Given the high alignment between deep learning and blended teaching objectives, blended teaching provides a feasible pathway for achieving deep learning. This study aimed to verify the efficacy of a deep learning-based blended teaching model integrating online immersive learning, offline scenario simulation and clinical case review in improving ECMO transport competency among ICU nurses. This was a single-center prospective parallel-group randomized controlled trial. A total of 130 ICU nurses were randomly assigned to experimental group (n = 68) and control group (n = 62). The experimental group received the blended teaching intervention, while the control group adopted traditional offline lecture teaching. Baseline questionnaires, learning style and motivation scales were collected before intervention. Eye-tracking technology and engagement scales were used to evaluate learning immersion during training. Post-training knowledge level and satisfaction were assessed, and paired t-tests were adopted for within-group comparison. Baseline data were comparable between the two groups. The experimental group showed significantly better eye-tracking immersion indicators and higher online learning participation than the control group (all P < 0.001). Post-training knowledge scores were significantly improved in the experimental group, while no obvious change was found in the control group. The intervention presented a small positive effect size. No significant between-group differences were observed in learning engagement, cognitive participation and satisfaction. Compared with traditional offline teaching, the deep learning-oriented blended teaching model can effectively improve ICU nurses’ ECMO transport knowledge mastery, learning immersion and online learning participation. It makes up for the competency deficiency of ICU nurses in ECMO transport and provides evidence for optimizing clinical nursing training strategies. Chinese Clinical Trial Registry, Registration Number: ChiCTR2500115168, Public Access Link https://www.chictr.org.cn/showproj.html?proj=292586. The registration date is December 23, 2025. • High incidence of adverse events during ECMO transport highlights urgent training needs due to insufficient transfer competency among ICU nurses. • Traditional offline lecture-based approaches suffer from inefficient knowledge delivery and inadequate skill transformation, failing to meet clinical demands. • While blended teaching shows promise in clinical education, existing research exhibits limitations including disconnection between theory and technology application, singular evaluation systems, and lack of hierarchical design considerations. • Constructed and validated a deep learning-oriented blended teaching model, specifying implementation standards for each component to provide a replicable ECMO transport training program. • Established a multidimensional evaluation system integrating process, outcome, and value-added assessments by combining eye-tracking technology with diverse scales, addressing limitations of traditional evaluation methods. • Demonstrated that this model significantly improves nurses’ ECMO transport knowledge, learning immersion, and online engagement, confirming its clinical applicability. • Provided data-driven support for personalized training optimization, enabling tailored solutions that meet varying competency levels among nurses.
Zhang et al. (Tue,) studied this question.