Predicting Methylphenidate Response in ADHD Using Machine Learning Approaches

Virtual reality-assisted prediction of adult ADHD based on eye tracking, EEG, actigraphy and behavioral indices: a machine learning analysis of independent training and test samples

Given the heterogeneous nature of attention-deficit/hyperactivity disorder (ADHD) and the absence of established biomarkers, accurate diagnosis and effective treatment remain a challenge in clinical practice. This study investigates the predictive utility of multimodal data, including eye tracking, EEG, actigraphy, and behavioral indices, in differentiating adults with ADHD from healthy individuals. Using a support vector machine model, we analyzed independent training (n = 50) and test (n = 36) samples from two clinically controlled studies. In both studies, participants performed an attention task (continuous performance task) in a virtual reality seminar room while encountering virtual distractions. Task performance, head movements, gaze behavior, EEG, and current self-reported inattention, hyperactivity, and impulsivity were simultaneously recorded and used for model training. Our final model based on the optimal number of features (maximal relevance minimal redundancy criterion) achieved a promising classification accuracy of 81% in the independent test set. Notably, the extracted EEG-based features had no significant contribution to this prediction and therefore were not included in the final model. Our results suggest the potential of applying ecologically valid virtual reality environments and integrating different data modalities for enhancing robustness of ADHD diagnosis.

Individualized prediction models in ADHD: a systematic review and meta-regression

There have been increasing efforts to develop prediction models supporting personalised detection, prediction, or treatment of ADHD. We overviewed the current status of prediction science in ADHD by: (1) systematically reviewing and appraising available prediction models; (2) quantitatively assessing factors impacting the performance of published models. We did a PRISMA/CHARMS/TRIPOD-compliant systematic review (PROSPERO: CRD42023387502), searching, until 20/12/2023, studies reporting internally and/or externally validated diagnostic/prognostic/treatment-response prediction models in ADHD. Using meta-regressions, we explored the impact of factors affecting the area under the curve (AUC) of the models. We assessed the study risk of bias with the Prediction Model Risk of Bias Assessment Tool (PROBAST). From 7764 identified records, 100 prediction models were included (88% diagnostic, 5% prognostic, and 7% treatment-response). Of these, 96% and 7% were internally and externally validated, respectively. None was implemented in clinical practice. Only 8% of the models were deemed at low risk of bias; 67% were considered at high risk of bias. Clinical, neuroimaging, and cognitive predictors were used in 35%, 31%, and 27% of the studies, respectively. The performance of ADHD prediction models was increased in those models including, compared to those models not including, clinical predictors (β = 6.54, p = 0.007). Type of validation, age range, type of model, number of predictors, study quality, and other type of predictors did not alter the AUC. Several prediction models have been developed to support the diagnosis of ADHD. However, efforts to predict outcomes or treatment response have been limited, and none of the available models is ready for implementation into clinical practice. The use of clinical predictors, which may be combined with other type of predictors, seems to improve the performance of the models. A new generation of research should address these gaps by conducting high quality, replicable, and externally validated models, followed by implementation research.

Prediction of methylphenidate treatment response for ADHD using conventional and radiomics T1 and DTI features: Secondary analysis of a randomized clinical trial

BACKGROUND

Attention-Deficit/Hyperactivity Disorder (ADHD) is commonly treated with methylphenidate (MPH). Although highly effective, MPH treatment still has a relatively high non-response rate of around 30%, highlighting the need for a better understanding of treatment response. Radiomics of T1-weighted images and Diffusion Tensor Imaging (DTI) combined with machine learning approaches could offer a novel method for assessing MPH treatment response.

PURPOSE

To evaluate the accuracy of both conventional and radiomics approaches in predicting treatment response based on baseline T1 and DTI data in stimulant-naive ADHD participants.

METHODS

We performed a secondary analysis of a randomized clinical trial (ePOD-MPH), involving 47 stimulant-naive ADHD participants (23 boys aged 11.4 ± 0.4 years, 24 men aged 28.1 ± 4.3 years) who underwent 16 weeks of treatment with MPH. Baseline T1-weighted and DTI MRI scans were acquired. Treatment response was assessed at 8 weeks (during treatment) and one week after cessation of 16-week treatment (post-treatment) using the Clinical Global Impressions - Improvement scale as our primary outcome. We compared prediction accuracy using a conventional model and a radiomics model. The conventional approach included the volume of bilateral caudate, putamen, pallidum, accumbens, and hippocampus, and for DTI the mean fractional anisotropy (FA) of the entire brain white matter, bilateral Anterior Thalamic Radiation (ATR), and the splenium of the corpus callosum, totaling 14 regional features. For the radiomics approach, 380 features (shape-based and first-order statistics) were extracted from these 14 regions. XGBoost models with nested cross-validation were used and constructed for the total cohort (n = 47), as well as children (n = 23) and adults (n = 24) separately. Exact binomial tests were employed to compare model performance.

RESULTS

For the conventional model, balanced accuracy (bAcc) in predicting treatment response during treatment was 63 % for the total cohort, 32 % for children, and 36 % for adults (Area Under the Receiver Operating Characteristic Curve (AUC-ROC): 0.69, 0.33, 0.41 respectively). Radiomics models demonstrated bAcc's of 68 %, 64 %, and 64 % during treatment (AUC-ROCs of 0.73, 0.62, 0.69 respectively). These predictions were better than chance for both conventional and radiomics models in the total cohort (p = 0.04, p = 0.003 respectively). The radiomics models outperformed the conventional models during treatment in children only (p = 0.02). At post-treatment, performance was markedly reduced.

CONCLUSION

While conventional and radiomics models performed equally well in predicting clinical improvement across children and adults during treatment, radiomics features offered enhanced structural information beyond conventional region-based volume and FA averages in children. Prediction of symptom improvement one week after treatment cessation was poor, potentially due to the transient effects of stimulant treatment on symptom improvement.

Predictors of Methylphenidate response in children and adolescents with ADHD: the role of sleep disturbances

Sleep Disturbances (SD) have been linked to children and adolescents with ADHD, impacting its progression and outcomes. Methylphenidate (MPH), a commonly used stimulant medication for ADHD treatment, has been observed to potentially influence SD as a side effect, while SD can in turn potentially affect the response to MPH. This study aimed to explore the potential role of SD on MPH response in children and adolescents with ADHD. At this aim, 43 children and adolescents with ADHD received a single dose of MPH and were assessed for attention before and after medication administration. As expected, the administration of MPH resulted in improved attention levels. Our data indicate that patients with higher SD experienced greater benefits from the medication, stabilizing Reaction Times Variability (VRTs). This suggests that SD might influence the response to MPH, with individuals exhibiting higher SD deriving more advantages from the treatment. In addition, we found that other factors, such as externalizing problems and IQ, interact with each other and with SD, influencing the response to stimulant medication. Early detection of SD, along with the study of cognitive and emotional-behavioral characteristics, could assist clinicians in predicting the effectiveness of MPH therapy in children and adolescents with ADHD. However, further research is necessary to gain a deeper understanding of the role of SD and other factors in the long-term effects of MPH.

Neural and Cognitive Predictors of Stimulant Treatment Efficacy in Medication-Naïve ADHD Adults: A Pilot Diffusion Tensor Imaging Study

OBJECTIVE

Stimulant medications are the main treatment for Attention Deficit Hyperactivity Disorder (ADHD), but overall treatment efficacy in adults has less than a 60% response rate. This study aimed to identify neural and cognitive markers predictive of longitudinal improvement in response to stimulant treatment in drug-naïve adults with ADHD.

METHOD

We used diffusion tensor imaging (DTI) and executive function measures with 36 drug-naïve adult ADHD patients in a prospective study design.

RESULTS

Structural connectivity (measured by fractional anisotropy, FA) in striatal regions correlated with ADHD clinical symptom improvement following stimulant treatment (amphetamine or methylphenidate) in better medication responders. A significant positive correlation was also found between working memory performance and stimulant-related symptom improvement. Higher pre-treatment working memory scores correlated with greater response.

CONCLUSION

These findings provide evidence of pre-treatment neural and behavioral markers predictive of longitudinal treatment response to stimulant medications in adults with ADHD.

