The relevance of implementing intelligent computing platforms in psychiatry and clinical psychology is dictated by the urgent need to improve existing diagnostic and therapeutic strategies, especially in the context of the increasing negative impact of military conflicts on the mental health of the population (for example, as noted in studies by the World Health Organization [WHO, 2023]). This study aims to develop and verify innovative approaches to overcoming psychological trauma based on the use of artificial intelligence (AI) technologies to identify subclinical symptoms and objectively assess mental status. The methodological basis of the study is a multifaceted analysis of big data, including the results of psychological surveys (e.g., the PCL-5 Post-Traumatic Stress Disorder Rating Scale [25]), neuroimaging data (fMRI, EEG), biometric parameters (heart rate variability, galvanic skin response), and the use of specialized neural networks and machine learning algorithms to identify patterns and predictive models. The results obtained during the research conducted by the Institute of Artificial Intelligence Problems (IAIP) demonstrated that the developed machine learning models, based on the analysis of linguistic markers of speech of participants in military conflicts, achieve 85% accuracy in identifying signs of post-traumatic stress disorder (PTSD). The use of deep neural network architectures (e.g., recurrent neural networks with long short-term memory, LSTM [Hochreiter & Schmidhuber, 1997]) allowed us to identify specific patterns of speech activity that statistically significantly correlate with anxiety and depression levels. The scientific novelty of this approach lies in the integration of fundamental psychological principles with advanced intelligent systems equipped with neural interfaces and physiological state monitoring sensors to optimize self-control processes and provide personalized therapeutic recommendations. The Institute of Artificial Intelligence Problems (IAIP) conducts in-depth research into the development of predictive models of mental health aimed at the timely detection of depressive and anxiety disorders, which is consistent with the concept of predictive psychiatry [Insel, 2009]. The findings confirm the potential of using intelligent systems for automated assessment and treatment of the mental state of individuals affected by military conflicts, which requires an interdisciplinary approach integrating advances in AI, neuroscience, and clinical psychology. Priority areas include the development of non-invasive neural interfaces for personalized biofeedback and the creation of complex machine learning models integrating various data sources. Particular attention must be paid to adherence to strict ethical standards and procedures to ensure the confidentiality and security of patients' personal data, in accordance with the GDPR and other regulatory requirements.
The relevance of implementing intelligent computing platforms in psychiatry and clinical psychology is dictated by the urgent need to improve existing diagnostic and therapeutic strategies, especially in the context of the increasing negative impact of military conflicts on the mental health of the population (for example, as noted in studies by the World Health Organization [WHO, 2023]). This study aims to develop and verify innovative approaches to overcoming psychological trauma based on the use of artificial intelligence (AI) technologies to identify subclinical symptoms and objectively assess mental status. The methodological basis of the study is a multifaceted analysis of big data, including the results of psychological surveys (e.g., the PCL-5 Post-Traumatic Stress Disorder Rating Scale [25]), neuroimaging data (fMRI, EEG), biometric parameters (heart rate variability, galvanic skin response), and the use of specialized neural networks and machine learning algorithms to identify patterns and predictive models. The results obtained during the research conducted by the Institute of Artificial Intelligence Problems (IAIP) demonstrated that the developed machine learning models, based on the analysis of linguistic markers of speech of participants in military conflicts, achieve 85% accuracy in identifying signs of post-traumatic stress disorder (PTSD). The use of deep neural network architectures (e.g., recurrent neural networks with long short-term memory, LSTM [Hochreiter & Schmidhuber, 1997]) allowed us to identify specific patterns of speech activity that statistically significantly correlate with anxiety and depression levels. The scientific novelty of this approach lies in the integration of fundamental psychological principles with advanced intelligent systems equipped with neural interfaces and physiological state monitoring sensors to optimize self-control processes and provide personalized therapeutic recommendations. The Institute of Artificial Intelligence Problems (IAIP) conducts in-depth research into the development of predictive models of mental health aimed at the timely detection of depressive and anxiety disorders, which is consistent with the concept of predictive psychiatry [Insel, 2009]. The findings confirm the potential of using intelligent systems for automated assessment and treatment of the mental state of individuals affected by military conflicts, which requires an interdisciplinary approach integrating advances in AI, neuroscience, and clinical psychology. Priority areas include the development of non-invasive neural interfaces for personalized biofeedback and the creation of complex machine learning models integrating various data sources. Particular attention must be paid to adherence to strict ethical standards and procedures to ensure the confidentiality and security of patients' personal data, in accordance with the GDPR and other regulatory requirements.