Subtle sarcasm and contextual sarcasm are particularly important topics of research due to their nature as linguistic phenomena. Indeed, sarcasm detection is a complex NLP task because sarcasm conveys meaning beyond the literal one and depends heavily on context. Thus, it requires considering discourse-level features and long-term relations between statements. Machine learning and deep learning models often perform poorly in recognising sarcasm because they cannot effectively incorporate contextual cues and polarity shifts. To address this problem, this paper introduces an improved sarcasm detection framework based on the Reformer transformer model to improve efficiency in modelling long sequences and extracting semantic features. We evaluated the proposed model on benchmark datasets (news, social media texts, conversations), using accuracy, precision, recall, F1-score, and AUC-ROC as performance indicators. The results showed that the recommended model outperformed several other available frameworks (BiLSTM, BERT, RoBERTa, DistilBERT).
Compared to traditional machine learning systems, the human visual system shows very high robustness against noise, occlusion and variation of the environment. This paper presents a brain-inspired sparse population coding framework that combines a neural manifold, mixed selectivity, and FPGA-based adaptive encoding for robust object recognition under sensory uncertainty. The model uses sparse distributed representations to improve feature discrimination and reduce redundancy and computational cost. The experiments on common vision datasets (n=200 samples) confirm that the model can achieve higher accuracy, about 3.1%-4.5% higher than traditional dense coding, in the face of noise and occlusion. FPGA implementation reduces processing latency by 18%-25%, enabling efficient real-time inference; paired t-tests confirm these performance gains (p<0.05). Additionally, the model shows greater robustness to adversarial attacks and illumination variations. These results indicate that the biologically-motivated sparse encoding scheme provides an effective and reliable visual recognition system. The proposed framework is well suited for edge robots, autonomous navigation, and surveillance systems where stable, efficient perception is critical.
Out-of-hospital cardiac arrest (OHCA) often results in hypoxic–ischemic brain injury with persistent neurocognitive sequelae that may not be fully captured by routine neuroimaging. We report a detailed neuropsychological and language profile of a 32-year-old Greek-speaking Cypriot male assessed six months after OHCA. The case was neurologically complex, as the patient had a history of right frontal traumatic brain injury, chronic right frontal gliosis, post-traumatic epilepsy, and long-term antiepileptic treatment prior to the OHCA. A standardized, culturally appropriate battery with qualitative analysis assessed attention, processing speed, executive functions, memory, language, and visuospatial abilities, alongside mood and quality of life. Findings showed a heterogeneous pattern, including severe impairments in verbal episodic memory and moderate-to-severe executive dysfunction, mild attentional inefficiency, and relatively preserved visuospatial skills and nonverbal memory. Language performance was characterized by impaired lexical access and discourse/narrative organization, consistent with executive–linguistic interaction. The profile was interpreted cautiously as potentially reflecting interacting contributions from post-arrest hypoxic-ischemic vulnerability, pre-existing frontal pathology, epilepsy-related factors, and medication effects. Patient reported outcomes indicated preserved emotional well-being and satisfactory quality of life. This case underscores the value of early cognitive-linguistic and psychosocial screening following OHCA and supports multidisciplinary follow-up assessment to identify impairments affecting communication, activity, and participation. To our knowledge, this is the first detailed post-OHCA neuropsychological characterization of a Greek speaking Cypriot patient, highlighting the need for culturally and linguistically tailored assessment. This contributes data to an underrepresented population and suggests priority targets for intervention.
Preterm infants admitted to neonatal intensive care units are exposed to environments that differ substantially from the intrauterine setting, with limited opportunities for positive sensory input. This randomised controlled trial examined whether a physiotherapy-led sensory–motor stimulation program could improve early growth outcomes among clinically stable preterm infants in a Malaysian NICU. Eighty-four infants born between 30 and 34 weeks’ gestation (birth weight 1.3–1.8 kg) were randomly assigned to receive either a combined massage therapy and kinaesthetic stimulation protocol (MT + KS) or kinaesthetic stimulation alone (KS). Mothers were trained and supervised by physiotherapists to deliver three daily sessions until discharge at a target weight of 1.8 kg. Average daily weight gain (ADWG) was the primary outcome, while study weight gain (StWG) and length of hospital stay (LOS) were secondary outcomes. Infants in the MT + KS group demonstrated higher ADWG compared with the KS-only group (mean difference +7.7 g/day, p=0.024), exceeding the predefined minimal clinically important difference of 5 g/day. Differences in StWG and LOS did not reach statistical significance, although both outcomes favoured the intervention group. No adverse events were observed, and caregiver adherence to the protocol was high. In this setting, a physiotherapy-supervised, mother-assisted MT + KS program was safe, feasible, and associated with improved growth velocity. The findings support incorporating structured sensory–motor stimulation into routine NICU care, particularly in resource-limited contexts.
Major depressive disorder (MDD) is a significant global health issue. Adolescence, in particular, represents a critical period marked by peak onset, heightened vulnerability, and poorer response to treatment, highlighting the urgent need for improved diagnostic and therapeutic strategies in this population. Accumulating evidence has proposed miR-134 as a potential molecular biomarker for the diagnosis of major depressive disorder (MDD). However, as with most research on miRNAs in depression, these investigations have been limited to adult populations, leaving the adolescent group underexplored. To extend this work, the present study examined circulating miR-134 expression in adolescents with MDD. A total of forty-two participants, including twenty adolescents with MDD and twenty-two healthy adolescents (HC), were recruited for plasma sampling, RNA extraction, and real-time RT-PCR to quantify the expression of circulating miR-134. Despite the relatively small sample size, circulating plasma miR-134 was significantly downregulated in adolescents with MDD compared with HC. Receiver operating characteristic (ROC) curve analysis showed that miR-134 had a moderate diagnostic potential for MDD (area under the curve 0.75). Plasma miR-134 distinguished MDD from HC with 70% sensitivity and 72,7% specificity. Our findings confirm decreased plasma miR-134 levels in adolescents with MDD, consistent with previous observations in adults. This cross-age consistency reinforces the potential of miR-134 as a reliable peripheral biomarker for major depressive disorder, relevant to both adolescent- and adult-onset depression.