Classification of attention deficit hyperactivity disorder using machine learning on an EEG dataset

The neurodevelopmental disorder, Attention Deficit Hyperactivity Disorder (ADHD), frequently affecting youngsters, is characterized by persistent patterns of inattention, hyperactivity, and impulsivity, the etiology of which may involve a variety of genetic, environmental, and neurological factors. Electroencephalography (EEG) measures the electrical activity in the brain through neuronal activity, which is a function of cognitive processes. In this study, a previously recorded sample set of 121 children containing unbiased data from both ADHD and control group classes and EEG signals were analyzed to classify the ADHD patients. The samples were tested under different cognitive conditions, and multiple features were extracted using Euclidean distance. Many machine learning algorithms use Euclidean distance as their default distance metric to compare two recorded data points. The extracted features were trained using four supervised machine learning algorithms (linear regression, random forest, extreme gradient boosting, and K nearest neighbor (KNN)) based on the results of various frequency bands. The results suggest that the KNN algorithm produces the highest accuracy over other machine learning approaches, and results can be further improved with the application of hyperparameter tuning and used for classifying sub-groups of ADHD to identify the severity of the disorder.

Development of a serious game-based cognitive rehabilitation system for patients with brain injury

BACKGROUND

Traumatic brain injury (TBI) resulting from a forceful impact to the head can cause severe functional disabilities, with cognitive impairment being a major hindrance to patients' return to daily life. Encouraging patients to engage in rehabilitation programs consistently poses a significant challenge for therapists. To address this issue, gamification has gained momentum as an effective approach. This study aims to develop a serious game-based cognitive rehabilitation system tailored for patients with brain injury.

METHODS

The study included four stages. Initially, the requirements were analyzed through focus groups. Then the system structure and game content were discussed and was agreed as a conceptual model. In second stage, the system design was drawn using various modeling diagrams. In third stage, a system prototype was developed using the Unity game engine and C# programming. Finally, a heuristic evaluation method was employed to assess usability.

RESULTS

Based on the focus group meetings with seven participants, a conceptual model of the system structure and game content was designed. Game's interface was developed for both the therapist and patient versions. The focus groups determined a 2D casual gaming genre with a postman character and 10 missions on the smartphone platform. For example, in the first mission, the postman must move from mailboxes 1 to 10 and pick up the letters. This is according to Trail Making Test task. The 16 tasks in different subcategories of attention were selected to make these missions. The usability evaluation highlighted privacy, help and documentation, and aesthetic and minimalist design as the areas with the highest percentage of problems.

CONCLUSIONS

Cognitive rehabilitation is vital in facilitating patients' faster return to daily routines and enhancing their quality-of-life following brain injury. Incorporating a game-based system provides patients with increased motivation to engage in various cognitive exercises. Additionally, continuous monitoring by specialists ensures effective patient management. The game-based system offers different game stages to strengthen and rehabilitate attention in patients with brain injury. In the next step, the clinical effects of this system will be evaluated.

Supervised machine learning: A new method to predict the outcomes following exercise intervention in children with autism spectrum disorder

The individual differences among children with autism spectrum disorder (ASD) may make it challenging to achieve comparable benefits from a specific exercise intervention program. A new method for predicting the possible outcomes and maximizing the benefits of exercise intervention for children with ASD needs further exploration. Using the mini-basketball training program (MBTP) studies to improve the symptom performance of children with ASD as an example, we used the supervised machine learning method to predict the possible intervention outcomes based on the individual differences of children with ASD, investigated and validated the efficacy of this method. In a long-term study, we included 41 ASD children who received the MBTP. Before the intervention, we collected their clinical information, behavioral factors, and brain structural indicators as candidate factors. To perform the regression and classification tasks, the random forest algorithm from the supervised machine learning method was selected, and the cross validation method was used to determine the reliability of the prediction results. The regression task was used to predict the social communication impairment outcome following the MBTP in children with ASD, and explainable variance was used to evaluate the predictive performance. The classification task was used to distinguish the core symptom outcome groups of ASD children, and predictive performance was assessed based on accuracy. We discovered that random forest models could predict the outcome of social communication impairment (average explained variance was 30.58%) and core symptom (average accuracy was 66.12%) following the MBTP, confirming that the supervised machine learning method can predict exercise intervention outcomes for children with ASD. Our findings provide a novel and reliable method for identifying ASD children most likely to benefit from a specific exercise intervention program in advance and a solid foundation for establishing a personalized exercise intervention program recommendation system for ASD children.

Predicting the Response of Children and Adolescents With ADHD to Methylphenidate: A Systematic Review

OBJECTIVE

To report the characteristics associated with response to methylphenidate (MPH) in children and adolescents with ADHD.

METHODS

Studies reporting potentials predictors of response to MPH were searched in Medline and Embase from January 1998 to March 2022. Narrative synthesis was performed.

RESULTS

Fifty-seven reports of 46 studies totaling 6,656 ADHD patients were included. No association appears between response to MPH and age, gender, MPH dosage, ADHD subtype, comorbidities nor socioeconomic status when considering a specific patient. No conclusion could be drawn about body weight, ADHD severity, intelligence quotient, and parental symptoms of depression or ADHD.

CONCLUSIONS

None of these potential predictors have proven their usefulness to predict response to MPH on an individual basis in clinical practice. In research, potential predictors should be measured, their association with response to MPH assessed, in order to control for confounding variables when modeling response to MPH.

Poor response to methylphenidate is associated with a smaller dorsal attentive network in adult Attention-Deficit/Hyperactivity Disorder (ADHD)

Stimulants, such as methylphenidate (MPH), are effective in treating attention-deficit/hyperactivity disorder (ADHD), but there is individual variability in response, especially in adults. To improve outcomes, we need to understand the factors associated with adult treatment response. This longitudinal study investigated whether pre-treatment anatomy of the fronto-striatal and fronto-parietal attentional networks was associated with MPH treatment response. 60 adults with ADHD underwent diffusion brain imaging before starting MPH treatment, and response was measured at two months. We tested the association between brain anatomy and treatment response by using regression-based approaches; and compared the identified anatomical characteristics with those of 20 matched neurotypical controls in secondary analyses. Finally, we explored whether combining anatomical with clinical and neuropsychological data through machine learning provided a more comprehensive profile of factors associated with treatment response. At a group level, a smaller left dorsal superior longitudinal fasciculus (SLF I), a tract responsible for the voluntary control of attention, was associated with a significantly lower probability of being responders to two-month MPH-treatment. The association between the volume of the left SLF I and treatment response was driven by improvement on both inattentive and hyperactive/impulsive symptoms. Only non-responders significantly differed from controls in this tract metric. Finally, our machine learning approach identified clinico-neuropsychological factors associated with treatment response, such as higher cognitive performance and symptom severity at baseline. These novel findings add to our understanding of the pathophysiological mechanisms underlying response to MPH, pointing to the dorsal attentive network as playing a key role.

Machine learning in attention-deficit/hyperactivity disorder: new approaches toward understanding the neural mechanisms

Attention-deficit/hyperactivity disorder (ADHD) is a highly prevalent and heterogeneous neurodevelopmental disorder in children and has a high chance of persisting in adulthood. The development of individualized, efficient, and reliable treatment strategies is limited by the lack of understanding of the underlying neural mechanisms. Diverging and inconsistent findings from existing studies suggest that ADHD may be simultaneously associated with multivariate factors across cognitive, genetic, and biological domains. Machine learning algorithms are more capable of detecting complex interactions between multiple variables than conventional statistical methods. Here we present a narrative review of the existing machine learning studies that have contributed to understanding mechanisms underlying ADHD with a focus on behavioral and neurocognitive problems, neurobiological measures including genetic data, structural magnetic resonance imaging (MRI), task-based and resting-state functional MRI (fMRI), electroencephalogram, and functional near-infrared spectroscopy, and prevention and treatment strategies. Implications of machine learning models in ADHD research are discussed. Although increasing evidence suggests that machine learning has potential in studying ADHD, extra precautions are still required when designing machine learning strategies considering the limitations of interpretability and generalization.

Prediction of Diagnosis and Treatment Response in Adolescents With Depression by Using a Smartphone App and Deep Learning Approaches: Usability Study

BACKGROUND

Lack of quantifiable biomarkers is a major obstacle in diagnosing and treating depression. In adolescents, increasing suicidality during antidepressant treatment further complicates the problem.

OBJECTIVE

We sought to evaluate digital biomarkers for the diagnosis and treatment response of depression in adolescents through a newly developed smartphone app.

