Virtual mouse brain histology from multi-contrast MRI via deep learning

  1. Zifei Liang
  2. Choong H Lee
  3. Tanzil M Arefin
  4. Zijun Dong
  5. Piotr Walczak
  6. Song Hai Shi
  7. Florian Knoll
  8. Yulin Ge
  9. Leslie Ying
  10. Jiangyang Zhang  Is a corresponding author
  1. New York University School of Medicine, United States
  2. University of Maryland, United States
  3. Memorial Sloan Kettering Cancer Center
  4. University at Buffalo, State University of New York, United States

Abstract

1H MRI maps brain structure and function non-invasively through versatile contrasts that exploit inhomogeneity in tissue micro-environments. Inferring histopathological information from MRI findings, however, remains challenging due to absence of direct links between MRI signals and cellular structures. Here, we show that deep convolutional neural networks, developed using co-registered multi-contrast MRI and histological data of the mouse brain, can estimate histological staining intensity directly from MRI signals at each voxel. The results provide three-dimensional maps of axons and myelin with tissue contrasts that closely mimics target histology and enhanced sensitivity and specificity compared to conventional MRI markers. Furthermore, the relative contribution of each MRI contrast within the networks can be used to optimize multi-contrast MRI acquisition. We anticipate our method to be a starting point for translation of MRI results into easy-to-understand virtual histology for neurobiologists and provide resources for validating novel MRI techniques.

Data availability

All data and source codes used in this study are available at https://www.github.com/liangzifei/MRH-net/. The data can also be found at datadryad.org

The following data sets were generated
The following previously published data sets were used
    1. Lein ES et al
    (2006) Allen Mouse Brain Atlas
    The reference data at http://connectivity.brain-map.org/static/referencedata.

Article and author information

Author details

  1. Zifei Liang

    Department of Radiology, New York University School of Medicine, New York, United States
    Competing interests
    The authors declare that no competing interests exist.
  2. Choong H Lee

    Department of Radiology, New York University School of Medicine, New York, United States
    Competing interests
    The authors declare that no competing interests exist.
  3. Tanzil M Arefin

    Department of Radiology, New York University School of Medicine, New York, United States
    Competing interests
    The authors declare that no competing interests exist.
  4. Zijun Dong

    Department of Radiology, New York University School of Medicine, New York, United States
    Competing interests
    The authors declare that no competing interests exist.
  5. Piotr Walczak

    Department of Diagnostic Radiology and Nuclear Medicine, University of Maryland, Baltimore, United States
    Competing interests
    The authors declare that no competing interests exist.
  6. Song Hai Shi

    Developmental Biology Program, Memorial Sloan Kettering Cancer Center
    Competing interests
    The authors declare that no competing interests exist.
  7. Florian Knoll

    Department of Radiology, New York University School of Medicine, New York, United States
    Competing interests
    The authors declare that no competing interests exist.
  8. Yulin Ge

    Department of Radiology, New York University School of Medicine, New York, United States
    Competing interests
    The authors declare that no competing interests exist.
  9. Leslie Ying

    Departments of Biomedical Engineering, University at Buffalo, State University of New York, Buffalo, United States
    Competing interests
    The authors declare that no competing interests exist.
  10. Jiangyang Zhang

    Department of Radiology, New York University School of Medicine, New York, United States
    For correspondence
    jiangyang.zhang@nyulangone.org
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-0740-2662

Funding

Eunice Kennedy Shriver National Institute of Child Health and Human Development (R01HD074593)

  • Jiangyang Zhang

National Institute of Neurological Disorders and Stroke (R01NS102904)

  • Jiangyang Zhang

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Reviewing Editor

  1. Saad Jbabdi, University of Oxford, United Kingdom

Ethics

Animal experimentation: This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All of the animals were handled according to approved institutional animal care and use committee (IACUC) protocols (s16-00145-133) of the New York University.

Version history

  1. Preprint posted: May 3, 2020 (view preprint)
  2. Received: July 20, 2021
  3. Accepted: January 27, 2022
  4. Accepted Manuscript published: January 28, 2022 (version 1)
  5. Version of Record published: February 11, 2022 (version 2)

Copyright

© 2022, Liang 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

  • 2,357
    views
  • 331
    downloads
  • 4
    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)

  1. Zifei Liang
  2. Choong H Lee
  3. Tanzil M Arefin
  4. Zijun Dong
  5. Piotr Walczak
  6. Song Hai Shi
  7. Florian Knoll
  8. Yulin Ge
  9. Leslie Ying
  10. Jiangyang Zhang
(2022)
Virtual mouse brain histology from multi-contrast MRI via deep learning
eLife 11:e72331.
https://doi.org/10.7554/eLife.72331

Share this article

https://doi.org/10.7554/eLife.72331

Further reading

    1. Medicine
    Christin Krause, Jan H Britsemmer ... Henriette Kirchner
    Research Article

    Background:

    The development of obesity-associated comorbidities such as type 2 diabetes (T2D) and hepatic steatosis has been linked to selected microRNAs in individual studies; however, an unbiased genome-wide approach to map T2D induced changes in the miRNAs landscape in human liver samples, and a subsequent robust identification and validation of target genes are still missing.

