This is a page that curates AI-related papers published worldwide. All content here is summarized using Google Gemini and operated on a non-profit basis. Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.
A dataset of primary nasopharyngeal carcinoma MRI with multi-modalities segmentation
Created by
Haebom
Author
Yin Li, Qi Chen, Kai Wang, Meige Li, Liping Si, Yingwei Guo, Yu Xiong, Qixing Wang, Yang Qin, Ling Xu, Patrick van der Smagt, Jun Tang, Nutan Chen
Outline
This paper presents the first comprehensive multimodal MRI dataset to improve the diagnosis and treatment planning of nasopharyngeal cancer (NPC). It includes T1-weighted, T2-weighted, and contrast-enhanced T1-weighted images (831 scans) from 277 primary NPC patients, along with high-quality segmented data manually annotated and labeled by an experienced radiologist. This dataset is expected to significantly contribute to the development of machine learning algorithms and clinical research related to NPC.
Takeaways, Limitations
•
Takeaways:
◦
Contributing to the advancement of research related to NPC diagnosis and treatment by releasing the first comprehensive NPC MRI dataset.
◦
Providing high-quality, manually segmented data, providing a useful resource for developing machine learning algorithms.
◦
Multi-faceted analysis possible by including various MRI sequences (T1, T2, contrast-enhanced T1)
◦
Comprehensive analysis and research possible with clinical data included
•
Limitations:
◦
The size of the dataset (277 patients) may not be sufficient for larger-scale studies.
◦
Only patients with primary NPC were included, which may limit studies of metastatic or recurrent NPC.
◦
Additional information is needed to determine whether the diversity of the dataset (age, gender, stage of disease, etc.) has been sufficiently considered.
◦
The generalization performance of the dataset may be degraded due to differences in images taken at different hospitals.