METHODS

We developed the Smart Healthcare System for Teens At Risk for Depression and Suicide app for Android-based smartphones. This app passively collected data reflecting the social and behavioral activities of adolescents, such as their smartphone usage time, physical movement distance, and the number of phone calls and text messages during the study period. Our study consisted of 24 adolescents (mean age 15.4 [SD 1.4] years, 17 girls) with major depressive disorder (MDD) diagnosed with Kiddie Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version and 10 healthy controls (mean age 13.8 [SD 0.6] years, 5 girls). After 1 week's baseline data collection, adolescents with MDD were treated with escitalopram in an 8-week, open-label trial. Participants were monitored for 5 weeks, including the baseline data collection period. Their psychiatric status was measured every week. Depression severity was measured using the Children's Depression Rating Scale-Revised and Clinical Global Impressions-Severity. The Columbia Suicide Severity Rating Scale was administered in order to assess suicide severity. We applied the deep learning approach for the analysis of the data. Deep neural network was employed for diagnosis classification, and neural network with weighted fuzzy membership functions was used for feature selection.

RESULTS

We could predict the diagnosis of depression with training accuracy of 96.3% and 3-fold validation accuracy of 77%. Of the 24 adolescents with MDD, 10 responded to antidepressant treatments. We predicted the treatment response of adolescents with MDD with training accuracy of 94.2% and 3-fold validation accuracy of 76%. Adolescents with MDD tended to move longer distances and use smartphones for longer periods of time compared to controls. The deep learning analysis showed that smartphone usage time was the most important feature in distinguishing adolescents with MDD from controls. Prominent differences were not observed in the pattern of each feature between the treatment responders and nonresponders. The deep learning analysis revealed that the total length of calls received as the most important feature predicting antidepressant response in adolescents with MDD.

CONCLUSIONS

Our smartphone app demonstrated preliminary evidence of predicting diagnosis and treatment response in depressed adolescents. This is the first study to predict the treatment response of adolescents with MDD by examining smartphone-based objective data with deep learning approaches.

Predicting efficacy of viloxazine extended-release treatment in adults with ADHD using an early change in ADHD symptoms: Machine learning Post Hoc analysis of a phase 3 clinical trial

Early response to viloxazine extended-release (viloxazine ER, Qelbree®) treatment predicted efficacy outcome in pediatric subjects with attention-deficit/hyperactivity disorder (ADHD). This study sought to determine whether the machine learning lasso model used in the pediatric study would predict response to viloxazine ER in an adult population based on early improvements in ADHD symptoms. We used data from a double-blind, placebo-controlled, flexible-dose (200-600 mg) study of viloxazine ER (N = 354; 18 to 60 years old). Area under the Receiver Operating Characteristic Curve (ROC AUC) statistics were computed using the lasso model from pediatric data to predict responder status in adults. Response was defined as ≥50% reduction from baseline in the Adult ADHD Investigator Symptoms Rating Scale (AISRS) Total score at Week 6. The adult study sample included 127 viloxazine ER-treated subjects with Week 6 data. Fifty-one subjects (40.2%) were categorized as responders. The ROC curves indicated that data collected up to Week 2 were sufficient to accurately predict treatment response at Week 6 with 68% positive predictive power, 80% sensitivity, and 74% specificity. This analysis demonstrated that the predictive model estimated from the child data generalizes to adults with ADHD, further supporting the consistency of viloxazine ER treatment across age groups.

Biological Predictors of Treatment Response in Adult Attention Deficit Hyperactivity Disorder (ADHD): A Systematic Review

Background: Attention-deficit/hyperactivity disorder (ADHD) is a highly prevalent condition with onset in childhood and in many cases persisting into adulthood. Even though an increasing number of studies have investigated the efficacy of pharmacotherapy in the management of adult ADHD, few authors have tried to identify the biological predictors of treatment response. Objectives: To summarize the available data about the biological markers of treatment response in adults affected by ADHD. Methods: A search on the main biomedical and psychological archives (PubMed, Embase, Scopus, and PsycINFO) was performed. Manuscripts in English, published up to May 2022 and having the biological predictors of treatment response in adults with ADHD as their main topic, were included. Results: A total of 3855 articles was screened. Twenty-two articles were finally included. Most of the manuscripts studied neuroimaging and electrophysiological factors as potential predictors of treatment response in adult ADHD patients. No reliable markers were identified until now. Promising findings on this topic regard genetic polymorphisms in snap receptor (SNARE) proteins and default mode network-striatum connectivity. Conclusions: Even though some biological markers seem promising for the prediction of treatment response in adults affected by ADHD, further studies are needed to confirm the available data in the context of precision medicine.

Treatment biomarkers for ADHD: Taking stock and moving forward

The development of treatment biomarkers for psychiatric disorders has been challenging, particularly for heterogeneous neurodevelopmental conditions such as attention-deficit/hyperactivity disorder (ADHD). Promising findings are also rarely translated into clinical practice, especially with regard to treatment decisions and development of novel treatments. Despite this slow progress, the available neuroimaging, electrophysiological (EEG) and genetic literature provides a solid foundation for biomarker discovery. This article gives an updated review of promising treatment biomarkers for ADHD which may enhance personalized medicine and novel treatment development. The available literature points to promising pre-treatment profiles predicting efficacy of various pharmacological and non-pharmacological treatments for ADHD. These candidate predictive biomarkers, particularly those based on low-cost and non-invasive EEG assessments, show promise for the future stratification of patients to specific treatments. Studies with repeated biomarker assessments further show that different treatments produce distinct changes in brain profiles, which track treatment-related clinical improvements. These candidate monitoring/response biomarkers may aid future monitoring of treatment effects and point to mechanistic targets for novel treatments, such as neurotherapies. Nevertheless, existing research does not support any immediate clinical applications of treatment biomarkers for ADHD. Key barriers are the paucity of replications and external validations, the use of small and homogeneous samples of predominantly White children, and practical limitations, including the cost and technical requirements of biomarker assessments and their unknown feasibility and acceptability for people with ADHD. We conclude with a discussion of future directions and methodological changes to promote clinical translation and enhance personalized treatment decisions for diverse groups of individuals with ADHD.

Toward Precision Medicine in ADHD

Attention-Deficit Hyperactivity Disorder (ADHD) is a complex and heterogeneous neurodevelopmental condition for which curative treatments are lacking. Whilst pharmacological treatments are generally effective and safe, there is considerable inter-individual variability among patients regarding treatment response, required dose, and tolerability. Many of the non-pharmacological treatments, which are preferred to drug-treatment by some patients, either lack efficacy for core symptoms or are associated with small effect sizes. No evidence-based decision tools are currently available to allocate pharmacological or psychosocial treatments based on the patient's clinical, environmental, cognitive, genetic, or biological characteristics. We systematically reviewed potential biomarkers that may help in diagnosing ADHD and/or stratifying ADHD into more homogeneous subgroups and/or predict clinical course, treatment response, and long-term outcome across the lifespan. Most work involved exploratory studies with cognitive, actigraphic and EEG diagnostic markers to predict ADHD, along with relatively few studies exploring markers to subtype ADHD and predict response to treatment. There is a critical need for multisite prospective carefully designed experimentally controlled or observational studies to identify biomarkers that index inter-individual variability and/or predict treatment response.

Review and Meta-analysis on the Impact of the ADRA2A Variant rs1800544 on Methylphenidate Outcomes in Attention-Deficit/Hyperactivity Disorder

Background

Methylphenidate is among the most prescribed medications for treating attention-deficit/hyperactivity disorder (ADHD). However, nearly half of pediatric patients with ADHD do not respond to methylphenidate treatment. Pharmacogenetic testing can aid in identifying patients for whom methylphenidate is unlikely to be safe or effective, leading to improved methylphenidate outcomes and increased use of alternative treatment options for ADHD. This article aimed to summarize findings from studies of the ADRA2A gene variant, rs1800544, and its association with methylphenidate outcomes in ADHD.

Methods

We systematically reviewed and meta-analyzed available literature on the impact of rs1800544 on methylphenidate outcomes in ADHD.