    Methods:

    Liver biopsies from age- and gender-matched obese individuals with (n=20) or without (n=20) T2D were used for microRNA microarray analysis. The candidate microRNA and target genes were validated in 85 human liver samples, and subsequently mechanistically characterized in hepatic cells as well as by dietary interventions and hepatic overexpression in mice.

    Results:

    Here, we present the human hepatic microRNA transcriptome of type 2 diabetes in liver biopsies and use a novel seed prediction tool to robustly identify microRNA target genes, which were then validated in a unique cohort of 85 human livers. Subsequent mouse studies identified a distinct signature of T2D-associated miRNAs, partly conserved in both species. Of those, human-murine miR-182–5 p was the most associated with whole-body glucose homeostasis and hepatic lipid metabolism. Its target gene LRP6 was consistently lower expressed in livers of obese T2D humans and mice as well as under conditions of miR-182–5 p overexpression. Weight loss in obese mice decreased hepatic miR-182–5 p and restored Lrp6 expression and other miR-182–5 p target genes. Hepatic overexpression of miR-182–5 p in mice rapidly decreased LRP6 protein levels and increased liver triglycerides and fasting insulin under obesogenic conditions after only seven days.

    Conclusions:

    By mapping the hepatic miRNA-transcriptome of type 2 diabetic obese subjects, validating conserved miRNAs in diet-induced mice, and establishing a novel miRNA prediction tool, we provide a robust and unique resource that will pave the way for future studies in the field. As proof of concept, we revealed that the repression of LRP6 by miR-182–5 p, which promotes lipogenesis and impairs glucose homeostasis, provides a novel mechanistic link between T2D and non-alcoholic fatty liver disease, and demonstrate in vivo that miR-182–5 p can serve as a future drug target for the treatment of obesity-driven hepatic steatosis.

    Funding:

    This work was supported by research funding from the Deutsche Forschungsgemeinschaft (KI 1887/2-1, KI 1887/2-2, KI 1887/3-1 and CRC-TR296), the European Research Council (ERC, CoG Yoyo LepReSens no. 101002247; PTP), the Helmholtz Association (Initiative and Networking Fund International Helmholtz Research School for Diabetes; MB) and the German Center for Diabetes Research (DZD Next Grant 82DZD09D1G).

    1. Immunology and Inflammation
    2. Medicine
    Joanna C Porter, Jamie Inshaw ... Venizelos Papayannopoulos
    Research Article

    Background:

    Prinflammatory extracellular chromatin from neutrophil extracellular traps (NETs) and other cellular sources is found in COVID-19 patients and may promote pathology. We determined whether pulmonary administration of the endonuclease dornase alfa reduced systemic inflammation by clearing extracellular chromatin.

    Methods:

    Eligible patients were randomized (3:1) to the best available care including dexamethasone (R-BAC) or to BAC with twice-daily nebulized dornase alfa (R-BAC + DA) for seven days or until discharge. A 2:1 ratio of matched contemporary controls (CC-BAC) provided additional comparators. The primary endpoint was the improvement in C-reactive protein (CRP) over time, analyzed using a repeated-measures mixed model, adjusted for baseline factors.

    Results:

    We recruited 39 evaluable participants: 30 randomized to dornase alfa (R-BAC +DA), 9 randomized to BAC (R-BAC), and included 60 CC-BAC participants. Dornase alfa was well tolerated and reduced CRP by 33% compared to the combined BAC groups (T-BAC). Least squares (LS) mean post-dexamethasone CRP fell from 101.9 mg/L to 23.23 mg/L in R-BAC +DA participants versus a 99.5 mg/L to 34.82 mg/L reduction in the T-BAC group at 7 days; p=0.01. The anti-inflammatory effect of dornase alfa was further confirmed with subgroup and sensitivity analyses on randomised participants only, mitigating potential biases associated with the use of CC-BAC participants. Dornase alfa increased live discharge rates by 63% (HR 1.63, 95% CI 1.01–2.61, p=0.03), increased lymphocyte counts (LS mean: 1.08 vs 0.87, p=0.02) and reduced circulating cf-DNA and the coagulopathy marker D-dimer (LS mean: 570.78 vs 1656.96 μg/mL, p=0.004).

    Conclusions:

    Dornase alfa reduces pathogenic inflammation in COVID-19 pneumonia, demonstrating the benefit of cost-effective therapies that target extracellular chromatin.

    Funding:

    LifeArc, Breathing Matters, The Francis Crick Institute (CRUK, Medical Research Council, Wellcome Trust).

    Clinical trial number:

    NCT04359654.