Zidong Yu

Recent advances in exoskeleton-assisted rehabilitation have highlighted the significance of lower limb movement intention recognition through deep learning. However, discrete motion phase classification and continuous real-time joint kinematics estimation are typically handled as independent tasks, leading to temporal misalignment or delayed assistance during dynamic movements. To address this issue, this study presents iTransformer-DTL, a dual-task learning framework with an improved Transformer designed to identify end-to-end locomotion modes and predict joint trajectories during sit-to-stand transitions. Employing a learnable query mechanism and a non-autoregressive decoding approach, the proposed iTransformer-DTL can produce the complete output sequence at once, without relying on any previously generated elements. The proposed framework has been tested with a dataset of lower limb movements involving seven healthy individuals and seven stroke patients. The experimental results indicate that the proposed framework achieves satisfactory performance in dual tasks. An average angle prediction Mean Absolute Error (MAE) of 3.84° and a classification accuracy of 99.42% were obtained in the healthy group, while 4.62° MAE and 99.01% accuracy were achieved in the stroke group. These results suggest that iTransformer-DTL could support adaptable rehabilitation exoskeleton controllers, enhancing human–robot interactions.

Scoliosis is a common spinal disease and it's early screening is essential for planning treatment and avoiding deterioration. The traditional screening methods for scoliosis have the disadvantages of unnecessary radiation exposure, the dependence on equipment, and the high demand on operators. Although the advent of deep learning techniques provides a new perspective for rapid and convenient screening of scoliosis, the existing related research faces challenges caused by issues such as image background diversity, image size inconsistency, and class imbalance. In order to solve the about problems, a method based on BAEN-SKCNN is proposed for early screening and severity diagnosis of scoliosis using back images. Specifically, BAEN is constructed to extract the back region to improve the diagnostic accuracy and model universality. Spatial pyramid pooling and selective kernel network are used to construct SKCNN for early screening and severity diagnosis of scoliosis. On a self-made scoliosis dataset, the proposed method achieves 98 % accuracy for early screening and 73 % accuracy for severity diagnosis, respectively. Combined with the APP software developed, the proposed method can easily and quickly complete the diagnosis of scoliosis without the limitation of venues, equipment and personnel. It has a certain application prospect in the large-scale screening of scoliosis, and has certain clinical significance for improving the diagnostic rate of scoliosis.