Results

Fourteen studies met inclusion criteria for review, 9 of which were eligible for meta-analysis. The included studies compared methylphenidate outcomes in patients with ADHD categorized by rs1800544 genotype. G-allele carriers experienced significantly greater improvements in ADHD symptom scores (Swanson, Nolan, and Pelham Version-IV Scale or ADHD Rating Scale-IV) relative to noncarriers (odds ratio 3.08, 95% confidence interval 1.71-5.56, p = .0002) and greater response rates as measured by a ≥50% improvement in symptom scores (odds ratio 2.68, 95% confidence interval 1.23-5.82, p = .01); no significant difference in response rate as measured by Clinical Global Impressions score ≤2 was found. Stouffer's z-score method showed significant improvement across all methylphenidate outcomes in G-allele carriers relative to noncarriers (z = 3.03, p = .002).

Conclusions

These findings suggest that carriers of rs1800544 may have improved ADHD outcomes following methylphenidate treatment. However, the extent to which these improvements are clinically impactful remain unclear. Additional studies are required to determine if rs1800544 carrier status should influence clinical recommendations for treatment of ADHD symptoms.

Predicting Children with ADHD Using Behavioral Activity: A Machine Learning Analysis

Attention deficit hyperactivity disorder (ADHD) is one of childhood’s most frequent neurobehavioral disorders. The purpose of this study is to: (i) extract the most prominent risk factors for children with ADHD; and (ii) propose a machine learning (ML)-based approach to classify children as either having ADHD or healthy. We extracted the data of 45,779 children aged 3–17 years from the 2018–2019 National Survey of Children’s Health (NSCH, 2018–2019). About 5218 (11.4%) of children were ADHD, and the rest of the children were healthy. Since the class label is highly imbalanced, we adopted a combination of oversampling and undersampling approaches to make a balanced class label. We adopted logistic regression (LR) to extract the significant factors for children with ADHD based on p-values (<0.05). Eight ML-based classifiers such as random forest (RF), Naïve Bayes (NB), decision tree (DT), XGBoost, k-nearest neighborhood (KNN), multilayer perceptron (MLP), support vector machine (SVM), and 1-dimensional convolution neural network (1D CNN) were adopted for the prediction of children with ADHD. The average age of the children with ADHD was 12.4 ± 3.4 years. Our findings showed that RF-based classifier provided the highest classification accuracy of 85.5%, sensitivity of 84.4%, specificity of 86.4%, and an AUC of 0.94. This study illustrated that LR with RF-based system could provide excellent accuracy for classifying and predicting children with ADHD. This system will be helpful for early detection and diagnosis of ADHD.

Influence of Atomoxetine on Relationship Between ADHD Symptoms and Prefrontal Cortex Activity During Task Execution in Adult Patients

Objective: We conducted this non-randomized prospective interventional study to clarify the relationship between improved attention-deficit hyperactivity disorder (ADHD) symptoms and regional brain activity. Methods: Thirty-one adult patients underwent near-infrared spectroscopy examinations during a go/no-go task, both before and 8 weeks after atomoxetine administration. Results: Clinical symptoms, neuropsychological results of the go/no-go task, and bilateral lateral prefrontal activity significantly changed. A positive correlation was observed between right dorsolateral prefrontal cortex activity and Conners' Adult ADHD Rating Scales scores. Before atomoxetine administration, no correlations between prefrontal cortex activity and clinical symptoms were observed in all cases. When participants were divided into atomoxetine-responder and non-responder groups, a positive correlation was observed between prefrontal cortex activity and clinical symptoms in the non-responder group before treatment but not in the responder group, suggesting that non-responders can activate the prefrontal cortex without atomoxetine. Conclusions: Individuals with increased ADHD symptoms appear to recruit the right dorsolateral prefrontal cortex more strongly to perform the same task than those with fewer symptoms. In clinical settings, individuals with severe symptoms are often observed to perform more difficultly when performing the tasks which individuals with mild symptoms can perform easily. The atomoxetine-responder group was unable to properly activate the right dorsolateral prefrontal cortex when necessary, and the oral administration of atomoxetine enabled these patients to activate this region. In brain imaging studies of heterogeneous syndromes such as ADHD, the analytical strategy used in this study, involving drug-responsivity grouping, may effectively increase the signal-to-noise ratio.

Noradrenergic genes polymorphisms and response to methylphenidate in children with ADHD: A systematic review and meta-analysis

BACKGROUND

Attention-deficit hyperactivity disorder (ADHD) is the most common childhood-onset neurodevelopmental disorder, and methylphenidate (MPH) is considered one of the first-line medicine for ADHD. Unfortunately, this medication is only effective for some children with ADHD. This meta-analysis was conducted to evaluate whether noradrenergic gene polymorphisms impact the efficacy of MPH in children with ADHD.

METHODS

Candidate gene studies published in English until March 1, 2020, were identified through literature searches on PubMed, Web of Science, and Embase. Data were pooled from individual clinical trials considering MPH pharmacogenomics. According to the heterogeneity, the odds ratio and mean differences were calculated by applying fixed-effects or random-effects models.

RESULTS

This meta-analysis includes 15 studies and 1382 patients. Four polymorphisms of the NET gene (rs5569, rs28386840, rs2242446, rs3785143) and 2 polymorphisms of the α2A-adrenergic receptor gene (ADRA2A) gene (MspI and DraI) were selected for the analysis. In the pooled data from all studies, T allele carriers of the rs28386840 polymorphism were significantly more likely to respond to MPH (P < .001, ORTcarriers = 2.051, 95% confidence interval [CI]:1.316, 3.197) and showed a relationship with significantly greater hyperactive-impulsive symptoms improvement (P < .001, mean difference:1.70, 95% CI:0.24, 3.16). None of the ADRA2A polymorphisms correlated significantly with MPH response as a whole. However, G allele carriers of the MspI polymorphism showed a relationship with significantly inattention symptoms improvement (P < .001, mean difference:0.31, 95% CI: 0.15, 0.47).

CONCLUSION

Our meta-analysis results indicate that the noradrenergic gene polymorphisms may impact MPH response. The NET rs28386840 is linked to improved MPH response in ADHD children. And the ADRA2A MspI is associated with inattention symptom improvements. Further investigations with larger samples will be needed to confirm these results.Registration: PROSPERO (no. CRD42021265830).

Patterns of Response to Methylphenidate Administration in Children with ADHD: A Personalized Medicine Approach through Clustering Analysis

Individual responses to methylphenidate (MPH) can significantly differ in children with attention-deficit/hyperactivity disorder (ADHD) in terms of the extent of clinical amelioration, optimal dosage needed, possible side effects, and short- and long-term duration of the benefits. In the present repeated-measures observational study, we undertook a proof-of-concept study to determine whether clustering analysis could be useful to characterize different clusters of responses to MPH in children with ADHD. We recruited 33 children with ADHD who underwent a comprehensive clinical, cognitive, and neurophysiological assessment before and after one month of MPH treatment. Symptomatology changes were assessed by parents and clinicians. The neuropsychological measures used comprised pen-and-paper and computerized tasks. Functional near-infrared spectroscopy was used to measure cortical hemodynamic activation during an attentional task. We developed an unsupervised machine learning algorithm to characterize the possible clusters of responses to MPH in our multimodal data. A symptomatology improvement was observed for both clinical and neuropsychological measures. Our model identified distinct clusters of amelioration that were related to symptom severity and visual-attentional performances. The present findings provide preliminary evidence that clustering analysis can potentially be useful in identifying different responses to MPH in children with ADHD, highlighting the importance of a personalized medicine approach within the clinical framework.

