Wernigg Róbert, Hajduska-Dér Bálint
Országos Kórházi Fôigazgatóság Alapellátási és Fejlesztési Fôosztály, Budapest, Hungary, E-mail:
Psychiatr Hung. 2024;39(1):24-35.
In recent decades a global problem in mental health has been the increase in the relative proportion of patients who do not receive care, which is associated with loss of life years and deterioration in quality of life. The practical application of artificial intelligence (AI) can help in the fields of data analysis, diagnosis, therapy planning, among others in psychiatric care, thus reducing the human resource input. Today's artificial narrow intelligence (ANI), also known as weak AI, can recognise patterns and correlations in large data sets with the help of machine learning procedures and to make autonomous decisions while making its own refinements. The use of AI-based systems may be effective in the classification of mental health disorders, in disease prevention, in clinical diagnosis and treatment without human input, and finally, it can play a supporting role in many areas of data analysis (quality care assessment, research). A key area of diagnostics is the estimation of suicidal risk and the assessment of mood status using machine learning, which can be used to make predictions with high accuracy, by analysing written text or speech. By examining correlations within large data sets, advances in precision medicine could also be made, allowing more accurate prediction of medication. Psychotherapeutic programs using artificial intelligence are already available today, which can provide users with easily accessible help, mainly using cognitive therapy tools. In addition to its obvious benefits, the use of artificial intelligence also raises ethical and methodological questions, making its regulation a key issue for the future.
近几十年来，心理健康领域的一个全球性问题是未接受治疗的患者相对比例增加，这与生命年数的损失和生活质量的下降有关。人工智能（AI）的实际应用可以在数据分析、诊断、治疗规划等精神科护理领域提供帮助，从而减少人力资源投入。当今的人工狭义智能（ANI），也称为弱人工智能，可以借助机器学习程序识别大数据集中的模式和相关性，并在自我完善的同时做出自主决策。基于人工智能的系统在心理健康障碍的分类、疾病预防、无需人工干预的临床诊断和治疗中可能会很有效，最终，它可以在数据分析的许多领域（优质护理评估、研究）发挥辅助作用。诊断的一个关键领域是使用机器学习估计自杀风险和评估情绪状态，通过分析书面文本或语音，机器学习可用于进行高精度预测。通过检查大数据集内的相关性，也可以在精准医学方面取得进展，从而更准确地预测药物治疗效果。如今已经有使用人工智能的心理治疗程序，主要使用认知治疗工具，可为用户提供易于获取的帮助。除了其明显的好处外，人工智能的使用也引发了伦理和方法问题，使其监管成为未来的一个关键问题。
