Searcharxiv⌕ Search

arXiv subjects

Dorry Segev

Publications and source records attributed to Dorry Segev.

2 recordsLinked to original sources

When Do Histology Foundation Models Need Geometric Test-Time Augmentation?

Test-time augmentation (TTA) aggregates model predictions across multiple transformed views of an input during inference. Because histology images lack a canonical orientation, $D_4$ TTA---pooling across the 8 rotations and reflections of a square---is a standard default in computational pathology. However, the rise of self-supervised histology foundation models (FMs) with inherent rotational invariance calls its continued utility into question. We benchmark 9 histology FMs and 6 general-purpose models across four patch-level and two whole-slide image (WSI) datasets. We find that $D_4$ TTA remains a reliable post hoc improvement, but benefits diminish from $+1.98$ percentage points (pp) for general-purpose backbones to $+0.50$ pp for histology FMs on TCGA-UT, and shrink further for the largest models. To eliminate computational overhead, our experiments demonstrate that entropy routing---applying TTA selectively to uncertain predictions---recovers most of the gain while augmenting only ${\sim}13\%$ of samples. Finally, we show that patch-level TTA improvements do not transfer to slide-level multiple instance learning (MIL). Code: https://github.com/lars-chen/histology-tta

cs.CV↗

Maximum Matchings in Graphs for Allocating Kidney Paired Donation

Relatives and friends of an end-stage renal disease patient who offer to donate a kidney are often found to be incompatible with their intended recipients. Kidney paired donation matches one patient and his incompatible donor with another patient and donor in the same situation for an organ exchange. Let patient- donor pairs be the vertices of an undirected graph G, with an edge connecting any two reciprocally compatible vertices. A matching in G is a feasible set of paired donations. We describe various optimization problems on kidney paired donation graphs G and the merits of each in clinical transplantation. Because some matches are geographically undesirable, and the expected lifespan of a transplanted kidney depends on the immunologic concordance of donor and recipient, we weight the edges of G and seek a maximum edge-weight matching. Unfortunately, such matchings might not have the maximum cardinality; there is a risk of an unpredictable trade-off between the quality and quantity of paired donations. We propose an edge-weighting of G which guarantees that every matching with maximum weight also has maximum cardinality, and also maximizes the number of transplants for an exceptional subset of recipients, while reducing travel and favoring immunologic concordance.

cs.GT↗