CellOMaps: A Compact Representation for Robust Classification of Lung Adenocarcinoma Growth Patterns

TL;DR

This paper introduces CellOMaps, a compact representation for classifying lung adenocarcinoma growth patterns from histology images, achieving state-of-the-art accuracy and potential for predicting tumor mutational burden.

Contribution

The paper presents a novel CellOMaps representation and a generalizable machine learning pipeline for accurate classification of LUAD growth patterns from WSIs.

Findings

State-of-the-art classification accuracy on internal and external datasets.

The pipeline outperforms existing methods.

Preliminary results suggest potential for TMB prediction.

Abstract

Lung adenocarcinoma (LUAD) is a morphologically heterogeneous disease, characterized by five primary histological growth patterns. The classification of such patterns is crucial due to their direct relation to prognosis but the high subjectivity and observer variability pose a major challenge. Although several studies have developed machine learning methods for growth pattern classification, they either only report the predominant pattern per slide or lack proper evaluation. We propose a generalizable machine learning pipeline capable of classifying lung tissue into one of the five patterns or as non-tumor. The proposed pipeline's strength lies in a novel compact Cell Organization Maps (cellOMaps) representation that captures the cellular spatial patterns from Hematoxylin and Eosin whole slide images (WSIs). The proposed pipeline provides state-of-the-art performance on LUAD growth pattern classification when evaluated on both internal unseen slides and external datasets, significantly outperforming the current approaches. In addition, our preliminary results show that the model's outputs can be used to predict patients Tumor Mutational Burden (TMB) levels.