# SAPSAM - Sparsely Annotated Pathological Sign Activation Maps - A novel approach to train Convolutional Neural Networks on lung CT scans using binary labels only

## TL;DR

This paper introduces a novel CNN training method using only regional labels to detect and localize lung pathological signs in CT scans, aiding CPA diagnosis and follow-up.

## Contribution

It presents a new approach to train CNNs with binary labels only, enabling spatial localization of pathological signs in lung CT scans.

## Key findings

- High classification accuracy on 352 subjects
- Precise localization of pathological signs
- Predictive power for 2-year survival

## Abstract

Chronic Pulmonary Aspergillosis (CPA) is a complex lung disease caused by infection with Aspergillus. Computed tomography (CT) images are frequently requested in patients with suspected and established disease, but the radiological signs on CT are difficult to quantify making accurate follow-up challenging. We propose a novel method to train Convolutional Neural Networks using only regional labels on the presence of pathological signs, to not only detect CPA, but also spatially localize pathological signs. We use average intensity projections within different ranges of Hounsfield-unit (HU) values, transforming input 3D CT scans into 2D RGB-like images. CNN architectures are trained for hierarchical tasks, leading to precise activation maps of pathological patterns. Results on a cohort of 352 subjects demonstrate high classification accuracy, localization precision and predictive power of 2 year survival. Such tool opens the way to CPA patient stratification and quantitative follow-up of CPA pathological signs, for patients under drug therapy.