Various disease detection models, based on deep learning algorithms using medical radiograph images (MRI, CT, and X-ray), have been actively explored in relation to medicine and computer vision. For diseases related to the spine, primarily MRI-based or CT-based studies have been conducted, but most studies were associated with the lumbar spine, not the cervical spine. Foraminal stenosis offers important clues in diagnosing cervical radiculopathy, which is usually detected based on MRI data because it is difficult even for experts to diagnose using only an X-ray examination. However, MRI examinations are expensive, placing a potential burden on patients. Therefore, this paper proposes a novel model for diagnosing foraminal stenosis using only X-ray images. In addition, we propose methods suitable for cervical spine X-ray images to improve the performance of the proposed classification model. First, the proposed model adopts data preprocessing and augmentation methods, including Histogram Equalization, Flip, and Spatial Transformer Networks. Second, we apply fine-tuned transfer learning using a pre-trained ResNet50 with cervical spine X-ray images. Compared to the basic ResNet50 model, the proposed method improves the performance of foraminal stenosis diagnosis by approximately 5.3–6.9%, 5.2–6.5%, 5.4–9.2%, and 0.8–4.3% in Accuracy, F1 score, specificity, and sensitivity, respectively. We expect that the proposed model can contribute towards reducing the cost of expensive examinations by detecting foraminal stenosis using X-ray images only.

Kim J, et al. Yonsei Med J. 2024 Apr;65:e16. https://doi.org/10.3349/ymj.2023.0091

The importance of a high-quality dataset availability in 3D human action analysis research cannot be overstated. This paper introduces DGU-HAO (Human Action analysis dataset with daily life Objects). This novel 3D human action multi-modality dataset encompasses four distinct data modalities accompanied by annotation data, including motion capture, RGB video, image, and 3D object modeling data. It features 63 action classes involving interactions with 60 common furniture and electronic devices. Each action class comprises approximately 1,000 motion capture data representing 3D skeleton data and corresponding RGB video and 3D object modeling data, resulting in 67,505 motion capture data samples. It offers comprehensive 3D structural information of the human, RGB images and videos, and point cloud data for 60 objects, collected through the participation of 126 subjects to ensure inclusivity and account for diverse human body types. To validate our dataset, we leveraged MMNet, a 3D human action recognition model, achieving Top-1 accuracy of 91.51% and 92.29% using the skeleton joint and bone methods, respectively. Beyond human action recognition, our versatile dataset is valuable for various 3D human action analysis research endeavors.

Constructing diverse and complex multi-modal datasets is crucial for advancing human action analysis research, providing ground truth annotations for training deep learning networks, and enabling the development of robust models across real-world scenarios. Generating natural and contextually appropriate nonverbal gestures is essential for enhancing immersive and effective human–computer interactions in various applications. These applications include video games, embodied virtual assistants, and conversations within a metaverse. However, existing speech-related human datasets are focused on style transfer, so they have limitations that make them unsuitable for 3D human action analysis studies, such as human action recognition and generation. Therefore, we introduce a novel multi-modal dataset, DGU-HAU, a dataset for 3D human action on utterances that commonly occurs during daily life. We validate the dataset using a human action generation model, Action2Motion (A2M), a state-of-the-art 3D human action generation model.