#jobim2024 Aurélien Beaude talks about
AttOmics: Attention-based architecture for diagnosis and prognosis from Omics data.
https://easychair.org/smart-program/JOBIM2024/2024-06-26.html#talk:261667

About personalized medicine, from omics data, using DL.

SoA notably talks about using ontological layers before auto-encoder layer, a method that I find interesting as well :-)

Introduces using self-attention, to pick relevant information from input. But a costly (quadratic) method! So proposes to group features and apply self-attention on groups.

Program for Wednesday, June 26th

Various grouping strategy are possible (again, one possibilities using ontologies ❤). Here, learn the grouping as well, taking into account inter-groups interactions.

Tested on TCGA, for diagnosis (cancer type) and prognosis.

Performances are better than other methods.

Looked at attention maps, to identify pathways leveraged by the NN, and found some correlations with existing knowledge.

May be applied to predicting subtypes on smaller datasets using transfer learning and after having redifined subtypes based on molecular profiles rather than on histopathology.