Dynamic chest radiography: clinical validation of ventilation and perfusion metrics derived from changes in radiographic lung density compared to nuclear medicine imaging

Rie Tanaka; Isao Matsumoto; Masaya Tamura; Munehisa Takata; Shuhei Yoshida; Daisuke Saito; Yusuke Tanaka; Dai Inoue; Noriyuki Ohkura; Kazuo Kasahara

doi:10.21037/qims-20-1217

Dynamic chest radiography: clinical validation of ventilation and perfusion metrics derived from changes in radiographic lung density compared to nuclear medicine imaging

Quant Imaging Med Surg. 2021 Sep;11(9):4016-4027. doi: 10.21037/qims-20-1217.

Authors

Affiliations

¹ College of Medical, Pharmaceutical & Health Sciences, Kanazawa University, Kanazawa, Ishikawa, Japan.
² Department of Thoracic Surgery, Kanazawa University, Kanazawa, Ishikawa, Japan.
³ Department of Radiology, Kanazawa University Hospital, Kanazawa, Ishikawa, Japan.
⁴ Department of Respiratory Medicine, Kanazawa University Hospital, Kanazawa, Ishikawa, Japan.

Abstract

Background: Dynamic chest radiography (DCR) is a type of non-contrast-enhanced functional lung imaging with a dynamic flat-panel detector (FPD). This study aimed to assess the clinical significance of ventilation and perfusion metrics derived from changes in radiographic lung density on DCR in comparison to nuclear medicine imaging-derived metrics.

Methods: DCR images of 42 lung cancer patients were sequentially obtained during respiration using a dynamic FPD imaging system. For each subdivided lung region, the maximum change in the averaged pixel value (Δ_max), i.e., lung density, due to respiration and cardiac function was calculated, and the percentage of Δ_max relative to the total of all lung regions (Δ_max%) was computed for ventilation and perfusion, respectively. The Δ_max% was compared to the accumulation of radioactive agents such as Tc-99m gas and Tc-99m macro-aggregated albumin (radioactive agents%) on ventilation and perfusion scans in the subdivided lung regions, by Spearman's correlation coefficient (r) and the Dice similarity coefficients (DSC). To facilitate visual evaluation, Δ_max% was visualized as a color scaling, where larger Δ_max values were indicated by higher color intensities.

Results: We found a moderate correlation between Δ_max% and radioactive agents% on ventilation and perfusion scans, with perfusion metrics (r=0.57, P<0.001) showing a higher correlation than ventilation metrics (r=0.53, P<0.001). We also found a good or strong correlation (r≥0.5) in 80.9% (34/42) of patients for perfusion metrics (r=0.60±0.16) and in 52.4% (22/42) of patients for ventilation metrics (r=0.53±0.16). DSC indicated a moderate correlation for both metrics. Decreased pulmonary function was observed in the form of reduced color intensities on color-mapping images.

Conclusions: DCR-derived ventilation and perfusion metrics correlated reasonably well with nuclear medicine imaging findings in lung subdivisions, suggesting that DCR could provide useful information on pulmonary function without the use of radioactive contrast agents.

Keywords: Diagnostic imaging; perfusion; radiography; thorax; ventilation.