Identification of impulsive adolescents with a functional near infrared spectroscopy (fNIRS) based decision support system

Background.The gold standard for diagnosing impulsivity relies on clinical interviews, behavioral questionnaires and rating scales which are highly subjective.Objective.The aim of this study was to develop a functional near infrared spectroscopy (fNIRS) based classification approach for correct identification of impulsive adolescents. Taking into account the multifaceted nature of impulsivity, we propose that combining informative features from clinical, behavioral and neurophysiological domains might better elucidate the neurobiological distinction underlying symptoms of impulsivity.Approach. Hemodynamic and behavioral information was collected from 38 impulsive adolescents and from 33 non-impulsive adolescents during a Stroop task with concurrent fNIRS recordings. Connectivity-based features were computed from the hemodynamic signals and a neural efficiency metric was computed by fusing the behavioral and connectivity-based features. We tested the efficacy of two commonly used supervised machine-learning methods, namely the support vector machines (SVM) and artificial neural networks (ANN) in discriminating impulsive adolescents from their non-impulsive peers when trained with multi-domain features. Wrapper method was adapted to identify the informative biomarkers in each domain. Classification accuracies of each algorithm were computed after 10 runs of a 10-fold cross-validation procedure, conducted for 7 different combinations of the 3-domain feature set.Main results.Both SVM and ANN achieved diagnostic accuracies above 90% when trained with Wrapper-selected clinical, behavioral and fNIRS derived features. SVM performed significantly higher than ANN in terms of the accuracy metric (92.2% and 90.16%, respectively,p= 0.005).Significance.Preliminary findings show the feasibility and applicability of both machine-learning based methods for correct identification of impulsive adolescents when trained with multi-domain data involving clinical interviews, fNIRS based biomarkers and neuropsychiatric test measures. The proposed automated classification approach holds promise for assisting the clinical practice of diagnosing impulsivity and other psychiatric disorders. Our results also pave the path for a computer-aided diagnosis perspective for rating the severity of impulsivity.

Leveraging Machine Learning to Identify Predictors of Receiving Psychosocial Treatment for Attention Deficit/Hyperactivity Disorder

This study aimed to identify factors associated with receiving psychosocial treatment for ADHD in a nationally representative sample. Participants were 6630 youth with a parent-reported diagnosis of ADHD from the 2016-2017 National Survey of Children's Health. Machine learning analyses were performed to identify factors associated with receipt of psychosocial treatment for ADHD. We examined potentially associated factors in the broad categories of variables hypothesized to affect problem recognition (e.g., severity, mental health comorbidities); the decision to seek treatment; service selection (e.g., insurance coverage) and service use. We found that three machine learning models unanimously identified parent-reported ADHD severity (mild vs. moderate/severe) as the factor that best distinguishes between children who receive psychosocial treatment for ADHD and those who do not. Receive operating characteristic curve analysis revealed the following model performance: classification and regression tree analysis (area under the curve; AUC = .68); an ensemble model (AUC = .71); and a deep, multi-layer neural network (AUC = .72), as well as comparison to a logistic regression model (AUC = .69). Further, insurance coverage of mental/behavioral health needs emerged as a salient factor associated with the receipt of psychosocial treatment. Machine learning models identified risk and protective factors that predicted the receipt of psychosocial treatment for ADHD, such as ADHD severity and health insurance coverage.

Predicting the course of ADHD symptoms through the integration of childhood genomic, neural, and cognitive features

Childhood attention deficit hyperactivity disorder (ADHD) shows a highly variable course with age: some individuals show improving, others stable or worsening symptoms. The ability to predict symptom course could help individualize treatment and guide interventions. By studying a cohort of 362 youth, we ask if polygenic risk for ADHD, combined with baseline neural and cognitive features could aid in the prediction of the course of symptoms over an average period of 4.8 years. Compared to a never-affected comparison group, we find that participants with worsening symptoms carried the highest polygenic risk for ADHD, followed by those with stable symptoms, then those whose symptoms improved. Participants with worsening symptoms also showed atypical baseline cognition. Atypical microstructure of the cingulum bundle and anterior thalamic radiation was associated with improving symptoms while reduction of thalamic volume was found in those with stable symptoms. Machine-learning algorithms, trained and tested on independent groups, performed well in classifying those never affected against groups with worsening, stable, and improving symptoms (area under the curve >0.79). We conclude that some measures of polygenic risk, cognition, and neuroimaging show significant associations with the future course of ADHD symptoms and may have modest predictive power. These features warrant further exploration as prognostic tools.

Individual Variability in Reaction Time and Prediction of Clinical Response to Methylphenidate in Adult ADHD: A Prospective Open Label Study Using Conners' Continuous Performance Test II

Objective: The aim of this study was to examine whether reaction time parameters in adult patients with ADHD could predict their response to methylphenidate (MPH). Method: Previously unmedicated patients (N = 123) were administered the Conners' Continuous Performance Test II (CPT II) at baseline and after 6 weeks of treatment with immediate-release MPH. In addition to traditional CPT measures, we extracted intraindividual raw data and analyzed time series using linear and nonlinear mathematical models. Results: Clinical responders, assessed with the Clinical Global Impression-Improvement scale, showed significant normalization of target failures, reduced variability and skewness, and increased complexity of reaction time series after 6 weeks of treatment, while nonresponders showed no significant changes. Prior to treatment, responders had significantly higher variability and skewness, combined with lower complexity, compared with nonresponders. Conclusion: These results show that the CPT test is useful in the evaluation of treatment response to MPH.

Early response to SPN-812 (viloxazine extended-release) can predict efficacy outcome in pediatric subjects with ADHD: a machine learning post-hoc analysis of four randomized clinical trials

Machine learning (ML) was used to determine whether early response can predict efficacy outcome in pediatric subjects with ADHD treated with SPN-812. We used data from four Phase 3 placebo-controlled trials of 100- to 600-mg/day SPN-812 (N=1397; 6-17 years of age). The treatment response was defined as having a ≥50% reduction in change from baseline (CFB) in ADHD Rating Scale-5 (ADHD-RS-5) Total score at Week 6. The variables used were: ADHD-RS-5 Total score, age, body weight, and body mass index at baseline; CFB ADHD-RS-5 Total score at Week 1, cumulative change in ADHD-RS-5 Total score at Week 2, and cumulative change in ADHD-RS-5 Total score at Week 3; Clinical Global Impressions-Improvement (CGI-I) score at Week 1, 2, and 3; and target dose. Using the best selected model, lasso regression, to generate importance scores, we found that change in ADHD-RS-5 Total score and CGI-I score were the best predictors of efficacy outcome. Change in ADHD-RS-5 Total score at Week 2 could predict treatment response at Week 6 (75% positive predictive power, 75% sensitivity, 74% specificity). Therefore, early response after two weeks of treatment with once-daily SPN-812 in pediatric patients with ADHD can predict efficacy outcome at Week 6.

A Machine-Based Prediction Model of ADHD Using CPT Data

Despite the popularity of the continuous performance test (CPT) in the diagnosis of attention-deficit/hyperactivity disorder (ADHD), its specificity, sensitivity, and ecological validity are still debated. To address some of the known shortcomings of traditional analysis and interpretation of CPT data, the present study applied a machine learning-based model (ML) using CPT indices for the Prediction of ADHD.Using a retrospective factorial fitting, followed by a bootstrap technique, we trained, cross-validated, and tested learning models on CPT performance data of 458 children aged 6–12 years (213 children with ADHD and 245 typically developed children). We used the MOXO-CPT version that included visual and auditory stimuli distractors. Results showed that the ML proposed model performed better and had a higher accuracy than the benchmark approach that used clinical data only. Using the CPT total score (that included all four indices: Attention, Timeliness, Hyperactivity, and Impulsiveness), as well as four control variables [age, gender, day of the week (DoW), time of day (ToD)], provided the most salient information for discriminating children with ADHD from their typically developed peers. This model had an accuracy rate of 87%, a sensitivity rate of 89%, and a specificity rate of 84%. This performance was 34% higher than the best-achieved accuracy of the benchmark model. The ML detection model could classify children with ADHD with high accuracy based on CPT performance. ML model of ADHD holds the promise of enhancing, perhaps complementing, behavioral assessment and may be used as a supportive measure in the evaluation of ADHD.

Interaction between lead and noradrenergic genotypes affects neurocognitive functions in attention-deficit/hyperactivity disorder: a case control study

BACKGROUND

Lead is known to be associated with attention-deficit/hyperactivity disorder (ADHD) even at low concentrations. We aimed to evaluate neurocognitive functions associated with lead in the blood and the interactions between lead and dopaminergic or noradrenergic pathway-related genotypes in youths with ADHD.

