Pubmed (TDAH) du 22/10/26
1. Geethapriya M, Pattabiraman K. Subject-independent classification of ADHD from pediatric EEG: a leakage-aware deep learning framework with interpretable neurophysiological mapping. Front Neurosci. 2026; 20: 1925266.
BACKGROUND: The use of electroencephalography (EEG) in automated screening of Attention-Deficit/Hyperactivity Disorder (ADHD) has created a large body of literature that suggests promising outcomes, with reported classification accuracies often above 90%. A major methodological issue suggests that a substantial part of these results are actually due to data leakage, whereby EEG epochs pertaining to the same subject end up in both training and testing sets, allowing classifiers to capitalize on features specific to the individual rather than on genuine disorder-related biomarkers. METHODS: We assessed the performance of three classical machine learning algorithms, namely, Logistic Regression (LR), Random Forest (RF), and Support Vector Machine (SVM), as well as three deep learning models: Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and CNN-BiLSTM, on a standard pediatric EEG dataset containing 121 children and 4,173 epochs. The performance of each model was compared in terms of both segment-wise cross-validation (Protocol A) and subject-wise cross-validation (without any data leakage, Protocol B).To evaluate neurotopographic plausibility without making unsupported causal assertions, we applied Deep SHapley Additive exPlanations (SHAP) to the best-generalizing architecture. RESULTS: The performance discrepancy was substantial. Under segment-wise evaluation, the SVM operating on power spectral density (PSD) features reached 90.0% accuracy (95% CI: [89.1%, 90.9%], AUC = 0.96). However, the same model dropped to 65.6% (95% CI: [62.1%, 69.1%], AUC = 0.72) under subject-wise evaluation-a statistically significant 24.4 percentage point collapse (p < 0.001, Wilcoxon signed-rank). In contrast, deep architectures learning directly from raw EEG were resilient: the stand-alone CNN lost only 2.5 percentage points, achieving 72.7% accuracy (95% CI: [66.0%, 79.4%], AUC = 0.72, recall = 88.0%) on completely unseen participants. Post-hoc SHAP attribution identified the highest relative contributions in frontal (F3, Fz, F4) and central (Cz) electrodes, aligning with established ADHD neurophysiology. CONCLUSION: Hand-engineered spectral features are highly vulnerable to identity leakage, whereas end-to-end convolutional models generalize across subjects with realistic, uninflated performance. Strict subject-wise validation is strongly recommended as standard practice for translational EEG research.
Lien vers le texte intégral (Open Access ou abonnement)
2. Huang YH, Huang KD, Zhan LF, Liu LQ, Liu LJ, Xiong J, Luo KJ, Yang F, Gong XX, Luo XM. [Neurophysiological differentiation of children with ADHD alone and ADHD comorbid with dyslexia: a P600-like component analysis]. Zhongguo Dang Dai Er Ke Za Zhi. 2026; 28(10): 1184-92.
OBJECTIVES: To investigate the clinical value of the latency of the P600-like event-related potential component during a spatial N-back task as a neurophysiological indicator for differentiating children with attention deficit hyperactivity disorder (ADHD) alone from those with ADHD comorbid with developmental dyslexia (DD). METHODS: In this prospective study, 66 ADHD children aged 6-10 years were enrolled from the Department of Child Health Care, Second Xiangya Hospital of Central South University between June and September 2025, including 28 with ADHD alone and 38 with ADHD+DD. An additional 20 healthy children were included solely as a visual reference for electroencephalogram (EEG) topographic mapping. Behavioral symptoms were assessed using the SNAP-IV, the Childhood Executive Functioning Inventory, the Conners Child Behavior Scale, and the Dyslexia Checklist for Chinese Children. EEG was recorded during a spatial N-back task, and the P600-like component was extracted from Fz, Cz, and Pz electrodes. A linear mixed model was applied to analyze the interactions among group, electrode site, and task load. Receiver operating characteristic (ROC) curves were used to evaluate discriminative performance, and 5-fold cross-validation was conducted to reduce overfitting risk. RESULTS: Under the 1-back condition, the latency of the P600-like component at Pz was significantly longer in the ADHD+DD group than in the ADHD-alone group (P<0.001). The area under the curve (AUC) for the P600-like component latency at Pz in distinguishing ADHD alone from ADHD+DD was 0.804 (95%CI: 0.697-0.910). Using a cut-off value of 688 ms, sensitivity was 63.2% and specificity 96.4%. Hierarchical logistic regression analysis showed that model 4 (P600-like component latency + N1 reaction time + age) achieved an AUC of 0.804, comparable to that of model 2 (model 1 + P600-like component peak value + P600-like component average amplitude; AUC=0.805), with superior cross-validation stability. CONCLUSIONS: The latency of the P600-like component at the Pz electrode during an N-back task may serve as a neurophysiological indicator to assist in differentiating whether children with ADHD have comorbid DD. A parsimonious model incorporating this indicator has discriminative performance comparable to the full-indicator model, with superior cross-validation stability.
