On March 21 and 22, 2026, students Dang Tri Hieu and Hoang Hiep – majoring in Informatics and Computer Engineering, Dao Duy Phat and Tran Quang Tiep – majoring in Applied Information Technology, from the International School, Vietnam National University, Hanoi, participated in presentations at the 6th International Conference on Intelligent Systems and Networks (ISN 2026).
The ISN 2026 Conference is a prestigious forum bringing together experts and scientists from home and abroad to discuss the latest technology trends. The direct participation of International School students in presenting at this forum is clear evidence of their research capabilities, academic spirit, and international integration.
Students Dang Tri Hieu and Dao Duy Phat confidently participate in the ISN 2026 International Conference.
During the specialized discussion sessions, representatives from student groups presented research works with high practical value and applicability.
“
Localization and Detection of Medical Abnormalities using YOLOv12 and Detectron2
Focuses on applying the most advanced deep learning models currently available to detect and localize medical abnormalities. The findings of this paper significantly contribute to improving the efficiency of clinical diagnostic support systems.
The paper “Hybrid NAS and Knowledge Distillation for CNN Optimization” proposes a two-branch CNN model that combines features from the spatial domain (original image) and the frequency domain (Fourier transform) to classify cervical cancer cells. The authors used the APACC dataset with over 41,000 images, applying preprocessing techniques such as hybrid CLAHE-PMD filter and Fourier transform to extract features.
Experimental results show that the feature fusion method, especially concatenation, yields superior performance compared to single-domain models, achieving an accuracy of 86.03%. This research confirms that integrating frequency domain information improves the stability and discriminative capability for cytological images, paving a new direction for more reliable automated Pap smear screening.
“
U-Mamba: Lightweight Ultrasound Segmentation Using Selective State Space Models
Presents a novel approach that optimizes the ultrasound image segmentation process through selective state space models, holding great potential for medical applications.
School students receive certificates from the Organizing Committee.
The students’ research projects are the result of their continuous efforts, coupled with the dedicated support and guidance from experienced scientific advisors, such as Dr. Le Xuan Hai, Dr. Kim Dinh Thai, and Dr. Pham Thi Viet Huong.
The ISN 2026 Conference is a prestigious forum bringing together experts and scientists from home and abroad.
The participation of School students in presenting at the international conference is evidence of their research capabilities, academic spirit, and international integration. At the same time, it is also the result of the School’s training orientation linked with scientific research and innovation, along with the intensive training environment in research labs.
