Revealing architectural order with quantitative label-free imaging and deep learning
- Chan Zuckerberg Biohub, United States
- UCSF School of Medicine, United States
- Stanford University, United States
Abstract
We report quantitative label-free imaging with phase and polarization (QLIPP) for simultaneous measurement of density, anisotropy, and orientation in unlabeled live cells and tissue slices. We combine QLIPP with deep neural networks to predict fluorescence images of diverse cell and tissue structures. QLIPP images reveal anatomical regions and axon tract orientation in prenatal human brain tissue sections that are not visible using brightfield imaging. We report a variant of UNet architecture, multi-channel 2.5D U-Net, for computationally efficient prediction of fluorescence images in three dimensions and over large fields of view. Further, we develop data normalization methods for accurate prediction of myelin distribution over large brain regions. We show that experimental defects in labeling the human tissue can be rescued with quantitative label-free imaging and neural network model. We anticipate that the proposed method will enable new studies of architectural order at spatial scales ranging from organelles to tissue.
Data availability
Our experiments generated imaging data from mouse kidney tissue and human brain tissue slices that are useful for machine learning and other analyses. We are in the process of uploading them to a public image archive.
Article and author information
Author details
Funding
Chan Zuckerberg Biohub
- Syuan-Ming Guo
- Li-Hao Yeh
- Jenny Folkesson
- Ivan E Ivanov
- Matthew G Keefe
- David Shin
- Bryant B Chhun
- Nathan H Cho
- Tomasz Nowakowski
- Shalin B Mehta
The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
Ethics
Human subjects: De-identified brain tissue samples were received with patient consent in accordance with a protocol approved by the Human Gamete, Embryo, and Stem Cell Research Committee (institutional review board) at the University of California, San Francisco.
Copyright
© 2020, Guo et al.
This article is distributed under the terms of the Creative Commons Attribution License permitting unrestricted use and redistribution provided that the original author and source are credited.
Metrics
-
- 7,330
- views
-
- 879
- downloads
-
- 71
- citations
Views, downloads and citations are aggregated across all versions of this paper published by eLife.
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Revealing architectural order with quantitative label-free imaging and deep learning
eLife 9:e55502.
https://doi.org/10.7554/eLife.55502
Further reading
-
- Neuroscience
This study examines whether auditory cortex anatomy reflects multilingual experience, specifically individuals’ phonological repertoire. Using data from over 200 participants exposed to 1–7 languages across 36 languages, we analyzed the role of language experience and typological distances between languages they spoke in shaping neural signatures of multilingualism. Our findings reveal a negative relationship between the thickness of the left and right second transverse temporal gyrus (TTG) and participants’ degree of multilingualism. Models incorporating phoneme-level information in the language experience index explained the most variance in TTG thickness, suggesting that a more extensive and more phonologically diverse language experience is associated with thinner cortices in the second TTG. This pattern, consistent across two datasets, supports the idea of experience-driven pruning and neural efficiency. Our findings indicate that experience with typologically distant languages appear to impact the brain differently than those with similar languages. Moreover, they suggest that early auditory regions seem to represent phoneme-level cross-linguistic information, contrary to the most established models of language processing in the brain, which suggest that phonological processing happens in more lateral posterior superior temporal gyrus (STG) and superior temporal sulcus (STS).
-
- Neuroscience
The brain is organized into systems and networks of interacting components. The functional connections among these components give insight into the brain's organization and may underlie some cognitive effects of aging. Examining the relationship between individual differences in brain organization and cognitive function in older adults who have reached oldest old ages with healthy cognition can help us understand how these networks support healthy cognitive aging. We investigated functional network segregation in 146 cognitively healthy participants aged 85+ in the McKnight Brain Aging Registry. We found that the segregation of the association system and the individual networks within the association system [the fronto-parietal network (FPN), cingulo-opercular network (CON) and default mode network (DMN)], has strong associations with overall cognition and processing speed. We also provide a healthy oldest-old (85+) cortical parcellation that can be used in future work in this age group. This study shows that network segregation of the oldest-old brain is closely linked to cognitive performance. This work adds to the growing body of knowledge about differentiation in the aged brain by demonstrating that cognitive ability is associated with differentiated functional networks in very old individuals representing successful cognitive aging.
