Efficient Deep Network Architectures for Fast Chest X-Ray Tuberculosis Screening and Visualization

By F. Pasa, Vladimir Golkov, Franz Pfeiffer, Daniel Cremers, Daniela Pfeiffer. Scientific Reports, 2019

351 citations recorded by the source corpus.

Automated diagnosis of tuberculosis (TB) from chest X-Rays (CXR) has been tackled with either hand-crafted algorithms or machine learning approaches such as support vector machines (SVMs) and convolutional neural networks (CNNs). Most deep neural network applied to the task of tuberculosis diagnosis have been adapted from natural image classification. These models have a large number of parameters as well as high hardware requirements, which makes them prone to overfitting and harder to deploy in mobile settings. We propose a simple convolutional neural network optimized for the problem which is faster and more efficient than previous models but preserves their accuracy. Moreover, the visualization capabilities of CNNs have not been fully investigated. We test saliency maps and grad-CAMs as tuberculosis visualization methods, and discuss them from a radiological perspective.

Subfields: Artificial Intelligence, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/809f8352133dbeb880e2df8c122a00f6d11928b1

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Maithra Raghu
2. Chiyuan Zhang
3. Jon M. Kleinberg
4. Samy Bengio
5. Mohammad Tariqul Islam
6. Tawsifur Rahman
7. Amith Abdullah Khandakar
8. Muhammad Enamul Hoque Chowdhury

In Radiology, Nuclear Medicine and Imaging

1. Mohammad Tariqul Islam
2. Muhammad Abdul Kadir
3. Tawsifur Rahman
4. Khandaker F. Islam
5. Khandaker Reajul Islam
6. Samy Bengio
7. Amith Abdullah Khandakar
8. Muhammad Enamul Hoque Chowdhury

Citations in the corpus

• Transfusion: Understanding Transfer Learning for Medical Imaging (2019, background): “This line of investigation is especially important in the medical setting, where large, computationally expensive models might signi cantly impede mobile and on-device applications (Pasa et al., 2019).”
• Reliable Tuberculosis Detection Using Chest X-Ray With Deep Learning, Segmentation and Visualization (2020, background): “[35] proposed a deep network architecture optimized for the screening of tuberculosis with an accuracy of 86.”