METHODS

A total of 259 youths with ADHD and 96 healthy controls (aged 5-18 years) enrolled in this study. The Korean Kiddie Schedule for Affective Disorders and Schizophrenia-Present and Lifetime version was conducted for psychiatric diagnostic evaluation. Blood lead levels were measured, and their interaction with dopaminergic or noradrenergic genotypes for ADHD; namely, the dopamine transporter (DAT1), dopamine receptor D4 (DRD4), and alpha-2A-adrenergic receptor (ADRA2A) genotypes were investigated. All participants were assessed using the ADHD Rating Scale-IV (ADHD-RS). Participants also completed the continuous performance test (CPT) and Stroop Color-Word Test (SCWT). Analysis of covariance was used for comparison of blood lead levels between ADHD and control groups. A multivariable linear regression model was used to evaluate the associations of blood lead levels with the results of ADHD-RS, CPT, and SCWT; adjusted for intelligence quotient (IQ), age, and sex. A path analysis model was used to identify the mediating effects of neurocognitive functions on the effects of blood lead on ADHD symptoms. To evaluate the effect of the interaction between blood lead and genes on neuropsychological functions, hierarchical regression analyses were performed.

RESULTS

There was a significant difference in blood lead levels between the ADHD and control groups (1.4 ± 0.5 vs. 1.3 ± 0.5 μg/dL, p = .005). Blood lead levels showed a positive correlation with scores on omission errors(r = .158, p = .003) and response time variability (r = .136, p = .010) of CPT. In the multivariable linear regression model, blood lead levels were associated with omission errors (B = 3.748, p = .045). Regarding the effects of lead on ADHD symptoms, hyperactivity-impulsivity was mediated by omission errors. An interaction effect was detected between ADRA2A DraI genotype and lead levels on omission errors (B = 5.066, p = .041).

CONCLUSIONS

Our results indicate that neurocognitive functions at least partly mediate the association between blood lead levels and ADHD symptoms, and that neurocognitive functions are affected by the interaction between blood lead levels and noradrenergic genotype.

ShinyLearner: A containerized benchmarking tool for machine-learning classification of tabular data

BACKGROUND

Classification algorithms assign observations to groups based on patterns in data. The machine-learning community have developed myriad classification algorithms, which are used in diverse life science research domains. Algorithm choice can affect classification accuracy dramatically, so it is crucial that researchers optimize the choice of which algorithm(s) to apply in a given research domain on the basis of empirical evidence. In benchmark studies, multiple algorithms are applied to multiple datasets, and the researcher examines overall trends. In addition, the researcher may evaluate multiple hyperparameter combinations for each algorithm and use feature selection to reduce data dimensionality. Although software implementations of classification algorithms are widely available, robust benchmark comparisons are difficult to perform when researchers wish to compare algorithms that span multiple software packages. Programming interfaces, data formats, and evaluation procedures differ across software packages; and dependency conflicts may arise during installation.

FINDINGS

To address these challenges, we created ShinyLearner, an open-source project for integrating machine-learning packages into software containers. ShinyLearner provides a uniform interface for performing classification, irrespective of the library that implements each algorithm, thus facilitating benchmark comparisons. In addition, ShinyLearner enables researchers to optimize hyperparameters and select features via nested cross-validation; it tracks all nested operations and generates output files that make these steps transparent. ShinyLearner includes a Web interface to help users more easily construct the commands necessary to perform benchmark comparisons. ShinyLearner is freely available at https://github.com/srp33/ShinyLearner.

CONCLUSIONS

This software is a resource to researchers who wish to benchmark multiple classification or feature-selection algorithms on a given dataset. We hope it will serve as example of combining the benefits of software containerization with a user-friendly approach.

Continuous Performance Test Is Sensitive to a Single Methylphenidate Challenge in Preschool Children With ADHD

Objective: There is a lack of evidence-based diagnostic paradigms and personalized interventions for preschoolers with ADHD. This study aimed to evaluate the performance of preschoolers diagnosed with ADHD on a continuous performance test (CPT) before and after a single methylphenidate (MPH) challenge. Method: The Test of Variables of Attention (TOVA)-a CPT-was administered to 61 preschoolers (5.64 ± 0.69 years; 74% boys) with ADHD before and after a single MPH challenge (0.3 or 0.5 mg/kg). Baseline TOVA performance was correlated with Conners' Rating Scales (CRS) and compared with post-MPH TOVA performance. Results: A high rate of omission errors and several significant correlations between TOVA values and CRS scores were found at baseline. A single MPH administration improved TOVA performance significantly and was well tolerated. Conclusion: TOVA assessment may assist in the evaluation of the effect of MPH in preschoolers with ADHD and may help in planning interventions for them.

A Predictive Model Offor Attention Deficit Hyperactivity Disorder Based on Clinical Assessment Tools

BACKGROUND

At present, clinicians diagnose that the clinical diagnosis of attention deficit hyperactivity disorder (ADHD) in children is mainly on the basis of the information provided by their parents, the behaviour of children in clinical clinics and the assessments of clinical rating scales and neuropsychological tests. Notably, no unified standard exists currently for analysing the results of various measurement tools for diagnosing ADHD. Therefore, clinicians interpret the results of clinical rating scales and neuropsychological tests entirely based on their clinical experience.

METHODS AND SUBJECTS

To provide guidance for clinicians on how to analyse the results of various clinical assessment tools when diagnosing ADHD, this study assessed children with ADHD and children in the control group using two clinical assessment scales-parent rating scale (PSQ) and Child Behavior Checklist (CBCL)-and one neuropsychological test (Integrated Visual and Auditory Continuous Performance Testing). The two-sample t-test (FDR correction) screened the parameters of the three assessment tools with significant inter-group differences. LibSVM was used to establish a classification prediction model for analysing the accuracy of ADHD prediction using parameters of the three assessment tools and weight values of each parameter for classification prediction.

RESULTS

A total of 19 parameters (16 from clinical rating scales, 3 from neuropsychological tests) with significant inter-group differences were screened. The accuracy of classification modelling was higher for the clinical rating scales (61.635%) than for the neuropsychological test (59.784%), whereas the accuracy of classification modelling was higher for the clinical rating scales combined with the neuropsychological test (70.440%) than for the former two parameters alone. The three parameters with the highest weight values were learning problem (0.731), hyperactivity/impulsivity (0.676) and activity capacity (0.569). The three parameters with the lowest weight values are integrated control force (0.028), visual attention (0.028) and integrated attention (0.034).

CONCLUSION

Our study findings indicate that the diagnosis of ADHD should be based on multidimensional assessment. For a more accurate diagnosis of ADHD, assessments and that more assessment parameters should be developed on the basis of different dimensions of physiology or psychology in the future to obtain a more accurate diagnosis of ADHD. Furthermore, the predictive model for ADHD may improve our understanding and help in optimisation of the treatment of such a condition.

Clinical and Neuropsychological Predictors of Methylphenidate Response in Children and Adolescents with ADHD: A Naturalistic Follow-up Study in a Spanish Sample

BACKGROUND

Methylphenidate (MPH) is the most commonly used medication for Attention-Deficit/Hyperactivity Disorder (ADHD), but to date, there are neither consistent nor sufficient findings on conditions differentiating responsiveness to MPH response in ADHD.

OBJECTIVE

To develop a predictive model of MPH response, using a longitudinal and naturalistic follow-up study, in a Spanish sample of children and adolescents with ADHD.

METHODS

We included all children and adolescents with ADHD treated with MPH in our outpatient Clinic (2005 to 2015), evaluated with the K-SADS interview. We collected ADHD-RS-IV.es and CGI-S scores at baseline and at follow up, and neuropsychological testing (WISC-IV, Continuous Performance Test (CPT-II) & Stroop). Clinical response was defined as >30% reduction from baseline of total ADHD-RS-IV.es score and CGI-S final score of 1 or 2 maintained for the previous 3 months.

RESULTS

We included 518 children and adolescents with ADHD, mean (SD) age of patients was 11.4 (3.3) years old; 79% male; 51.7% had no comorbidities; and 75.31% had clinical response to a mean MPH dose of 1.2 mg/kg/day. Lower ADHD-RS-IV.es scores, absence of comorbidities (oppositional-defiant symptoms, depressive symptoms and alcohol/cannabis use), fewer altered neuropsychological tests, higher total IQ and low commission errors in CPT-II, were significantly associated with a complete clinical response to methylphenidate treatment.

CONCLUSION

Oppositional-defiant symptoms, depressive symptoms, and a higher number of impaired neuropsychological tests are associated with worse clinical response to methylphenidate. Other stimulants or non-stimulants treatment may be considered when these clinical and neuropsychological variables converged in the first clinical interview.

