Deep learning based Multiple Sclerosis lesion detection utilizing synthetic data generation and soft attention mechanism
In person: 22 February 2022 • 4:30 PM - 4:50 PM PST
In this work, we suggest a new network architecture, based on Y-net and EfficientNet models, with attention layers to improve the network performance and reduce overfitting. Furthermore, the attention layers allow us to extract lesion locations. In addition, we show an innovative regularization scheme on the attention weight mask to make it focus on the lesions while letting it search in different areas. Finally, we explore an option to add synthetic lesions in the training process. Based on recent work, we generate artificial lesions in healthy brain MRI scans to augment our training data. Our system achieves 90% accuracy in identifying cases that contain lesions (vs. healthy) with more than 12% improvement over an equivalent system without the attention and the data added.
Tel Aviv Univ. (Israel)
Omer Shmueli has a B.Sc in Electrical Engineer from Ben Gurion university. Currently, He is working on his masters degree, specifies in the field of Electrical and Electronics engineering, under prof. Hayit Greenspan, at Tel Aviv University, At the moment, He works as computer vision and Machine Learning Researcher at Juganu. He previously worked as computer vision and deep learning researcher at Deep AI Technologies,2Sens and Xsight Systems.