Electrocardiogram (ECG) signal is an important basis for the diagnosis of arrhythmia and myocardial infarction. In order to further improve the classification effect of arrhythmia and myocardial infarction, an ECG classification algorithm based on Convolutional vision Transformer (CvT) and multimodal image fusion was proposed. Through Gramian summation angular field (GASF), Gramian difference angular field (GADF) and recurrence plot (RP), the one-dimensional ECG signal was converted into three different modes of two-dimensional images, and fused into a multimodal fusion image containing more features. The CvT-13 model could take into account local and global information when processing the fused image, thus effectively improving the classification performance. On the MIT-BIH arrhythmia dataset and the PTB myocardial infarction dataset, the algorithm achieved a combined accuracy of 99.9% for the classification of five arrhythmias and 99.8% for the classification of myocardial infarction. The experiments show that the high-precision computer-assisted intelligent classification method is superior and can effectively improve the diagnostic efficiency of arrhythmia as well as myocardial infarction and other cardiac diseases.
Objective To explore the educational effectiveness of multimodal image fusion combined with three-dimensional (3D) reconstruction technology in the standardized training of orthopedic residents in bone and soft tissue tumors. Methods Orthopedic residents who participated in the bone and soft tissue tumor training program at orthopedic center of West China Hospital of Sichuan University between September 2023 and September 2024 were selected. The orthopedic residents were divided into an experimental group and a control group using a random number table method. The experimental group received multimodal image fusion combined with 3D reconstruction teaching, whereas the control group received conventional imaging-based teaching. Pre-class and post-class scores across four sessions, final assessment scores at 3 weeks after course completion, and questionnaire scores were compared between groups. Gain scores were calculated as post-class scores minus pre-class scores. Analysis of covariance was performed separately for each session, with the corresponding pre-class score included as a covariate. Results A total of 60 residents were included, with 30 people in each group. No significant between-group differences were observed in pre-class scores for Sessions 1 and 2 (P>0.05), whereas the experimental group had significantly higher pre-class scores than the control group in Sessions 3 and 4 (P<0.05). In both groups, post-class scores increased significantly after all four sessions (P<0.001), and the gain scores were significantly greater in the experimental group than in the control group across all sessions (P<0.01). Analysis of covariance showed no significant interactions between group and the corresponding pre-class score. After adjustment for pre-class scores, the experimental group continued to have significantly higher post-class scores in all four sessions, with adjusted mean differences ranging from 8.91 to 9.91 points (P<0.001). At 3 weeks after course completion, the final assessment score was significantly higher in the experimental group than in the control group (89.70±4.30 vs. 72.10±5.60, P<0.001). The experimental group also scored significantly higher in satisfaction with the teaching method, learning interest, evaluation of teaching content, and perceived knowledge mastery (P<0.001), whereas no significant between-group difference was found in perceived self-improvement (P>0.05). Conclusions Multimodal image fusion combined with 3D reconstruction may improve residents’ understanding of complex anatomical relationships and surgical planning, as well as immediate learning outcomes, short-term knowledge retention, and learning experience. It may serve as an adjunctive approach in residency training for bone and soft tissue tumors; however, its educational effectiveness requires further validation in larger, multicenter studies.