PATHOMICS – A NOVEL OMICS APPROACH FOR HISTOPATHOLOGY
Artificial Intelligence (AI) and Deep Learning (DL) hold great promise to transform pathology practice. Currently, the majority of commercially available products and AI research focuses on end-toend AI, i.e., an approach in which the model learns all the steps between the initial input and the fin...
Saved in:
| Format: | Article |
|---|---|
| Language: | English |
| Published: |
PAGEPress Publications
2025-08-01
|
| Series: | European Journal of Histochemistry |
| Subjects: | |
| Online Access: | https://www.ejh.it/ejh/article/view/4284 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Artificial Intelligence (AI) and Deep Learning (DL) hold great promise to transform pathology practice. Currently, the majority of commercially available products and AI research focuses on end-toend AI, i.e., an approach in which the model learns all the steps between the initial input and the final output, providing qualitative or semiquantitative (class) data. A complementary or alternative approach to analyse histomorphology is using DL-based segmentation of relevant histological compartments and cells, followed by extraction of relevant quantitative data (features). If done on a large scale, it is termed pathomics, representing a novel -omics approach for morphology at the microscopical level. Pathomics complements molecular omics, like genomics or transcriptomics, or radiomics, which aims at quantifying radiology images at the macroscopic level. This lecture will explore the potential of pathomics and compare it with end-to-end models, focusing on kidney pathology.
|
|---|---|
| ISSN: | 1121-760X 2038-8306 |