Role of deep learning in infant brain MRI analysis

Deep learning algorithms and in particular convolutional networks have shown tremendous success in medical image analysis applications, though relatively few methods have been applied to infant MRI data due numerous inherent challenges such as inhomogenous tissue appearance across the image, considerable image intensity variability across the first year of life, and a low signal to noise setting. This paper presents methods addressing these challenges in two selected applications, specifically infant brain tissue segmentation at the isointense stage and presymptomatic disease prediction in neurodevelopmental disorders. Corresponding methods are reviewed and compared, and open issues are identified, namely low data size restrictions, class imbalance problems, and lack of interpretation of the resulting deep learning solutions. We discuss how existing solutions can be adapted to approach these issues as well as how generative models seem to be a particularly strong contender to address them.

Psychoradiological Biomarkers for Psychopharmaceutical Effects

The application of personalized medicine to psychiatry is challenging. Psychoradiology could provide biomarkers based on objective tests in support of the diagnostic classifications and treatment planning. We review potential psychoradiological biomarkers for psychopharmaceutical effects. Although none of the biomarkers reviewed are yet of sufficient clinical utility to inform the selection of a specific pharmacologic compound for an individual patient, there is strong consensus that advanced multimodal approaches will contribute to discovery of novel treatment predictors in psychiatric disorders. Progress has been sufficient to warrant enthusiasm, in which application of neuroimaging-based biomarkers would represent a paradigm shift and modernization of psychiatric practice.

Imaging-Based Subtyping for Psychiatric Syndromes

Despite considerable research evidence demonstrating significant neurobiological alterations in psychiatric disorders, incorporating neuroimaging approaches into clinical practice remains challenging. There is an urgent need for biologically validated psychiatric disease constructs that can inform diagnostic algorithms and targeted treatment development. In this article, we present a conceptual review of the most robust and impactful findings from studies that use neuroimaging methods in efforts to define distinct disease subtypes, while emphasizing cross-diagnostic and dimensional approaches. In addition, we discuss current challenges in psychoradiology and outline potential future strategies for clinically applicable translation.

Prediction of sleep side effects following methylphenidate treatment in ADHD youth

•

Sleep problems after stimulant use in ADHD were predicted using machine learning.

•

Step-wise combination of multi-level features enhanced prediction performance.

•

Fronto-striatal connectivity and DAT1, ADRA2A, SLC6A2 SNPs were key features.

•

An accuracy of 95.5% was achieved by Logistic Ridge Regression in the training data.

•

An accuracy of 86.1% was achieved by J48 in the independent validation analysis.

Objective

Sleep problems is the most common side effect of methylphenidate (MPH) treatment in ADHD youth and carry potential to negatively impact long-term self-regulatory functioning. This study aimed to examine whether applying machine learning approaches to pre-treatment demographic, clinical questionnaire, environmental, neuropsychological, genetic, and neuroimaging features can predict sleep side effects following MPH administration.

Method

The present study included 83 ADHD subjects as a training dataset. The participants were enrolled in an 8-week, open-label trial of MPH. The Barkley Stimulant Side Effects Rating Scale was used to determine the presence/absence of sleep problems at the 2nd week of treatment. Prediction of sleep side effects were performed with step-wise addition of variables measured at baseline: demographics (age, gender, IQ, height/weight) and clinical variables (ADHD Rating Scale-IV (ADHD-RS) and Disruptive Behavior Disorder rating scale) at stage 1, neuropsychological test (continuous performance test (CPT), Stroop color word test) and genetic/environmental variables (dopamine and norepinephrine receptor gene (DAT1, DRD4, ADRA2A, and SLC6A2) polymorphisms, blood lead, and urine cotinine level) at stage 2, and structural connectivities of frontostriatal circuits at stage 3. Three different machine learning algorithms ((Logistic Ridge Regression (LR), support vector machine (SVM), J48) were used for data analysis. Robustness of classifier model was validated in the independent dataset of 36 ADHD subjects.

Results

Classification accuracy of LR was 95.5% (area under receiver operating characteristic curve (AUC) 0.99), followed by SVM (91.0%, AUC 0.85) and J48 (90.0%, AUC 0.87) at stage 3 for predicting sleep problems. The inattention symptoms of ADHD-RS, CPT response time variability, the DAT1, ADRA2A DraI, and SLC6A2 A-3081T polymorphisms, and the structural connectivities between frontal and striatal brain regions were identified as the most differentiating subset of features. Validation analysis achieved accuracy of 86.1% (AUC 0.92) at stage 3 with J48.

Conclusions

Our results provide preliminary support to the combination of multimodal classifier, in particular, neuroimaging features, as an informative method that can assist in predicting MPH side effects in ADHD.

The Application of Machine Learning Techniques in Clinical Drug Therapy

INTRODUCTION

The development of a novel drug is an extremely complicated process that includes the target identification, design and manufacture, and proper therapy of the novel drug, as well as drug dose selection, drug efficacy evaluation, and adverse drug reaction control. Due to the limited resources, high costs, long duration, and low hit-to-lead ratio in the development of pharmacogenetics and computer technology, machine learning techniques have assisted novel drug development and have gradually received more attention by researchers.

METHODS

According to current research, machine learning techniques are widely applied in the process of the discovery of new drugs and novel drug targets, the decision surrounding proper therapy and drug dose, and the prediction of drug efficacy and adverse drug reactions.

RESULTS AND CONCLUSION

In this article, we discussed the history, workflow, and advantages and disadvantages of machine learning techniques in the processes mentioned above. Although the advantages of machine learning techniques are fairly obvious, the application of machine learning techniques is currently limited. With further research, the application of machine techniques in drug development could be much more widespread and could potentially be one of the major methods used in drug development.

Alterations in resting-state functional connectivity in substance use disorders and treatment implications

Substance use disorders (SUD) are diseases of the brain, characterized by aberrant functioning in the neural circuitry of the brain. Resting state functional connectivity (rsFC) can illuminate these functional changes by measuring the temporal coherence of low-frequency fluctuations of the blood oxygenation level-dependent magnetic resonance imaging signal in contiguous or non-contiguous regions of the brain. Because this data is easy to obtain and analyze, and therefore fairly inexpensive, it holds promise for defining biological treatment targets in SUD, which could help maximize the efficacy of existing clinical interventions and develop new ones. In an effort to identify the most likely "treatment targets" obtainable with rsFC we summarize existing research in SUD focused on 1) the relationships between rsFC and functionality within important psychological domains which are believed to underlie relapse vulnerability 2) changes in rsFC from satiety to deprived or abstinent states 3) baseline rsFC correlates of treatment outcome and 4) changes in rsFC induced by treatment interventions which improve clinical outcomes and reduce relapse risk. Converging evidence indicates that likely "treatment target" candidates, emerging consistently in all four sections, are reduced connectivity within executive control network (ECN) and between ECN and salience network (SN). Other potential treatment targets also show promise, but the literature is sparse and more research is needed. Future research directions include data-driven prediction analyses and rsFC analyses with longitudinal datasets that incorporate time since last use into analysis to account for drug withdrawal. Once the most reliable biological markers are identified, they can be used for treatment matching, during preliminary testing of new pharmacological compounds to establish clinical potential ("target engagement") prior to carrying out costly clinical trials, and for generating hypotheses for medication repurposing.

Medical treatment of Attention Deficit/Hyperactivity Disorder (ADHD) and children's academic performance

Attention Deficit/Hyperactivity Disorder (ADHD) is negatively associated with a range of academic achievement measures. We use Danish administrative register data to study the impact of medical treatment of ADHD on children's academic performance assessed by student grade point average (GPA). Using administrative register data on children, who begin medical treatment, we conduct a natural experiment and exploit plausible exogenous variation in medical nonresponse to estimate the effect of medical treatment on school-leaving GPA. We find significant effects of treatment on both exam and teacher evaluated GPAs: Compared to consistent treatment, part or full discontinuation of treatment has large significant negative effects reducing teacher evaluation and exam GPA with .18 and .22 standard deviations, respectively. The results demonstrate that medical treatment may mitigate the negative social consequences of ADHD. Placebo regressions indicate that a causal interpretation of our findings is plausible.

