Deep learning ­– promises for 3D nuclear imaging: a guide for biologists

Abstract

For the past century, the nucleus has been the focus of extensive investigations in cell biology. However, many questions remain about how its shape and size are regulated during development, in different tissues, or during disease and aging. To track these changes, microscopy has long been the tool of choice. Image analysis has revolutionized this field of research by providing computational tools that can be used to translate qualitative images into quantitative parameters. Many tools have been designed to delimit objects in 2D and, eventually, in 3D in order to define their shapes, their number or their position in nuclear space. Today, the field is driven by deep-learning methods, most of which take advantage of convolutional neural networks. These techniques are remarkably adapted to biomedical images when trained using large datasets and powerful computer graphics cards. To promote these innovative and promising methods to cell biologists, this Review summarizes the main concepts and terminologies of deep learning. Special emphasis is placed on the availability of these methods. We highlight why the quality and characteristics of training image datasets are important and where to find them, as well as how to create, store and share image datasets. Finally, we describe deep-learning methods well-suited for 3D analysis of nuclei and classify them according to their level of usability for biologists. Out of more than 150 published methods, we identify fewer than 12 that biologists can use, and we explain why this is the case. Based on this experience, we propose best practices to share deep-learning methods with biologists.

NISNet3D: three-dimensional nuclear synthesis and instance segmentation for fluorescence microscopy images

“…In addition, we also compare NISNet3D to several non-deep learning segmentation methods including the 3D Watershed transform 46 , Squassh 47 , and the “IdentifyPrimaryObject” module from CellProfiler 48 . We used the above methods for comparison with NISNet3D because they have been commonly used in the literature 49 – 53 . We also use the 2D watershed transform and connected component analysis from VTEA for comparison 54 .…”

UNETRIS: transformer-based nuclear instance segmentation for three-dimensional fluorescence microscopy images

“…The non-deep learning methods we compared to are the 3D Watershed transform 40 , Squassh 41 , the 2D watershed transform and connected component analysis from VTEA for comparison 43 , and the "IdentifyPrimaryObject" module from CellProfiler 42 . The above methods for comparison with UNETRIS are used because they have been commonly used in the literature [45][46][47][48][49] . For deep learning methods, the same image volumes used to train UNETRIS are also used to train the deep learning methods.…”

3D Nuclei Segmentation by Combining GAN Based Image Synthesis and Existing 3D Manual Annotations

“…In this work, our objective was not to develop a method achieving state of the art accuracy for segmenting nuclei in 3D. In the past years (Mougeot et al, 2022), nuclei segmentation has seen the development of hundreds of methods competing for the leading position in term of segmentation accuracy. Unfortunately, most of these techniques are out of reach for life science labs and imaging facilities because they can be limited to one operating system or do not provide source code, tutorial or toy datasets (Mougeot et al, 2022).…”

“…In the past years (Mougeot et al, 2022), nuclei segmentation has seen the development of hundreds of methods competing for the leading position in term of segmentation accuracy. Unfortunately, most of these techniques are out of reach for life science labs and imaging facilities because they can be limited to one operating system or do not provide source code, tutorial or toy datasets (Mougeot et al, 2022). We therefore wanted to show that it is also possible to achieve good qualitative segmentation by using already established methods such as Stardist (Schmidt et al, 2018; Weigert et al, 2020) that are widely used in labs and facilities.…”

“…In parallel, the last few years have witnessed the rapid emergence of convolutional neural networks (CNNs) as a method of choice for microscopy image segmentation, achieving an accuracy unheard of for a wide range of segmentation tasks. Among those, deep learning-based nuclei segmentation has been particularly active, with more than one hundred methods being developed since 2019 (Mougeot et al, 2022).…”