Deep learning segmentation projects of FIB-SEM dataset of a U2-OS cell
Автор: Ilya Belevich
Загружено: 2024-02-16
Просмотров: 388
Описание:
Video explains how to train and use the trained convolutional neural networks for segmentation of mitochondria, nuclear envelope, Golgi, and endoplasmic reticulum from volume electron microscopy dataset.
Link to the datasets and trained networks:
https://doi.org/10.5281/zenodo.10043461
Link to the full dataset at EMPIAR:
https://doi.org/10.6019/EMPIAR-11746
Link to Microscopy Image Browser:
http://mib.helsinki.fi/
00:00 Introduction
01:18 Zenodo document description
01:48 0_Segmentation: Ground-truth models
02:50 Protocols for generating training sets
05:01 Detection of ER using 2.5D CNN
09:03 Detection of Golgi using 2.5D CNN
14:38 Detection of Mitochondria using 2.5D CNN
15:36 Detection of Nuclear Envelope using 2D CNN
16:51 Concluding remarks
Повторяем попытку...
Доступные форматы для скачивания:
Скачать видео
-
Информация по загрузке:
![Как сжимаются изображения? [46 МБ ↘↘ 4,07 МБ] JPEG в деталях](https://imager.clipsaver.ru/Kv1Hiv3ox8I/max.jpg)
