RT Conference Proceedings
T1 Search for temporal cell segmentation robustness in phase-contrast microscopy videos
A1 Gómez de Mariscal, Estíbaliz
A1 Jayatilaka, Hasini
A1 Çiçek, Özgün
A1 Brox, Thomas
A1 Wirtz, Denis
A1 Muñoz Barrutia, María Arrate
AB This work presents a deep learning-based workflow to segment cancer cells embedded in D collagen matrices and imaged with phase-contrast microscopy under low magnification and strong background noise conditions. Due to the experimental and imaging setup, cell and protrusion appearance change largely from frame to frame. We use transfer learning and recurrent convolutional long-short term memory units to exploit the temporal information and provide temporally stable results. Our results show that the proposed approach is robust to weight initialization and training data sampling.
YR 2022
FD 2022
LK https://hdl.handle.net/10016/37399
UL https://hdl.handle.net/10016/37399
LA eng
NO Proceeding of: Medical Imaging with Deep Learning (MIDL 2022), Zürich, Switzerland, 6-8July 2022
NO This work was co-financed by ERDF, "A way of making Europe" (AMB), partially funded under Grant PID2019-109820RB-I00, MCIN/AEI/10.13039/501100011033/; the US NIH under Grants UO1AG060903 (DW) and U54CA143868 (DW). We acknowledge NVIDIA Corporation for the donation of the Titan X (Pascal) GPU.
DS e-Archivo
RD 20 may. 2024