3D Nucleus Instance Segmentation for Whole-Brain Microscopy Images

Ma, Junbo; Krupa, Oleh; Glass, Madison R.; McCormick, Carolyn M.; Borland, David; Kim, Minjeong; Stein, Jason L.; Wu, Guorong

doi:10.1007/978-3-030-78191-0_39

Cited by 1 publication

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The context prediction and the jigsaw puzzle tasks could have more than one possible answer (particularly for images where structures are far apart), rotations are not defined in the plane of the image, and pseudo-colors are arbitrarily defined from photon-counts. Instance discrimination, 29 geometric self-distillation, 30 classification of image parameters (e.g., scale 31 ), and image prediction 32 , 33 are all pretext tasks that are applicable to microscopy images, do not require semantic labels, and yet still enable the model to learn generalizable representations of the data. Image prediction can be used as a pretext task to learn denoising in temporal imaging data, improving signal-to-noise ratio in calcium 34 , 35 and voltage 36 imaging without the need for ground-truth denoised images, which are difficult to obtain.…”

Section: Discussionmentioning

confidence: 99%

Addressing annotation and data scarcity when designing machine learning strategies for neurophotonics

Bouchard,

Bernatchez,

Lavoie-Cardinal

2023

Neurophoton.

View full text Add to dashboard Cite

Machine learning has revolutionized the way data are processed, allowing information to be extracted in a fraction of the time it would take an expert. In the field of neurophotonics, machine learning approaches are used to automatically detect and classify features of interest in complex images. One of the key challenges in applying machine learning methods to the field of neurophotonics is the scarcity of available data and the complexity associated with labeling them, which can limit the performance of data-driven algorithms. We present an overview of various strategies, such as weakly supervised learning, active learning, and domain adaptation that can be used to address the problem of labeled data scarcity in neurophotonics. We provide a comprehensive overview of the strengths and limitations of each approach and discuss their potential applications to bioimaging datasets. In addition, we highlight how different strategies can be combined to increase model performance on those datasets. The approaches we describe can help to improve the accessibility of machine learning-based analysis with limited number of annotated images for training and can enable researchers to extract more meaningful insights from small datasets.

show abstract