More people die every year from kidney disease than breast or prostate cancer. Kidneytransplantation is life-saving, yet the donor organ shortage and high organ discard ratecontributes to 13 deaths daily among patients awaiting transplant. The decision to use ordiscard a donor kidney relies heavily on microscopic quantitation of chronic damage bypathologists. The current standard of care relies on a manual process that is subject tosignificant human variability and inefficiency, resulting in potentially healthy kidneys beingdiscarded and potentially damaged kidneys being transplanted inappropriately. Our teamdeveloped the first Deep Learning model to quantify percent global glomerulosclerosis in donorkidney frozen section biopsy whole slide images. We developed a cloud-based platform to applythe Deep Learning model to analyze kidney biopsy whole slide images in under 6 minutes withaccuracy and precision equal to or greater than current standard of care pathologists. We havealso developed a Deep Learning model to quantify interstitial fibrosis on donor kidney biopsywhole slide images. This innovative approach has the potential to transform donor kidney biopsyevaluation by improving pathologist efficiency, accuracy, and precision ultimately resulting inoptimized donor organ utilization, improved patient outcomes, and diminished health care costs.The goal of this project is to develop a Deep Learning technique for quantification ofarteriosclerosis, to support evaluation of donor kidneys prior to transplantation. This will beachieved by assembling a team of expert pathologists and computer scientists specializing inmachine learning. The proposal will evaluate the accuracy and precision of the arteriosclerosisDeep Learning model. The functionality of the Trusted Kidney software platform will beimproved beyond the current usable product into a commercially viable solution for multiplelaboratories.
Public Health Relevance Statement: PUBLIC HEALTH RELEVANCE STATEMENT
Before kidneys can be transplanted, they must be examined using a microscope to ensure the
kidney is healthy enough for transplant. A limitation of microscopic examination by pathologists
is the inherent human variability in quantifying the degree of chronic vascular damage. The
result is potentially healthy organs being discarded or damaged kidneys being used
inappropriately. This funding will support developing artificial intelligence tools to assist
pathologists with quantifying chronic vascular damage in donor kidneys prior to transplantation,
resulting in more consistent and objective biopsy evaluations, minimizing discard of potentially
healthy kidneys, and optimizing placement of kidneys for transplant.
Project Terms: 