FNIRS Classification of Adults with ADHD Enhanced by Feature Selection.

In this study, the retinal nerve fiber layer (RNFL) thicknesses were compared between adult attention deficit hyperactivity disorder (ADHD) cases and healthy controls. The study included adults diagnosed with ADHD based on the DSM-5 criteria and age and gender matched healthy controls. Spectral area optical coherence tomography (OCT) was performed on the 52 eyes of 26 participants with ADHD and the 52 eyes of the 26 healthy control individuals. Comparing the data on the 52 eyes of 26 ADHD participants and the 52 eyes of 26 healthy control participants indicated that the central macular thickness (CMT) and the RNFL thicknesses, the ganglion cell complex (GCC), the mean inner macular ring (MIRAVG) and the mean outer macular ring (MOR-AVG) thicknesses were significantly lower in the ADHD group. This is the first study in the literature on the RNFL thickness in adult ADHD patients. Our findings demonstrated that RNFL thickness is lower in ADHD cases as the unmyelinated axons are reduced in ADHD. Hence, the quantitative and reproducible nature of Spectral Domain-OCT thickness measurements can be used as biomarkers to monitor disease progression in ADHD cases.

The ability of music to spread joy and excitement across lives, makes it widely acknowledged as the human race's universal language. The phrase "music genre" is frequently used to group several musical styles together as following a shared custom or set of guidelines. According to their unique preferences, people now make playlists based on particular musical genres. Due to the determination and extraction of appropriate audio elements, music genre identification is regarded as a challenging task. Music information retrieval, which extracts meaningful information from music, is one of several real - world applications of machine learning. The objective of this paper is to efficiently categorise songs into various genres based on their attributes using various machine learning approaches. To enhance the outcomes, appropriate feature engineering and data pre-processing techniques have been performed. Finally, using suitable performance assessment measures, the output from each model has been compared. Compared to other machine learning algorithms, Random Forest along with efficient feature selection and hyperparameter tuning has produced better results in classifying music genres.

Electrophysiological indices of error monitoring in juvenile and adult attention deficit hyperactivity disorder (ADHD)—A meta-analytic appraisal

Both feature selection and hyperparameter tuning are key tasks in machine learning. Hyperparameter tuning is often useful to increase model performance, while feature selection is undertaken to attain sparse models. Sparsity may yield better model interpretability and lower cost of data acquisition, data handling and model inference. While sparsity may have a beneficial or detrimental effect on predictive performance, a small drop in performance may be acceptable in return for a substantial gain in sparseness. We therefore treat feature selection as a multi-objective optimization task. We perform hyperparameter tuning and feature selection simultaneously because the choice of features of a model may influence what hyperparameters perform well. We present, benchmark, and compare two different approaches for multi-objective joint hyperparameter optimization and feature selection: The first uses multi-objective model-based optimization. The second is an evolutionary NSGA-II-based wrapper approach to feature selection which incorporates specialized sampling, mutation and recombination operators. Both methods make use of parameterized filter ensembles. While model-based optimization needs fewer objective evaluations to achieve good performance, it incurs computational overhead compared to the NSGA-II, so the preferred choice depends on the cost of evaluating a model on given data.

Explainable machine learning-based fractional vegetation cover inversion and performance optimization – A case study of an alpine grassland on the Qinghai-Tibet Plateau

Identification of high-risk genes and classification of acute myocardial infarction patients utilizing deep learning in a restricted cohort.

A machine-learning-based model for predicting the effective stiffness of precast concrete columns

Using network analysis and random forest regression, this study identified attention-deficit/hyperactivity disorder (ADHD) symptoms most important for indicating impairment in various functional domains. Participants comprised a nationally representative sample of 1249 adults in the United States. Bridge symptoms were identified as those demonstrating unique relations with impairment domains that, in total, were stronger than those involving other symptoms. Results suggested three inattentive (i.e., difficulty organizing; does not follow through; makes careless mistakes) and one hyperactive (difficulty engaging in leisure activities) bridge symptoms. Random forest regression results supported bridge symptoms as most important (compared to other symptoms) for predicting global and specific impairment domains. Hyperactive/impulsive symptoms appeared more strongly related to impairment in women, whereas difficulty organizing and easily distracted appeared more related to impairment in men. Clarification of bridge symptoms may help identify core characteristics of ADHD in adulthood and specify screening and intervention targets to reduce risk for related impairment.

