Back

qPaLM: quantifying occult microarchitectural relationships in histopathological landscapes

Kendall, T. J.; Duff, C. M.; Thomson, A. M.; Iredale, J. P.

2019-11-02 pathology
10.1101/828004 bioRxiv
Show abstract

Optimal tissue imaging methods should be easy to apply, not require use-specific algorithmic training, and should leverage feature relationships central to subjective gold-standard assessment. We reinterpret histological images as landscapes to describe quantitative pathological landscape metrics (qPaLM), a generalisable framework defining topographic relationships in tissue using geoscience approaches. qPaLM requires no user-dependent training to operate on all image datasets in a classifier-agnostic manner to quantify occult abnormalities, derive mechanistic insights, and define a new feature class for machine-learning diagnostic classification.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.