[행사/세미나] IPHC Seminar Series / 4. 13.
- 지능형정밀헬스케어융합학과
- 조회수628
- 2021-04-15
Neurological Imaging and Microvascular Function
(Prof. Seunghong Choi, Seoul National University Hospital)
Quantitative analysis of brain MRI is routine for many neurological diseases and conditions and relies
on accurate evaluation of structures of interest. Deep learning-based approaches for brain MRI are
gaining interest due to their self-learning and generalization ability over large amounts of data. The
convolutional neural network (CNN) is a class of deep artificial neural networks inspired by biologic
processes of the visual cortex. CNNs respond to inputs through overlapping restricted regions called
receptive fields of different neurons, similar to the cortical neurons responding to external stimuli in the
entire visual cortex through overlapping restricted visual fields of individual neurons. Distinct from
traditional algorithms, CNNs can capture hierarchical contextual information automatically and can
perform a specific task without defining image features. Applications of CNNs to medical imaging
have rapidly increased during the last 2 years. CNNs have been successfully implemented in
radiographs to suppress noise and automatic segmentation. In this presentation, the speaker will
introduce deep learning applications for brain tumors, where their strength and weakness are also
discussed.
References
1. Matsugu M, Mori K, Mitari Y, Kaneda Y. Subject independent facial expression recognition with
robust face detection using a convolutional neural network. Neural Netw 2003;16(5-6):555–559.
2. Greenspan H, van Ginneken B, Summers RM. Deep Learning in Medical Imaging: Overview and
Future Promise of an Exciting New Technique. IEEE Trans Med Imaging. 2016;35(5):1153–1159.
3. Kim KH, Choi SH, Park SH. Improving Arterial Spin Labeling by Using Deep Learning. Radiology.
2018;287(2):658-666.