Global average pooling as a global context prior is a good baseline model, but for the complex scene image of ultrasound thyroid, this strategy is not enough to cover the necessary information. The pixels in these ultrasound images are labeled for many structures and tissues. If they are fused...
Domain adaptation is an approach to alleviate the above problem, which transfers the labeled source domain to the target domain. It can reduce the data labeling workload in the target domain and significantly improve the network's generalizability. However, the differences between foreground areas ...
In recent years, the poor generalizability of deep neural networks in multi-model medical images has attracted widespread attention. Domain adaptation is an approach to alleviate the above problem, which transfers the labeled source domain to the target domain. It can reduce the data labeling worklo...
Ultrasound examination cannot distinguish a benign from a malignant process. Thyroid scans using radioactive iodine are often performed to evaluate the function of thyroid nodules. The thyroid is the only location in the body that takes up iodine, so when radioactively labeled iodine is given, it...
Deep learning-based CAD system design for thyroid tumor characterization using ultrasound images Niranjan Yadav Rajeshwar Dass Jitendra Virmani Multimedia Tools and Applications (2023) A comparison of artificial intelligence versus radiologists in the diagnosis of thyroid nodules using ultrasonography: a ...
Thyroid Nodule Segmentation and Classification in Ultrasound Images Dataset In the challenge, the participants will be provided the US nodular thyroid dataset supplied by Shanghai Ruijin Hospital. This data comes from the Chinese Medical Ultrasound Artificial Intelligence Alliance (CMUAA) initiated by doct...
Additionally, integrating multidimensional data (i.e., ultrasound images, gene information, blood tests, etc.) can depict the state of the tumor more comprehensively and view the tumor from different perspectives, thus obtaining a more accurate assessment, which of course cannot be achieved without ...
The experiment results on ultrasound images of 1030 patients suggest that the MCDLM model can achieve almost the same classification performance as the fully supervised learning (an accuracy of 90.01% and an AUC of 91.07%) using only about 35% of the full labeled dataset, which saves a lot ...
The technique of ultrasound assessment of lymph node metastasis is becoming more sophisticated, but a certain number of missed diagnoses cannot be avoided, which has a great impact on the diagnosis and treatment of patients20,21. At the same time, PTC has a certain probability of recurrence22....
Ultrasound is useful for delineating the size of the mass, its anatomic relationship to the rest of the thyroid, and the presence of cystic structures. 123I imaging that reveals a single nonfunctioning mass (“cold” nodule) suggests a carcinoma or adenoma and is a clear indication for ...