Identifying actionable driver mutations in lung cancer using an efficient Asymmetric Transformer Decoder

Biagio Brattoli, Jack Shi, Jongchan Park, Taebum Lee, Donggeun Yoo, Sergio Pereira
Proceedings of the MICCAI Workshop on Computational Pathology, PMLR 316:75-85, 2026.

Abstract

Identifying actionable driver mutations in non-small cell lung cancer (NSCLC) can impact treatment decisions and significantly improve patient outcomes. Despite guideline recommendations, broader adoption of genetic testing remains challenging due to limited availability and lengthy urnaround times. Machine Learning (ML) methods for Computational Pathology (CPath) offer a potential solution; however, research often focuses on only one or two common mutations, limiting the clinical value of these tools and the pool of patients who can benefit from them. This study evaluates various Multiple Instance Learning (MIL) techniques to detect six key actionable NSCLC driver mutations: ALK, BRAF, EGFR, ERBB2, KRAS, and MET ex14. Additionally, we introduce an Asymmetric Transformer Decoder model that employs queries and key-values of varying dimensions to maintain a low query dimensionality. This approach efficiently extracts information from patch embeddings and minimizes overfitting risks, proving highly adaptable to the MIL setting. Moreover, we present a method to directly utilize tissue type in the model, addressing a typical MIL limitation where either all regions or only some specific regions are analyzed, neglecting biological relevance. Our method outperforms top MIL models by an average of 3%, and over 4% when predicting rare mutations such as ERBB2 and BRAF, moving ML-based tests closer to being practical alternatives to standard genetic testing.

Cite this Paper

BibTeX
@InProceedings{pmlr-v316-brattoli26a, title = {Identifying actionable driver mutations in lung cancer using an efficient Asymmetric Transformer Decoder}, author = {Brattoli, Biagio and Shi, Jack and Park, Jongchan and Lee, Taebum and Yoo, Donggeun and Pereira, Sergio}, booktitle = {Proceedings of the MICCAI Workshop on Computational Pathology}, pages = {75--85}, year = {2026}, editor = {Studer, Linda and Ciompi, Francesco and Khalili, Nadieh and Faryna, Khrystyna and Faryna, Khrystyna and Yeong, Joe and Lau, Mai Chan and Chen, Hao and Liu, Ziyi and Brattoli, Biagio}, volume = {316}, series = {Proceedings of Machine Learning Research}, month = {27 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v316/main/assets/brattoli26a/brattoli26a.pdf}, url = {https://proceedings.mlr.press/v316/brattoli26a.html}, abstract = {Identifying actionable driver mutations in non-small cell lung cancer (NSCLC) can impact treatment decisions and significantly improve patient outcomes. Despite guideline recommendations, broader adoption of genetic testing remains challenging due to limited availability and lengthy urnaround times. Machine Learning (ML) methods for Computational Pathology (CPath) offer a potential solution; however, research often focuses on only one or two common mutations, limiting the clinical value of these tools and the pool of patients who can benefit from them. This study evaluates various Multiple Instance Learning (MIL) techniques to detect six key actionable NSCLC driver mutations: ALK, BRAF, EGFR, ERBB2, KRAS, and MET ex14. Additionally, we introduce an Asymmetric Transformer Decoder model that employs queries and key-values of varying dimensions to maintain a low query dimensionality. This approach efficiently extracts information from patch embeddings and minimizes overfitting risks, proving highly adaptable to the MIL setting. Moreover, we present a method to directly utilize tissue type in the model, addressing a typical MIL limitation where either all regions or only some specific regions are analyzed, neglecting biological relevance. Our method outperforms top MIL models by an average of 3%, and over 4% when predicting rare mutations such as ERBB2 and BRAF, moving ML-based tests closer to being practical alternatives to standard genetic testing.} }
Endnote
%0 Conference Paper %T Identifying actionable driver mutations in lung cancer using an efficient Asymmetric Transformer Decoder %A Biagio Brattoli %A Jack Shi %A Jongchan Park %A Taebum Lee %A Donggeun Yoo %A Sergio Pereira %B Proceedings of the MICCAI Workshop on Computational Pathology %C Proceedings of Machine Learning Research %D 2026 %E Linda Studer %E Francesco Ciompi %E Nadieh Khalili %E Khrystyna Faryna %E Khrystyna Faryna %E Joe Yeong %E Mai Chan Lau %E Hao Chen %E Ziyi Liu %E Biagio Brattoli %F pmlr-v316-brattoli26a %I PMLR %P 75--85 %U https://proceedings.mlr.press/v316/brattoli26a.html %V 316 %X Identifying actionable driver mutations in non-small cell lung cancer (NSCLC) can impact treatment decisions and significantly improve patient outcomes. Despite guideline recommendations, broader adoption of genetic testing remains challenging due to limited availability and lengthy urnaround times. Machine Learning (ML) methods for Computational Pathology (CPath) offer a potential solution; however, research often focuses on only one or two common mutations, limiting the clinical value of these tools and the pool of patients who can benefit from them. This study evaluates various Multiple Instance Learning (MIL) techniques to detect six key actionable NSCLC driver mutations: ALK, BRAF, EGFR, ERBB2, KRAS, and MET ex14. Additionally, we introduce an Asymmetric Transformer Decoder model that employs queries and key-values of varying dimensions to maintain a low query dimensionality. This approach efficiently extracts information from patch embeddings and minimizes overfitting risks, proving highly adaptable to the MIL setting. Moreover, we present a method to directly utilize tissue type in the model, addressing a typical MIL limitation where either all regions or only some specific regions are analyzed, neglecting biological relevance. Our method outperforms top MIL models by an average of 3%, and over 4% when predicting rare mutations such as ERBB2 and BRAF, moving ML-based tests closer to being practical alternatives to standard genetic testing.
APA
Brattoli, B., Shi, J., Park, J., Lee, T., Yoo, D. & Pereira, S.. (2026). Identifying actionable driver mutations in lung cancer using an efficient Asymmetric Transformer Decoder. Proceedings of the MICCAI Workshop on Computational Pathology, in Proceedings of Machine Learning Research 316:75-85 Available from https://proceedings.mlr.press/v316/brattoli26a.html.

Related Material