introduces a new multi‐criteria decision‐making (MCDM) framework to evaluate trauma injury detection models in intensive care units (ICUs). research addresses the challenges associated with diverse machine learning (ML) models, inconsistencies, conflicting priorities, and importance of metrics. The developed methodology consists three phases: dataset identification pre‐processing, hybrid model development, an evaluation/benchmarking framework. Through meticulous is tailored focus on adult patients. Forty were by combining eight ML algorithms four filter‐based feature‐selection methods principal component analysis (PCA) as dimensionality reduction method, these evaluated using seven weight coefficients for metrics are determined 2‐tuple Linguistic Fermatean Fuzzy‐Weighted Zero‐Inconsistency (2TLF‐FWZIC) method. Vlsekriterijumska Optimizcija I Kompromisno Resenje (VIKOR) approach applied rank models. According 2TLF‐FWZIC, classification accuracy (CA) precision obtained highest weights 0.2439 0.1805, respectively, while F1, training time, test time lowest 0.1055, 0.0886, 0.1111, respectively. benchmarking results revealed following top‐performing models: Gini index logistic regression (GI‐LR), decision tree (GI_DT), information gain (IG_DT), VIKOR Q score values 0.016435, 0.023804, 0.042077, proposed MCDM assessed examined systematic ranking, sensitivity analysis, validation best‐selected two unseen datasets, mode explainability SHapley Additive exPlanations (SHAP) We benchmarked against other benchmark studies achieved 100% across six key areas. provides several insights into empirical synthesis this study. It contributes advancing medical informatics enhancing understanding selection ICUs.

Language: Английский

---

An Enhanced Detection System of Autism Spectrum Disorder Using Thermal Imaging and Deep Learning

SN Computer Science, Journal Year: 2025, Volume and Issue: 6(2)

Published: Feb. 14, 2025

Language: Английский

---

AI-assisted early screening, diagnosis, and intervention for autism in young children

Frontiers in Psychiatry, Journal Year: 2025, Volume and Issue: 16

Published: April 14, 2025

Autism is a serious threat to an individual’s physical and mental health. Early screening, diagnosis, intervention can effectively reduce the level of deficits in individuals with autism. However, traditional methods rely on professionalism psychiatrists require great deal time effort, resulting large proportion autism being diagnosed after age 6. Artificial intelligence (AI) combined machine learning used improve efficiency early young children. This review aims summarize AI-assisted for children (infants, toddlers, preschoolers). To achieve screening diagnosis children, AI have built predictive models automation behavioral analyzed brain imaging genetic data break barrier established intelligent systems mass screening. For education optimize teaching environment provide individualized interventions, constructed monitoring dynamic tracking, created support continuous meet diverse needs As continues develop, further research needed build shared database autism, generalize migrate effects appearance performance AI-powered robots, failure rates costs technologies.

Language: Английский

---

Prediction Level Fusion of Vision Transformers (PF-ViTs) based Network for the Detection of Autism Spectrum Disorder using sMRI

Intelligent Data Analysis, Journal Year: 2025, Volume and Issue: unknown

Published: April 16, 2025

There has been an unanticipated increase in the number of cases Autism Spectrum Disorder (ASD) present era. Its late detection due to negligence its early symptoms aggravates complications day-to-day life autistic person. Artificial Intelligence (AI)-based classification framework can assist doctors detection, and it help people ameliorate their lifestyle. The less works using Structural Magnetic Resonance Imaging (sMRI) compared Functional (fMRI) with AI-based approaches gives motivation develop system for ASD sMRI scans. In past few years, huge numbers involvement CNN-based computer-vision application have witnessed by research community. Vision Transformer (ViT) network based on idea Transformers Natural Language Processing done revelation performances image recognition. proposed work focuses development a utilizing ViT detection. two different variants i.e., ViT-B16 ViT-B32 utilized additional modification experimentation. Prediction Level Fusion (PF-ViTs) exhibited impressive sMRI-based state-of-the-art (SOTAW) achieving accuracy 94.24%, precision 96.03%, sensitivity 92.36%, specificity 96.14%, F1 score 94.16%, AUC 98.45% towards ASD.

Language: Английский

---