Advances in data collection have resulted in an exponential growth of high-dimensional microarray datasets for binary classification in bioinformatics and medical diagnostics. These datasets generally possess many features but relatively few samples, resulting in challenges associated with the “curse of dimensionality”, such as feature redundancy and an elevated risk of overfitting. While traditional feature selection approaches, such as filter-based and wrapper-based approaches, can help to reduce dimensionality, they often struggle to capture feature interactions while adequately preserving model generalization. Therefore, this paper introduces the Adaptive Cluster-Guided Simple, Fast, and Efficient (ACG-SFE) feature selection, a hybrid approach designed to address the challenges of high-dimensional microarray data in binary classification. ACG-SFE enhances the Simple, Fast, and Efficient (SFE) evolutionary feature selection model by integrating hierarchical clustering to dynamically group correlated features based on the optimal number of clusters determined by the Silhouette index, Davies-Bouldin score, and the feature-to-observation ratio while adaptively selecting representative features within clusters using mutual information and adjusting the selection threshold through a progress factor. This hybrid filter-wrapper approach improves feature interactions, effectively minimizing redundancy and overfitting while enhancing classification performance. The proposed model is assessed against four state-of-the-art evolutionary feature selection models on 11 high-dimensional microarray datasets. Experimental results indicate that ACG-SFE effectively selects a small yet pertinent feature subset, minimizing redundancy while attaining enhanced classification accuracy and F-measure. Furthermore, its reduced RMSE between train and test accuracy substantiates its capability to reduce overfitting, outperforming comparative models. These findings establish ACG-SFE as an effective feature selection model for handling high-dimensional microarray data in binary classification, enhancing classification accuracy while selecting minimal relevant features to reduce unnecessary complexity and the risk of overfitting.

Background The number of individuals who experience the onset of attention-deficit/hyperactivity disorder (ADHD)-like symptoms after the age of 12 years has been growing, which does not meet the diagnostic criteria for adult ADHD. We designated this condition as "pseudo-ADHD" and investigated its psychological implications by comparing the parenting attitudes in childhood and attachment styles among individuals with "pseudo-ADHD," those with adult ADHD, and healthy controls. Methods We conducted an online cross-sectional survey of adults aged 18 years. Participants in the analysis included individuals with "pseudo-ADHD" (n = 46), adults with ADHD (n = 23), and healthy controls (n = 758). The Parental Bonding Instrument (PBI) was used to evaluate parenting attitudes, and the Experiences in Close Relationships Inventory for the Generalized Other (ECR-GO) was used to evaluate attachment styles. One-way analysis of variance was performed to compare the PBI and ECR-GO scores among the three groups. Results Regarding the PBI scores, a significant main effect of group was found for the paternal care scores (p = 0.002), and the "pseudo-ADHD" group did not differ significantly from the adult ADHD group or healthy control group (p = 0.378, p = 0.228, respectively). A significant main effect of the group was found for the maternal care scores (p < 0.001). The "pseudo-ADHD" group scored significantly lower than the adult ADHD group (p = 0.005), whereas there was no significant difference compared to the healthy control group (p = 1.000). A significant main effect of the group was also found for the paternal overprotection scores (p < 0.001). The "pseudo-ADHD" group scored significantly lower than the adult ADHD group (p = 0.002), whereas there was no significant difference compared to the healthy control group (p = 0.571). A significant main effect of group was found for the maternal overprotection scores (p < 0.001), and the "pseudo-ADHD" group did not differ significantly from either the adult ADHD group or the healthy control group (p = 0.060, p = 0.161, respectively). Regarding ECR-GO scores, a significant main effect of group was found for the fear of abandonment score (p < 0.001), and the "pseudo-ADHD" group scored significantly lower than the adult ADHD group (p < 0.001) and significantly higher than the healthy control group (p = 0.005). Conclusion This study suggested that the parenting attitudes of parents of individuals with "pseudo-ADHD" were not different from those of the healthy controls. For attachment styles, the fear of abandonment was higher in individuals with "pseudo-ADHD" than in healthy controls, but not as high as in patients with adult ADHD. This study highlights, for the first time, the importance of considering attachment styles in the diagnosis and treatment of individuals with ADHD symptoms in adulthood.

The design of this system shows the efficient prediction of heart diseases by the implementation of machine learning multiple classifiers algorithm. The UCI Heart Disease Dataset is used, while the data preprocessing involves filling missing values, encoding of categorical values, scaling of features, and selection of important features. Three machine learning classifiers, namely Random Forest Classifier, Gradient Boosting Classifier, and Logistic Regression are implemented. The implementation involves training the machine learning models using the RandomizedSearchCV technique and cross-validation technique. The evaluation of the model's performance is done using measures such as ACCURACY, PRECISION, RECALL, and F1-SCORE. A fair comparison is conducted to select the best performing model automatically using accuracy. Results show that the Random Forest model performed the best with an accuracy of 81.67%, compared to the rest of the models. Evaluation of feature importance shows key clinical features that significantly contribute to the accuracy. Keywords— Heart Disease Prediction, Machine Learning, Random Forest, Gradient Boosting, Logistic Regression, Feature Selection, Classification, Healthcare Analytics, UCI Dataset, Predictive Modeling

P 150 The correlation of attention and neurophysiological characteristics in attention deficit hyperactivity disorder (ADHD)