Lien vers le texte intégral (Open Access ou abonnement)
3. Iacobucci G. AD-Smart: Alzheimer’s trial will test diabetes and ADHD drugs on patients. Bmj. 2026; 395: e101219.
Lien vers le texte intégral (Open Access ou abonnement)
4. Lal S, Khan E, Rasheed M, O’Shea M. Anaesthetic Management of a Child with Chromosome 7q32.1 Microdeletion Undergoing Ligation of Patent Processus Vaginalis. Ir Med J. 2026; 119(9): 182.
PRESENTATION: A six-year-old boy with chromosome 7q32.1 microdeletion, autism spectrum disorder, attention deficit hyperactivity disorder, hypotonia and asthma presented for elective ligation of a patent processus vaginalis under general anaesthesia. DIAGNOSIS: Chromosome 7q32.1 microdeletion is a rare genetic disorder with developmental, behavioural and neuromuscular features. Coexisting reactive airway disease and autism spectrum disorder heighten the risk of perioperative respiratory adverse events (PRAEs) and complicate induction and emergence. TREATMENT: A perioperative strategy minimising airway stimulation and preserving spontaneous ventilation was employed. Anaesthesia was induced with sevoflurane in oxygen; a laryngeal mask airway (LMA) was used instead of endotracheal intubation, and neuromuscular blockade was avoided. The perioperative course was uneventful, with no respiratory complications. DISCUSSION: Use of a supraglottic airway with maintenance of spontaneous ventilation, and a calm behaviourally tailored approach, facilitated safe anaesthesia in this child. This case adds to the limited literature on the anaesthetic implications of chromosome 7q32.1 microdeletion.
Lien vers le texte intégral (Open Access ou abonnement)
5. Milne S, Blake N, Jones E, Ingold C, Cihan B, Bridge L, Robinson JL, Taylor GM. Smoking cessation interventions in adults with attention deficit hyperactivity disorder (ADHD) or autism spectrum disorder (ASD). Cochrane Database Syst Rev. 2026; 10(10): Cd016400.
This is a protocol for a Cochrane review (intervention). The objectives are as follows: Primary objective To assess the effects of smoking cessation interventions, compared with no intervention, usual care, placebo, or alternative interventions, on sustained abstinence from smoking in adults with attention deficit hyperactivity disorder (ADHD), autism spectrum disorder (ASD), or co-occurring ADHD and ASD. Secondary objectives To assess whether the effects of smoking cessation interventions differ between adults with ADHD, ASD, and co‑occurring ADHD and ASD. To assess whether intervention effects vary according to participant characteristics (e.g. ADHD or ASD symptom profiles, co‑occurring mental health conditions, and alcohol- or substance-use disorders). To assess the safety and acceptability of smoking cessation interventions, including serious and non-serious adverse events and treatment discontinuation.
Lien vers le texte intégral (Open Access ou abonnement)
6. Şahin B, Yildiz M, Tonkaz GY. Feasibility and interpretational limits of bright light therapy and exercise for depression prevention in youth with ADHD: methodological and conceptual considerations. Eur Arch Psychiatry Clin Neurosci. 2026.