What is the level of evidence for the use of currently available technologies in facilitating the self-management of difficulties associated with ADHD in children and young people? A systematic review

A number of technologies to help self-manage attention deficit hyperactivity disorder (ADHD) in children and young people (YP) have been developed. This review will assess the level of evidence for the use of such technologies. The review was undertaken in accordance with the general principles recommended in the Preferred Reporting Items for Systematic Reviews and Meta-Analysis. 7545 studies were screened. Fourteen studies of technology that aim to self-manage difficulties associated with ADHD in children and YP were included. Primary outcome measures were measures that assessed difficulties related to ADHD. Databases searched were MEDLINE, Web of Science (Core collection), CINAHL, the Cochrane Library, ProQuest ASSIA, PsycINFO and Scopus. The methodological quality of the studies was assessed. This review highlights the potential for the use of technology in paediatric ADHD management. However, it also demonstrates that current research lacks robustness; using small sample sizes, non-validated outcome measures and little psychoeducation component. Future research is required to investigate the value of technology in supporting children and YP with ADHD and a focus psychoeducation is needed.

Effects of Interaction Between DRD4 Methylation and Prenatal Maternal Stress on Methylphenidate-Induced Changes in Continuous Performance Test Performance in Youth with Attention-Deficit/Hyperactivity Disorder

OBJECTIVES

Environmental factors may interact with genetic factors via the epigenetic process, and this interaction can contribute to inter-individual variability in the treatment response. The purpose of this study was to investigate the interaction effects between dopamine receptor D4 (DRD4) methylation and prenatal maternal stress on the methylphenidate (MPH) response of youth with attention-deficit/hyperactivity disorder (ADHD).

METHODS

This study was an 8-week open-label trial of MPH that included 74 ADHD youth. We investigated the associations between MPH treatment response, which was defined as a score ≤2 on the Clinical Global Impressions-Improvement (CGI-I) scale, and the methylation of 28 cytosine-guanine dinucleotide (CpG) sites of DRD4. Additionally, the interaction effects between DRD4 methylation and prenatal maternal stress on changes in Continuous Performance Test (CPT) scores after MPH treatment were investigated.

RESULTS

Although there were no significant sites that showed significant association with treatment response, there was a significant interaction effect of the methylation of CpG7 and prenatal maternal stress on changes in omission errors of the CPT following treatment (p = 0.0001).

CONCLUSIONS

The present findings indicate that the interaction between methylation of CpG7 of DRD4 and prenatal maternal stress may be predictive of the treatment response to MPH in youth with ADHD.

Pharmacogenetics predictors of methylphenidate efficacy in childhood ADHD

Stimulant medication has long been effective in treating attention-deficit/hyperactivity disorder (ADHD) and is currently the first-line pharmacological treatment for children. Both methylphenidate and amphetamine modulate extracellular catecholamine levels through interaction with dopaminergic, adrenergic and serotonergic system components; it is therefore likely that catecholaminergic molecular components influence the effects of ADHD treatment. Using meta-analysis, we sought to identify predictors of pharmacotherapy to further the clinical implementation of personalized medicine. We identified 36 studies (3647 children) linking the effectiveness of methylphenidate treatment with DNA variants. Pooled-data revealed a statistically significant association between single nucleotide polymorphisms (SNPs) rs1800544 ADRA2A (odds ratio: 1.69; confidence interval: 1.12-2.55), rs4680 COMT (odds ratio (OR): 1.40; confidence interval: 1.04-1.87), rs5569 SLC6A2 (odds ratio: 1.73; confidence interval: 1.26-2.37) and rs28386840 SLC6A2 (odds ratio: 2.93; confidence interval: 1.76-4.90), and, repeat variants variable number tandem repeat (VNTR) 4 DRD4 (odds ratio: 1.66; confidence interval: 1.16-2.37) and VNTR 10 SLC6A3 (odds ratio: 0.74; confidence interval: 0.60-0.90), whereas the following variants were not statistically significant: rs1947274 LPHN3 (odds ratio: 0.95; confidence interval: 0.71-1.26), rs5661665 LPHN3 (odds ratio: 1.07; confidence interval: 0.84-1.37) and VNTR 7 DRD4 (odds ratio: 0.68; confidence interval: 0.47-1.00). Funnel plot asymmetry among SLC6A3 studies was identified and attributed largely to small study effects. Egger's regression test and Duval and Tweedie's 'trim and fill' were used to examine and correct for publication bias. These findings have major implications for advancing our therapeutic approach to childhood ADHD treatment.

SLC6A3 polymorphism and response to methylphenidate in children with ADHD: A systematic review and meta-analysis

Methylphenidate (MPH) is the most commonly used treatment for attention-deficit hyperactivity disorder (ADHD) in children. However, the response to MPH is not similar in all patients. This meta-analysis investigated the potential role of SLC6A3 polymorphisms in response to MPH in children with ADHD. Clinical trials or naturalistic studies were selected from electronic databases. A meta-analysis was conducted using a random-effects model. Cohen's d effect size and 95% confidence intervals (CIs) were determined. Sensitivity analysis and meta-regression were performed. Q-statistic and Egger's tests were conducted to evaluate heterogeneity and publication bias, respectively. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) system was used to assess the quality of evidence. Sixteen studies with follow-up periods of 1-28 weeks were eligible. The mean treatment acceptability of MPH was 97.2%. In contrast to clinical trials, the meta-analysis of naturalistic studies indicated that children without 10/10 repeat carriers had better response to MPH (Cohen's d: -0.09 and 0.44, respectively). The 9/9 repeat polymorphism had no effect on the response rate (Cohen's d: -0.43). In the meta-regression, a significant association was observed between baseline severity of ADHD, MPH dosage, and combined type of ADHD in some genetic models. Sensitivity analysis indicated the robustness of our findings. No publication bias was observed in our meta-analysis. The GRADE evaluations revealed very low levels of confidence for each outcome of response to MPH. The results of clinical trials and naturalistic studies regarding the effect size between different polymorphisms of SLC6A3 were contradictory. Therefore, further research is recommended.

Prediction of outcome in internet-delivered cognitive behaviour therapy for paediatric obsessive-compulsive disorder: A machine learning approach

BACKGROUND

There are no consistent predictors of treatment outcome in paediatric obsessive-compulsive disorder (OCD). One reason for this might be the use of suboptimal statistical methodology. Machine learning is an approach to efficiently analyse complex data. Machine learning has been widely used within other fields, but has rarely been tested in the prediction of paediatric mental health treatment outcomes.

OBJECTIVE

To test four different machine learning methods in the prediction of treatment response in a sample of paediatric OCD patients who had received Internet-delivered cognitive behaviour therapy (ICBT).

METHODS

Participants were 61 adolescents (12-17 years) who enrolled in a randomized controlled trial and received ICBT. All clinical baseline variables were used to predict strictly defined treatment response status three months after ICBT. Four machine learning algorithms were implemented. For comparison, we also employed a traditional logistic regression approach.

RESULTS

Multivariate logistic regression could not detect any significant predictors. In contrast, all four machine learning algorithms performed well in the prediction of treatment response, with 75 to 83% accuracy.

CONCLUSIONS

The results suggest that machine learning algorithms can successfully be applied to predict paediatric OCD treatment outcome. Validation studies and studies in other disorders are warranted.

The opportunities of virtual reality in the rehabilitation of children with attention deficit hyperactivity disorder: a literature review

Attention deficit hyperactivity disorder (ADHD) is one of the most common psychiatric disorders in childhood. This disorder, in addition to its main symptoms, creates significant difficulties in education, social performance, and personal relationships. Given the importance of rehabilitation for these patients to combat the above issues, the use of virtual reality (VR) technology is helpful. The aim of this study was to highlight the opportunities for VR in the rehabilitation of children with ADHD. This narrative review was conducted by searching for articles in scientific databases and e-Journals, using keywords including VR, children, and ADHD. Various studies have shown that VR capabilities in the rehabilitation of children with ADHD include providing flexibility in accordance with the patients' requirements; removing distractions and creating an effective and safe environment away from real-life dangers; saving time and money; increasing patients' incentives based on their interests; providing suitable tools to perform different behavioral tests and increase ecological validity; facilitating better understanding of individuals' cognitive deficits and improving them; helping therapists with accurate diagnosis, assessment, and rehabilitation; and improving working memory, executive function, and cognitive processes such as attention in these children. Rehabilitation of children with ADHD is based on behavior and physical patterns and is thus suitable for VR interventions. This technology, by simulating and providing a virtual environment for diagnosis, training, monitoring, assessment and treatment, is effective in providing optimal rehabilitation of children with ADHD.