Background: Current research indicates clear distinctions between the general population and individuals diagnosed with Attention-Deficit/Hyperactivity Disorder (ADHD), specifically in cognitive processes and physiological brain activity. Stimulant medications have been shown to improve and, in some cases, normalize dysfunction in both these areas. The majority of extant brain imaging literature has included functional magnetic resonance imaging (fMRI) in tandem with neuropsychological tests to explore these relationships. However, functional near-infrared spectroscopy (fNIRS) is a promising neuroimaging technology that offers some unique features including portability, ecological validity, and the ability to detect brain activity via concentrations of oxygenated and deoxygenated blood. Very few studies have used fNIRS as a tool to investigate the impact of medication on adults with ADHD, both in their cognitive functioning and brain hemodynamics. However, there is evidence that this technology may-in addition to providing novel information about cognitive and physiological functioning-actually be better suited to evaluating the ADHD population. Objectives: The present study sought to: 1) compare differences in cognition (using neuropsychological tests) and neurophysiology (using fNIRS) between adults withADHD (when unmedicated) and age-, gender-, and education-matched healthy control(HC) adults, and 2) compare differences in cognition and neurophysiology betweenmedicated and unmedicated states in adults with ADHD. Ultimately, the goal was to investigate the potential application of fNIRS as an assessment tool for both diagnostic(of the physiological underpinnings of the disorder) and monitoring (of the effectiveness of cognition-improving medications) purposes. Participants: Nine individuals aged 18-25 diagnosed with ADHD and nine age-, gender-, and education-matched healthy control participants (HCs). Method: All participants completed two testing sessions. HCs completed both sessions under normal conditions whereas ADHD participants completed one testing session while medicated and the other while unmedicated. Cognitive testing focused on working memory, a domain relevant to ADHD. Testing included 1) a two-subtest WAIS-IV; 2) a Sternberg delayed recognition task (a test of working memory); and 3) a visual n-back task (another working memory test). The latter two tasks were completed while participants were connected to a continuous wave fNIRS system to record cerebral hemodynamic activity during working memory performance. Demographic variables, confirmation of diagnosistic status, medications, and recent daily activities regarding caffeine, alcohol, tobacco, and other behaviors that could impact fNIRS data were recorded. Results: Cognitively, while healthy controls performed better on working memory tasks than ADHD participants, and ADHD participants generally performed better when medicated than unmedicated, neither comparison reached significance. Physiologically, there were no significant differences in PFC activation between the HC and ADHD groups, nor between medicated and unmedicated states within the ADHD group. Conclusions: The findings of this study mirror previous findings of the cognitive effects of medication in ADHD. Not surprisingly, individuals with ADHD performed working memory tasks less accurately and more slowly than controls and the intake of stimulant medication led to an improvement in performance that reached levels similar to healthy individuals. By contrast, the medication-induced working memory improvements among adults with ADHD were not reflected in corresponding physiological changes measured by fNIRS. The lack of significant findings in this study may be due to low statistical power, though alternative explanations are also explored. Nonetheless, before fNIRS can contribute to clinical diagnosis and treatment of ADHD, more research is needed to establish its clinical application.

P-633 - Self-reported chronic insomnia and psychiatric symptoms: preliminary results from an online survey in romanian adults

Increasing evidence shows that unintentional mind wandering is linked to Attention Deficit Hyperactivity Disorder (ADHD) and that its frequency contributes to symptom severity and functional impairment in ADHD. However, empirical data on mind wandering in adult ADHD are still scarce, and a validated scale to assess mind wandering in German adult ADHD patients is lacking. The primary aim of this study is to assess the psychometric properties of the German version of the recently published Mind Excessively Wandering Scale (MEWS-G) in terms of factorial structure and factor stability, internal consistency and construct validity. Analyses were performed in 128 adults with ADHD, clinical and healthy controls. As described for the original English 15-item version of the scale, we found lowest item-total-correlations for items 6, 10 and 14 with item-total correlation of all: 0.54/ADHD: 0.32 (item 6), all: 0.55/ADHD: 0.39 (item 10) and all: 0.11/ADHD: -0.04 (item 14). Item-total correlations for the remaining items were 0.65-0.86 and Cronbach Alpha was 0.96 indicating good internal consistency of the 12-item version of scale, on which we based all further analyses. Principal component analysis indicated a one- and two- factorial scale structure respectively explaining 71.7 % and 78.7 % of variance. Both factors showed good stability with lower stability of the factor-2 solution if sample size was reduced. The two-factorial solution also had many cross-loadings and a strong correlation of both factors in confirmatory factorial analysis (rf1f2=0.87). It probably describes related and interdependent, but not distinct facets of mind wandering, which strongly argues for the one factorial structure of the scale. Mean MEWS-G score in ADHD was 23.77±7.85 compared to 7.64±7.27 in controls (p<.0001). According to ROC, the optimal cut-off point to discriminate ADHD and controls is at MEWS-G score=13. On the symptom level, MEWS-G score was correlated with ADHD, depressive and total psychiatric symptom scores, on the personality level with neuroticsm and negatively with conscientiousness and on the functional level with social interaction difficulties and impaired self-efficacy. In summary, our study shows that MEWS-G is a reliable, valid instrument to assess spontaneous mind wandering in adult ADHD and to discriminate between ADHD and controls.