U-Net: deep learning for cell counting, detection, and morphometry

By Thorsten Falk, Dominic Mai, R. Bensch, Özgün Çiçek, A. Abdulkadir, Yassine Marrakchi, Anton Böhm, Jan Deubner, Zoe Jäckel, Katharina Seiwald and 12 more. 2018

1,774 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Biophysics, Computer Vision and Pattern Recognition, Medical Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Biophysics

In Computer Vision and Pattern Recognition

In Medical Imaging

Citations in the corpus

Application of Artificial Intelligence Technology in Oncology: Towards the Establishment of Precision Medicine (2020, methodology): “Typical deep learning techniques used in medical imaging include convolutional neural networks for image classification [6], recurrent neural networks for reporting tasks [7], and a U-net for image segmentation [8].”
Medical Image Segmentation Review: The Success of U-Net (2022, background): “Semantic segmentation transforms a plain biomedical image modality into a meaningful and spatially wellorganized pattern that could aid medical discoveries and diagnoses [97], [98] and sometimes is beneficial to patients too as they may be able to avoid an invasive medical procedure [99].”
nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation (2020, background): “emantic segmentation transforms raw biomedical image data into meaningful, spatially structured information and thus plays an essential role in scientific discovery 1,2 .”
Survey of Explainable AI Techniques in Healthcare (2023, background): “For example, the development of deep learning (DL) models for healthcare has resulted in advanced performance, such as the U-Net model in image segmentation [11].”
Deep learning for cellular image analysis (2019, methodology): “We recommend that users focus their initial efforts on the existing DeepCell, CDeep3M, U-Net, CellProfiler, or Mask R-CNN software libraries, as these methods have been applied successfully to a variety of data types, provide pretrained models, and support both model training and deployment on new d”
Evolutionary Compression of Deep Neural Networks for Biomedical Image Segmentation (2020, methodology): “Finally, we present a pruning operator for eliminating filters convolved on the concatenated feature maps from different layers, which is commonly used for capturing multilevel features from images [9].”
Deep-learning-based information mining from ocean remote-sensing imagery (2020, background): “An instance of a weight map is given in [16].”
U-Sleep: resilient high-frequency sleep staging (2021, background): “They mark the state-of-the-art in image segmentation, with the U-Net arguably being the most popular architecture so far 40,41 .”
Evaluation of Deep Learning Strategies for Nucleus Segmentation in Fluorescence Images (2018, methodology): “Also, a U-Net plugin was independently developed for ImageJ for running generic cell segmentation and quanti ﬁ cation tasks (42).”
Democratising deep learning for microscopy with ZeroCostDL4Mic (2021, methodology): “These include several supervised and self-supervised DL networks for image segmentation and object detection (using U-Net 2,14,28 , StarDist 5,6 and YOLOv2 3 ), image denoising and restoration (using CARE 7 and Noise2Void 8 ), super-resolution microscopy (using Deep-STORM 29 ) and image-to-image…”
Spherical Fourier Neural Operators: Learning Stable Dynamics on the Sphere (2023, background): “…is also computationally more efficient compared to standard vision transformers, which have quadratic time complexity, and more effective in capturing global dependencies compared to convolutional neural networks, graph neural networks, and other local models (McCormick, 1987; Falk et al., 2019).”