Taming Domain Shift in Multi-source CT-Scan Classification via Input-Space Standardization

TL;DR

This paper presents a preprocessing pipeline combining spatial-slice feature learning and kernel-density slice sampling to standardize input CT scans, effectively reducing domain shifts and improving multi-source classification performance.

Contribution

The study introduces a novel input-space standardization method that enhances cross-source generalization in multi-source CT classification, validated through empirical experiments and challenge success.

Findings

Consistent performance improvements across architectures.

Effective reduction of inter-source variance.

First place in a competitive challenge.

Abstract

Multi-source CT-scan classification suffers from domain shifts that impair cross-source generalization. While preprocessing pipelines combining Spatial-Slice Feature Learning (SSFL++) and Kernel-Density-based Slice Sampling (KDS) have shown empirical success, the mechanisms underlying their domain robustness remain underexplored. This study analyzes how this input-space standardization manages the trade-off between local discriminability and cross-source generalization. The SSFL++ and KDS pipeline performs spatial and temporal standardization to reduce inter-source variance, effectively mapping disparate inputs into a consistent target space. This preemptive alignment mitigates domain shift and simplifies the learning task for network optimization. Experimental validation demonstrates consistent improvements across architectures, proving the benefits stem from the preprocessing itself. The approach's effectiveness was validated by securing first place in a competitive challenge, supporting input-space standardization as a robust and practical solution for multi-institutional medical imaging.