• Citer

Understanding the Pathophysiology of Mental Diseases and Early Diagnosis Thanks to Electrophysiological Tools: Some Insights and Empirical Facts

par Sumiyoshi, Tomiki;Campanella, Salvatore ;Giordano, Giulia Maria;Ishii, Ryouhei;Pogarell, Oliver
Référence Clinical EEG and neuroscience
Publication Publié, 2024-09-01

Article révisé par les pairs

• Accès en ligne
• Détails
• Contenu
• Statistiques

Résumé :

Objective. Neurophysiological tools remain indispensable instruments in the assessment of psychiatric disorders. These techniques are widely available, inexpensive and well tolerated, providing access to the assessment of brain functional alterations. In the clinical psychiatric context, electrophysiological techniques are required to provide important information on brain function. While there is an immediate benefit in the clinical application of these techniques in the daily routine (emergency assessments, exclusion of organic brain alterations), these tools are also useful in monitoring the progress of psychiatric disorders or the effects of therapy. There is increasing evidence and convincing literature to confirm that electroencephalography and related techniques can contribute to the diagnostic workup, to the identification of subgroups of disease categories, to the assessment of long-term causes and to facilitate response predictions. Methods and Results. In this report we focus on 3 different novel developments of the use of neurophysiological techniques in 3 highly prevalent psychiatric disorders: (1) the value of EEG recordings and machine learning analyses (deep learning) in order to improve the diagnosis of dementia subtypes; (2) the use of mismatch negativity in the early diagnosis of schizophrenia; and (3) the monitoring of addiction and the prevention of relapse using cognitive event-related potentials. Empirical evidence was presented. Conclusion. Such information emphasized the important role of neurophysiological tools in the identification of useful biological markers leading to a more efficient care management. The potential of the implementation of machine learning approaches together with the conduction of large cross-sectional and longitudinal studies was also discussed.