Search

EP-4420078-B1 - SYSTEM AND METHOD FOR EVALUATING A PULMONARY VENTILATION AND PERFUSION GRADIENT

EP4420078B1EP 4420078 B1EP4420078 B1EP 4420078B1EP-4420078-B1

Inventors

  • WIEMKER, RAFAEL
  • SABCZYNSKI, Jörg
  • HENDRIKS, CORNELIS PETRUS

Dates

Publication Date
20260513
Application Date
20221012

Claims (15)

  1. A processing system (130) for evaluating a pulmonary ventilation and perfusion gradient of at least one lung of a subject (110), the processing system being configured to: receive CT imaging data (125) containing a representation of at least one lung of the subject; perform segmentation on the CT imaging data to identify the representation of at least one lung of the subject; generate a plurality of local histograms of Hounsfield density frequency distribution by processing the CT imaging data, wherein each of the plurality of local histograms is representative of a different region of the at least one lung represented in the CT imaging data; generate a global histogram of Hounsfield density frequency distribution for the at least one lung of the subject; and for each of the plurality of local histograms: generate a plurality of shifted local histograms, each shifted local histogram being a version of the local histogram shifted by different shift value; cross-correlate each shifted local histogram with the global histogram to produce a respective plurality of correlation values; and process the shift values and the correlation values to produce a final shift value, wherein the final shift value represents a difference between the local histogram and the global histogram, and wherein an overall range in magnitudes of the final shift values is representative of a magnitude of the pulmonary ventilation and perfusion gradient.
  2. The processing system (130) of claim 1, wherein the final shift value for each of the plurality of local histograms is the shift value corresponding to the correlation value having the maximum value amongst the plurality of correlation values of the local histogram.
  3. The processing system (130) of claim 1, wherein the final shift value for each of the plurality of local histograms is a correlation-weighted mean shift value.
  4. The processing system (130) of any of claims 1 to 3, further configured to provide, at a user interface (140), a visualization of the final shift values.
  5. The processing system (130) of claim 4, wherein negative final shift values are represented by a first color, and positive final shift values are represented by a second, different color.
  6. The processing system (130) of claim 4 or 5, wherein the visualization of the final shift values is generated by: generating, for each local histogram, a visualization of the maximum correlation values for that local histogram; and generating an overlay representative of the final shift values on the visualization of the maximum correlation values.
  7. The processing system (130) of any of claims 1 to 6, wherein shift values are in a predetermined range, for instance, from between -200 Hounsfield units and +200 Hounsfield units.
  8. The processing system (130) of any of claims 1 to 7, wherein the plurality of local histograms comprise ray-wise histograms in a lateral direction and/or an axial, i.e. cranio-caudal, direction.
  9. The processing system (130) of any of claims 1 to 7, wherein the plurality of local histograms comprise two-dimensional patch-wise histograms in a coronal plane.
  10. The processing system (130) of any of claims 1 to 7, wherein each of the plurality of local histograms is generated by: generating a plurality of sub-local histograms, wherein each of the sub-local histograms is representative of a different subregion of the region of the at least one lung corresponding to the local histogram; and cumulating the sub-local histograms into the local histogram.
  11. The processing system (130) of any of claims 1 to 10, wherein the global histogram is generated by cumulating the local histograms into the global histogram.
  12. A system (100) for evaluating at least one lung of a subject (110), the system comprising: a CT scanner (120) configured to produce CT imaging data (125); and the processing system (130) of any of claims 1 to 11, configured to receive the CT imaging data from the CT scanner.
  13. The system (100) of claim 12, wherein the processing system (130) is configured to generate a visualization of the final shift values, and wherein the system further comprises: a user interface (140), configured to receive, from the processing system, and display the generated visualization of the final shift values.
  14. A computer-implemented method (300) for evaluating a pulmonary ventilation and perfusion gradient of at least one lung of a subject (110), the computer-implemented method comprising: receiving CT imaging data (125) containing a representation of at least one lung of the subject; performing segmentation on the CT imaging data to identify the representation of at least one lung of the subject; generating a plurality of local histograms of Hounsfield density frequency distribution by processing the CT imaging data, wherein each of the plurality of local histograms is representative of a different region of the at least one lung represented in the CT imaging data; generating a global histogram of Hounsfield density frequency distribution for the at least one lung of the subject; and for each of the plurality of local histograms: generating a plurality of shifted local histograms, each shifted local histogram being a version of the local histogram shifted by different shift value; cross-correlating each shifted local histogram with the global histogram to produce a respective plurality of correlation values; and processing the shift values and the correlation values to produce a final shift value, wherein the final shift value represents a difference between the local histogram and the global histogram, and wherein an overall range in magnitudes of the final shift values is representative of a magnitude of the pulmonary ventilation and perfusion gradient.
  15. A computer program product comprising computer program code means which, when executed on a computer device having a processing system, cause the processing system to perform all of the steps of the method (300) according to claim 14.

