Publications by authors named "Nhu-Tai Do"
Article Synopsis
- Recent studies on tumor classification and segmentation have focused mostly on brain, lung, and liver cancers, leaving a gap in research related to knee bone tumors using deep learning.
- The study introduces the Seg-Unet model, employing both global and patch-based methods to tackle challenges like small lesion size and variability in knee bone tumors, aiming to assist physicians in identifying normal, benign, or malignant regions.
- Results from experiments on a dataset created in collaboration with Chonnam National University Hospital indicate the proposed method achieved high accuracy (99.05%) in classification and a significant Mean IoU (84.84%) for segmentation, highlighting its potential usefulness for medical professionals.
Emotion recognition plays an important role in human-computer interactions. Recent studies have focused on video emotion recognition in the wild and have run into difficulties related to occlusion, illumination, complex behavior over time, and auditory cues. State-of-the-art methods use multiple modalities, such as frame-level, spatiotemporal, and audio approaches.
View Article and Find Full Text PDF