ASN's Mission

To create a world without kidney diseases, the ASN Alliance for Kidney Health elevates care by educating and informing, driving breakthroughs and innovation, and advocating for policies that create transformative changes in kidney medicine throughout the world.

learn more

Contact ASN

1401 H St, NW, Ste 900, Washington, DC 20005

email@asn-online.org

202-640-4660

The Latest on Twitter

Kidney Week

Abstract: TH-PO535

Visualizing and Quantifying Biological Age in the Kidney

Session Information

  • Pathology and Lab Medicine
    November 03, 2022 | Location: Exhibit Hall, Orange County Convention Center‚ West Building
    Abstract Time: 10:00 AM - 12:00 PM

Category: Pathology and Lab Medicine

  • 1700 Pathology and Lab Medicine

Authors

  • Sheehan, Susan Marie, The Jackson Laboratory, Bar Harbor, Maine, United States
  • Mawe, Seamus, The Jackson Laboratory, Bar Harbor, Maine, United States
  • Chen, Mandy, The Jackson Laboratory, Bar Harbor, Maine, United States
  • Korstanje, Ron, The Jackson Laboratory, Bar Harbor, Maine, United States
  • Mahoney, John Matthew, The Jackson Laboratory, Bar Harbor, Maine, United States
Background

The kidney is one of the organs with noticeable age-related changes in structure and function. However, we do not understand the mechanisms and timing involved in kidney aging. The Geropathology Research Network (GRN) published a grading scheme for aged mice in several tissues, including the kidney. Use of this grading scheme is labor intensive, requires expert knowledge to implement, and has large interscorer variability. Using image analysis and machine learning approaches, we developed a high-throughput classifier that can be used to improve the quantification and resolve the unknown mechanisms of aging.

Methods

Using the same slides as the GRN, which includes male mice from two strains (C57BL/6 and CB6F1) and combining them with additional C57BL/6J mice, we developed a training set that includes 12 mice at four ages (8, 13, 20 and 32 months). We adopted a light deep learning architecture (LinkNet) and trained the model on images from ImageNet to develop an ordinal classifier that provides us with pixel level information about aging. By looking at the predictions, we can ascertain how likely a tissue is young or old age at a given pixel. Additionally, we can generate an overall age score for each kidney.

Results

After we trained the classifier, we tested the classifier in an additional set of male and female C57BL/6J mice. We can distinguish between young and old tissue and see a strong correlation with the scores using the GRN grading. We wanted to determine whether kidneys age uniformly or if specific regions or structures age faster than others. We developed a method that allows us to ‘paint’ our aging scores onto kidneys and see that aging happens first in a thin band in the renal cortex, which aligns with the findings of pathologists.

Conclusion

Our tool can be used to measure biological age in the kidney and is critical in developing and evaluating interventions that affect kidney health span and overall lifespan. The tool, and tutorials about how to apply this to your own data set, are made publicly available at agingmice.org. We are now applying this tool to more diverse and larger datasets and refining our ability to quantify and localize kidney aging such that we can understand the mechanisms that change spatially and structurally with age.

Funding

  • Other NIH Support