Detecting genetic alterations in BRAF and NTRK as oncogenic drivers in digital pathology images: towards model generalization within and across multiple thyroid cohorts

Johannes Höhne, Jacob de Zoete, Arndt A. Schmitz, Tricia Bal, Emmanuelle di Tomaso, Matthias Lenga
Proceedings of the MICCAI Workshop on Computational Pathology, PMLR 156:105-116, 2021.

Abstract

In this paper, we describe the machine learning problem of identifying different types of tumors based on digital pathology images. Given a set of Hematoxylin and Eosin (H&E) stained images of thyroid tumors, we train deep learning models to detect two known molecular oncogenic drivers: \textit{BRAF} mutations and \textit{NTRK} gene fusions. We implement an attention-based multiple instance learning (MIL) classifier and we assess its generalization within and across three independent cohorts. We find that the model can detect both oncogenic drivers with the MIL approach, however the problem remains challenging: our exhaustive evaluation scenarios exemplify unknown data drifts and batch effects in digital pathology as the model performance decreases when processing images from an unseen cohort. These findings highlight the necessity of rich and diverse datasets for training and evaluation as well as methods for domain-agnostic learning.

Cite this Paper

BibTeX
@InProceedings{pmlr-v156-hohne21a, title = {Detecting genetic alterations in BRAF and NTRK as oncogenic drivers in digital pathology images: towards model generalization within and across multiple thyroid cohorts}, author = {H\"ohne, Johannes and de Zoete, Jacob and A.~Schmitz, Arndt and Bal, Tricia and di Tomaso, Emmanuelle and Lenga, Matthias}, booktitle = {Proceedings of the MICCAI Workshop on Computational Pathology}, pages = {105--116}, year = {2021}, editor = {Atzori, Manfredo and Burlutskiy, Nikolay and Ciompi, Francesco and Li, Zhang and Minhas, Fayyaz and Müller, Henning and Peng, Tingying and Rajpoot, Nasir and Torben-Nielsen, Ben and van der Laak, Jeroen and Veta, Mitko and Yuan, Yinyin and Zlobec, Inti}, volume = {156}, series = {Proceedings of Machine Learning Research}, month = {27 Sep}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v156/hohne21a/hohne21a.pdf}, url = {https://proceedings.mlr.press/v156/hohne21a.html}, abstract = {In this paper, we describe the machine learning problem of identifying different types of tumors based on digital pathology images. Given a set of Hematoxylin and Eosin (H&E) stained images of thyroid tumors, we train deep learning models to detect two known molecular oncogenic drivers: \textit{BRAF} mutations and \textit{NTRK} gene fusions. We implement an attention-based multiple instance learning (MIL) classifier and we assess its generalization within and across three independent cohorts. We find that the model can detect both oncogenic drivers with the MIL approach, however the problem remains challenging: our exhaustive evaluation scenarios exemplify unknown data drifts and batch effects in digital pathology as the model performance decreases when processing images from an unseen cohort. These findings highlight the necessity of rich and diverse datasets for training and evaluation as well as methods for domain-agnostic learning.} }
Endnote
%0 Conference Paper %T Detecting genetic alterations in BRAF and NTRK as oncogenic drivers in digital pathology images: towards model generalization within and across multiple thyroid cohorts %A Johannes Höhne %A Jacob de Zoete %A Arndt A. Schmitz %A Tricia Bal %A Emmanuelle di Tomaso %A Matthias Lenga %B Proceedings of the MICCAI Workshop on Computational Pathology %C Proceedings of Machine Learning Research %D 2021 %E Manfredo Atzori %E Nikolay Burlutskiy %E Francesco Ciompi %E Zhang Li %E Fayyaz Minhas %E Henning Müller %E Tingying Peng %E Nasir Rajpoot %E Ben Torben-Nielsen %E Jeroen van der Laak %E Mitko Veta %E Yinyin Yuan %E Inti Zlobec %F pmlr-v156-hohne21a %I PMLR %P 105--116 %U https://proceedings.mlr.press/v156/hohne21a.html %V 156 %X In this paper, we describe the machine learning problem of identifying different types of tumors based on digital pathology images. Given a set of Hematoxylin and Eosin (H&E) stained images of thyroid tumors, we train deep learning models to detect two known molecular oncogenic drivers: \textit{BRAF} mutations and \textit{NTRK} gene fusions. We implement an attention-based multiple instance learning (MIL) classifier and we assess its generalization within and across three independent cohorts. We find that the model can detect both oncogenic drivers with the MIL approach, however the problem remains challenging: our exhaustive evaluation scenarios exemplify unknown data drifts and batch effects in digital pathology as the model performance decreases when processing images from an unseen cohort. These findings highlight the necessity of rich and diverse datasets for training and evaluation as well as methods for domain-agnostic learning.
APA
Höhne, J., de Zoete, J., A. Schmitz, A., Bal, T., di Tomaso, E. & Lenga, M.. (2021). Detecting genetic alterations in BRAF and NTRK as oncogenic drivers in digital pathology images: towards model generalization within and across multiple thyroid cohorts. Proceedings of the MICCAI Workshop on Computational Pathology, in Proceedings of Machine Learning Research 156:105-116 Available from https://proceedings.mlr.press/v156/hohne21a.html.

Related Material