loading page

Quantitative lung microstructure mapping with X-ray dark-field grating interferometry on a laboratory source
  • +4
  • Matteo Abis,
  • Goran Lovrić,
  • Johannes Schittny,
  • Marco Stampanoni,
  • Goran Lovric,
  • Tiziana Cremona,
  • Charaf Benarafa
Matteo Abis
Paul Scherrer Institut, ETH Zurich
Author Profile
Goran Lovrić
Paul Scherrer Institut, ETH Zurich
Johannes Schittny
University of Bern
Marco Stampanoni
Paul Scherrer Institut, ETH Zurich
Goran Lovric
Swiss Light Source, Paul Scherrer Institute, 5234 Villigen, Switzerland, Centre d'Imagerie BioMédicale, École Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerland
Author Profile
Tiziana Cremona
Charaf Benarafa


A wide range of pulmonary diseases affect the lungs microstructure, i.e. the alveolar epithelium, where diffusion-driven gas exchange takes place.  Thereby, early and definite diagnosis is not only important to prevent treatment delays, but also to counteract acute exacerbation and initiate treatment before reaching an irreversible state. In a number of works, dark-field grating interferometry has been identified as the most promising technique in the screening of disease-induced changes to the lungs microstucture, most notably in the early diagnosis of pulmonary emphysema. Particularly the dependency of the small-angle scattering (dark-field) signal on the mean alveolar size has the potential to improve diagnosis. On the other hand, it has also been shown that the dark-field signal can be strongly biased and system-dependent when acquired on a conventional X-ray tube. In the present work, we present a full model-driven framework for quantitatively mapping the lungs microstructure to the dark-field image. Starting from first principles, our quantitative model for dark-field radiography on a laboratory source is verified  with high-resolution (sub-micron scale) tomographic volumes of the same samples. We show that our model is capable of establishing a direct correlation between microstructure and the macroscopic X-ray radiographic image. This provides a deeper understanding of dark-field imaging that can be beneficial for the design of an interferometer as a specific diagnostic tool, as well as for similar analyses in different fields of foam dynamics and material sciences. (Goran: last sentence I would still adapt depending on what really the conclusion is: ideally already suggest an idealistic interferometer and/or imaging technique...)