bioRxivJClub
Linking clinical border morphology to cell-scale growth mechanisms in melanoma: a segmentation-to-simulation alignment study
bioRxiv · · Preprint · Open access
Liu, Zu
Abstract ↗AI summary
The abstract is read at the publisher; the summary is JClub's.
In 1,581 melanoma lesions, lobulation-linked shape metrics shift from benign to invasive (compactness 1.351→1.810; solidity 0.910→0.853), while box-counting fractal dimension changes only 0.024.
- Why it matters: Irregular melanoma borders are clinically informative, but box-counting fractal dimension has not been tied to the cell-scale growth mechanism that generates it.
- What they did: Seven boundary descriptors were measured on 1,581 lesions from the 2018 International Skin Imaging Collaboration archive, then a hybrid cellular automaton simulated boundaries using the same measurement code while sweeping four canonical cell-level mechanisms.
- The result: A long-range spatial heterogeneity in growth rate reproduced the malignant lobular signature to a distance of 0.187, with correlation length dominating by total-effect index, whereas empty-neighbour division rule variants did not.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.
- Open access