Conformity Certifications
MFDS
Republic of Korea
01
02
03
Validations
Brain hemorrhage diagnostic function has been proven to deliver outstanding performance.
0
sensitivity
0
specificity
0
Accuracy
0
AUROC
The brain hemorrhage diagnostic assistance software, aview NeuroCAD,
has been extensively tested across a range of conditions involving similar patient scenarios,
resulting in retrospective clinical evidence (Clinical Trials for the Certificate of Conformity).
Publications
Refer to the following research papers
These contents represent summaries of scientific
publications and are unrelated to any form of advertising.
Recent advancements in deep learning have led to the proposal of SMART-Net, a supervised multi-task aiding representation transfer learning network, for computer-aided diagnosis (CAD) tasks involving the classification and segmentation of intracranial hemorrhage (ICH) using non-contrast head computed tomography (NCCT).
To prove our SMART-Net framework in real-emergency medical situations, the SMART-Net was evaluated with one internal and three external test sets, including an open-source dataset of ICH, PhysioNet.
The volume-level classification and segmentation results indicated that SMART-Net provides the best balance between sensitivity and specificity and also robust performance to even external datasets.
The deep learning algorithms have been commercialized as part of the aview NeuroCAD system by Coreline Soft, showing potential for improved emergency medical care.
Kyung S, Shin K, Jeong H, Kim KD, Park J, Cho K, Lee JH, Hong G, Kim N. “Improved performance and robustness of multi-task representation learning with consistency loss between pretexts for intracranial hemorrhage identification in head CT.” Med Image Anal. 2022 Oct;81:102489.
