We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

q-bio.QM

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Quantitative Biology > Quantitative Methods

Title: A robust and lightweight deep attention multiple instance learning algorithm for predicting genetic alterations

Abstract: Deep-learning models based on whole-slide digital pathology images (WSIs) become increasingly popular for predicting molecular biomarkers. Instance-based models has been the mainstream strategy for predicting genetic alterations using WSIs although bag-based models along with self-attention mechanism-based algorithms have been proposed for other digital pathology applications. In this paper, we proposed a novel Attention-based Multiple Instance Mutation Learning (AMIML) model for predicting gene mutations. AMIML was comprised of successive 1-D convolutional layers, a decoder, and a residual weight connection to facilitate further integration of a lightweight attention mechanism to detect the most predictive image patches. Using data for 24 clinically relevant genes from four cancer cohorts in The Cancer Genome Atlas (TCGA) studies (UCEC, BRCA, GBM and KIRC), we compared AMIML with one popular instance-based model and four recently published bag-based models (e.g., CHOWDER, HE2RNA, etc.). AMIML demonstrated excellent robustness, not only outperforming all the five baseline algorithms in the vast majority of the tested genes (17 out of 24), but also providing near-best-performance for the other seven genes. Conversely, the performance of the baseline published algorithms varied across different cancers/genes. In addition, compared to the published models for genetic alterations, AMIML provided a significant improvement for predicting a wide range of genes (e.g., KMT2C, TP53, and SETD2 for KIRC; ERBB2, BRCA1, and BRCA2 for BRCA; JAK1, POLE, and MTOR for UCEC) as well as produced outstanding predictive models for other clinically relevant gene mutations, which have not been reported in the current literature. Furthermore, with the flexible and interpretable attention-based MIL pooling mechanism, AMIML could further zero-in and detect predictive image patches.
Subjects: Quantitative Methods (q-bio.QM); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Genomics (q-bio.GN)
Cite as: arXiv:2206.00455 [q-bio.QM]
  (or arXiv:2206.00455v1 [q-bio.QM] for this version)

Submission history

From: Bangwei Guo [view email]
[v1] Tue, 31 May 2022 15:45:29 GMT (2971kb)

Link back to: arXiv, form interface, contact.