Description

FIELD OF THE INVENTION The invention relates to the field of evaluating computed tomography (CT) images. BACKGROUND OF THE INVENTION Gravity can cause a pulmonary ventilation and perfusion gradient between ventral and dorsal regions of the lungs, especially in the lungs of a subject who has been lying down for a long period. The potential adverse effects caused by this gradient should be taken into account when treating patients with acute respiratory syndrome (ARDS), which is often caused by infectious pulmonary diseases such as COVID-19, by, for example, applying mechanical ventilation while the patient is in a prone position in addition to the standard supine position. However, the presence and magnitude of a gravity-caused gradient can vary widely between patients. Some diseases increase or decrease the magnitude of this gradient: for example, patients with systemic sclerosis-related vasculopathy show a reduced gravity-dependent attenuation effect, due to the reduced elastic compliance of the pulmonary arterial tree. Several diffuse lung diseases (e.g. hypersensitivity pneumosis, cellular non-specific interstitial pneumonia, and subacute diffuse alveolar damage) show a subtle, relatively homogenous increase in lung attenuation, which may overlap with the gravity-induced gradient. The pulmonary ventilation and perfusion gradient in a patient is visible, in three-dimensional images from a CT chest scan, as a subtle shift in overall Hounsfield densities between ventral and dorsal lung regions (with dorsal regions having higher Hounsfield densities when a subject is in a supine position). However, the effect is difficult to quantify objectively. Further, the pulmonary parenchyma surrounds regular anatomical structures (such as vessels, bronchi, lymph nodes and the heart). In patients with pulmonary diseases, disease lesions, effusions, atelectasis, mesothelioma and scars may be embedded in the pulmonary parenchyma. This "anatomical noise" means that there are far greater variations in Hounsfield density in a CT image of the lungs than the variation caused by the pulmonary ventilation and perfusion gradient. Simple quantitative measures, such as regional average Hounsfield density, cannot therefore be used to provide information about the ventro-dorsal gravity effect. Treatment decisions for patients with acute respiratory syndrome often need to be made urgently in an emergency situation, in which there is not sufficient time to take careful manual measurements to determine information about the pulmonary ventilation and perfusion gradient. There is therefore a need for improved information about the pulmonary ventilation and perfusion gradient between ventral and dorsal regions of a subject's lungs. US 2021/065361 A1 discloses a method and system for determining regions of hyperdense lung parenchyma in an image of a lung. WO 2020/231904 A1 discloses a method for imaging a lung of a patient. Mulreany, D G et al., "Volumetric xenon-CT imaging of conventional and high-frequency oscillatory ventilation", Academic Radiology, 16(6):718-725, 1 June 2009 discloses obtaining high-resolution volumetric ventilation maps of a lung. Kircher, Michael et al., "Regional lung perfusion analysis in experimental ARDS by electrical impedance and computed tomography", IEEE Transactions on Medical Imaging, 40(1):251-261, 21 September 2020 discloses a method for detecting pulmonary diffusion based on indicator-enhanced electrical impedance tomography. Gunnar, Elke et al., "Quantification of ventilation distribution in regional lung injury by electrical impedance tomography and xenon computed tomography", Physiological Measurement, 34(10):1303-1318, 11 September 2013 discloses an electrical impedance tomography based assessment of regional ventilation under pathological conditions. SUMMARY OF THE INVENTION The invention is defined by the claims. According to examples in accordance with an aspect of the invention, there is provided a processing system for evaluating a pulmonary ventilation and perfusion gradient of at least one lung of a subject. The processing system is configured to: receive CT imaging data containing a representation of at least one lung of the subject; perform segmentation on the CT imaging data to identify the representation of at least one lung of the subject; generate a plurality of local histograms of Hounsfield density frequency distribution by processing the CT imaging data, wherein each of the plurality of local histograms is representative of a different region of the at least one lung represented in the CT imaging data; generate a global histogram of Hounsfield density frequency distribution for the at least one lung of the subject; and, for each of the plurality of local histograms: generate a plurality of shifted local histograms, each shifted local histogram being a version of the local histogram shifted by different shift value; cross-correlate each shifted local histogram with the global histogram