Background/Objectives: The COVID-19 pandemic has led to increased mental health issues, particularly among long-COVID patients, who experience persistent symptoms post-recovery, potentially leading to chronic conditions. The psychological impact of long-COVID is still largely unknown, but it may contribute to mental disorders like Post-Traumatic Stress Disorder (PTSD). Given the global rise in anxiety and depression, exploring therapies like Eye Movement Desensitization and Reprocessing (EMDR) for long-COVID traumatic disorders is crucial. This study explores the effectiveness of remote EMDR therapy for PTSD-like symptoms in long-COVID conditions (LCC), assessing their emergence, the impact of LCC on mental health, and identifying key commonalities. It also examines the potential advantages of an artificial intelligence (AI)-powered platform for EMDR treatments for both therapists and patients, evaluating the response differences between remote and in-person treatment. Methods: We enrolled a total of 160 participants divided into two groups of 80, with the experimental group receiving EMDR treatment for PTSD-like symptoms via a remote AI-powered platform, and the control group receiving traditional in-person therapy. We compared the ANOVA for Subjective Units of Disturbance (SUDs) scores, PTSD Checklist for DSM-5 (PCL-5) scores, and Impact of Event Scale-Revised (IES-R) scores between our two groups for three cases: pre-treatment, post-treatment, and decrement. Results: Statistical significance analysis showed a consistent absence of significant differences between online AI-powered platforms and traditional in-presence sessions. This effectively confirms our hypothesis and highlights that no significant differences were observed between the two groups. Conclusions: The AI-supported remote platform demonstrates comparable efficacy in delivering EMDR therapy, confirming its potential as an effective alternative to traditional in-person methods while providing added advantages in accessibility and adaptability (e.g., remote areas, hikikomori, natural disasters).

Despite the growing interest in physiological stress monitoring, an objective measure of stress is currently lacking, especially in clinical and rehabilitation contexts. With the emerging integration of artificial intelligence (AI) in data analytics, heart rate variability (HRV) has gained attention as an effective biomarker; however, the literature remains fragmented across disciplines, stress types, and methodological approaches. This scoping review aims to investigate how AI techniques are applied to HRV analysis for stress detection and prediction in adult populations. Although this review does not focus on a specific subtype of stress, its primary objective is to explore the current methodological state of the art as reported in the literature, without restrictions on stress typology. Following PRISMA-ScR guidelines, a systematic search was conducted across PubMed, Scopus, and Google Scholar for studies published between 2005 and 2025, using MeSH terms including “HRV”, “Rehabilitation”, “SCI” (for Spinal Cord Injury), “Stress”, “Sympathetic”, “Parasympathetic”, “Non-linear”, “Gamification”, “AI” and “Machine Learning”. Inclusion criteria targeted adult human populations and studies employing HRV features as input for AI and machine learning techniques for psychophysical stress assessment. Of the 566 records identified, 15 studies met the eligibility criteria. The reviewed studies exhibit substantial heterogeneity in terms of settings, populations, sensors, and algorithms with most employing supervised methods (e.g., random forest, support vector machine), alongside several applications of deep learning and explainable AI. Only one study focused specifically on physiological stress, none focused on SCI populations, and rehabilitation-related research was scarce, thereby underscoring important gaps in the current literature. Overall, HR variability analysis, especially when combined with artificial intelligence techniques, represents a promising approach for stress assessment; however, the field is methodologically fragmented and clinically underdeveloped in critical areas, underscoring the need for a multidisciplinary methodological framework.

For this work, a preliminary study proposed virtual interfaces for remote psychotherapy and psychology practices. This study aimed to verify the efficacy of such approaches in obtaining results comparable to in-presence psychotherapy, when the therapist is physically present in the room. In particular, we implemented several joint machine-learning techniques for distance detection, camera calibration and eye tracking, assembled to create a full virtual environment for the execution of a psychological protocol for a self-induced mindfulness meditative state. Notably, such a protocol is also applicable for the desensitization phase of EMDR therapy. This preliminary study has proven that, compared to a simple control task, such as filling in a questionnaire, the application of the mindfulness protocol in a fully virtual setting greatly improves concentration and lowers stress for the subjects it has been tested on, therefore proving the efficacy of a remote approach when compared to an in-presence one. This opens up the possibility of deepening the study, to create a fully working interface which will be applicable in various on-field applications of psychotherapy where the presence of the therapist cannot be always guaranteed.

Cybersecurity is a field in which integration of artificial intelligence (AI) represents a significant direction towards protection against cyber threats. This scoping review explores the current impact and future prospects of AI in four key areas of cybersecurity: threat detection, endpoint security, phishing and fraud detection, and network security. The main goal was to answer the research question, ‘Is AI an effective method to enhance current infrastructures’ cybersecurity?’ Method: Through the PRISMA-ScR protocol, 2548 records were identified from the Google Scholar database from January 2020 to April 2025. The following search terms were used to identify available literature: “Artificial Intelligence Cybersecurity”, “Machine Learning Cybersecurity”, “Cybersecurity Innovation AI”, “AI Future Perspective Cybersecurity”, “Machine Learning Innovation Cybersecurity”. The search only included articles in English. No grey literature has been included. Articles with a focus on performance optimization, cost analysis and business models without a focus on privacy and security have been discarded. Results: The impact and performance of AI algorithms have been highlighted through a selection of 20 articles. Both Machine Learning and Neural Network methods have been employed in the literature, with Decision Trees and Random Forest being the most common approaches. Discussion: The main common limitations of the analyzed articles have been discussed, highlighting possible future directions of research to tackle them. Conclusions: Despite the evidenced limitations, AI showed promising results in improving cybersecurity, especially concerning cyber attack detection and classification, with methods able to grant very high accuracy and trustworthiness.

Background: Early diagnosis of Invasive Pulmonary Aspergillosis (IPA) in non-neutropenic ICU patients remains challenging due to the low specificity of radiological findings and the limited reliability of individual biomarkers. This diagnostic uncertainty often delays antifungal treatment and contributes to the high mortality associated with IPA. We aimed to develop a clinically interpretable model for early bedside risk stratification. Methods: We conducted a retrospective study including 298 ICU patients. A Logistic LASSO regression approach was used to identify the most informative predictors of IPA and to develop a parsimonious predictive model. Model performance was assessed through internal exploration and robustness testing under noise perturbation. Results: Three routinely available clinical variables were retained in the final model: pulmonary galactomannan, non-specific pulmonary infiltrates, and patient age, reflecting fungal burden, lung involvement, and host vulnerability, respectively, and identifying patients with a higher probability of culture-positive Aspergillus detection within a clinical context compatible with IPA. The final model showed promising discriminative performance, with an AUC of 0.836 ± 0.075, sensitivity of 0.72, and specificity of 0.74. Despite relying on only three routinely available clinical variables, the model maintained stable performance during internal exploration and robustness analyses. Conclusions: Within this cohort, the combination of positive pulmonary galactomannan, radiological infiltrates, and advanced age identifies a subgroup of ICU patients with a higher probability of microbiological positivity and clinical suspicion of IPA, supporting early risk assessment rather than definitive diagnosis. This simple and transparent predictive model may support early clinical evaluation and risk stratification of ICU patients and assist clinicians in identifying patients requiring further diagnostic assessment.