Abstract
In biopsy diagnosis not everything seen is understood for what it means and uncertainty lurks when key features are sparse or signal different possibilities. Therefore, insights that recalibrate interpretation are welcome. These most usefully come from independent approaches—the rejection expression signature for late allograft interface hepatitis, 1 or serologic data informing histologic criteria for antibody-mediated rejection. 2 However, there is potential in mining histologic datasets for inherent feature patterns using unbiased data-driven algorithms. In the context of rejection, we expect to recapitulate known phenotypes, but new approaches might better delineate margins between entities, overlooked feature combinations, or poorly characterized conditions.
Original language | English |
---|---|
Pages (from-to) | 892-893 |
Journal | American Journal of Transplantation |
Volume | 24 |
Issue number | 6 |
DOIs | |
Publication status | Published - 5 Feb 2024 |
Keywords / Materials (for Non-textual outputs)
- liver
- transplantation
- pathology
- rejection
- allograft
- archetypal analysis