Interpretable Tile-Based Classification of Paclitaxel Exposure

TL;DR

This paper introduces a tile-based classification method for detecting paclitaxel exposure in cell images, achieving high accuracy and interpretability, which can accelerate drug discovery processes.

Contribution

The study presents a simple tiling-and-aggregation pipeline that outperforms previous models in classifying drug exposure from microscopy images, with enhanced interpretability.

Findings

Achieved around 20% improvement over baseline accuracy.

Enhanced model interpretability using Grad-CAM and attention analyses.

Validated robustness through cross-validation.

Abstract

Medical image analysis is central to drug discovery and preclinical evaluation, where scalable, objective readouts can accelerate decision-making. We address classification of paclitaxel (Taxol) exposure from phase-contrast microscopy of C6 glioma cells -- a task with subtle dose differences that challenges full-image models. We propose a simple tiling-and-aggregation pipeline that operates on local patches and combines tile outputs into an image label, achieving state-of-the-art accuracy on the benchmark dataset and improving over the published baseline by around 20 percentage points, with trends confirmed by cross-validation. To understand why tiling is effective, we further apply Grad-CAM and Score-CAM and attention analyses, which enhance model interpretability and point toward robustness-oriented directions for future medical image research. Code is released to facilitate reproduction and extension.