Intensive care unit (ICU) psychosis, commonly recognised as ICU delirium, is a frequent neuropsychiatric complication among critically ill patients and is associated with prolonged hospitalisation, increased morbidity, and higher mortality. Nurses play a pivotal role in its early identification and assessment. However, inadequate knowledge and organisational barriers may hinder effective practice. This study aimed to assess ICU nurses' perception regarding ICU psychosis, identify the personal and professional barriers influencing its assessment, and examine the association between selected demographic variables and both perception and perceived barriers. In this quantitative descriptive cross-sectional study conducted among 100 ICU nurses working in selected hospitals of Anand district (Gujarat), participants were selected using purposive sampling. Data were collected using a structured questionnaire, a 25-item perception scale, and a 15-item barriers scale. Data were analysed using descriptive statistics and Chi-square tests. Overall, 70 percent of nurses demonstrated adequate perception of ICU psychosis, while 30 percent had partially adequate perception. The mean perception score was 87.92 ± 6.89, indicating an overall satisfactory level of perception. Personal and professional barriers were reported at comparable levels, with inadequate staffing, heavy workload, limited training, and the absence of standardised assessment protocols identified as the principal barriers. Social media was the predominant source of prior knowledge. Gender was significantly associated with both perception and perceived barriers (p <0.05), whereas no significant associations were observed with other demographic variables. Although ICU nurses demonstrated generally adequate perception of ICU psychosis, persistent educational and organisational barriers may impede routine assessment. Strengthening continuing nursing education, implementing validated delirium assessment tools, and establishing institutional protocols are essential to improve early recognition and optimise patient outcomes in intensive care settings.