Quantitative comparison between single-photon emission computed tomography and positron emission tomography imaging of lung ventilation with 99mTc-technegas and 68Ga‑gallgas in patients with chronic obstructive pulmonary disease: a pilot study

Cuña Rodríguez, Enrique Gustavo - Gambini, Juan Pablo - Servente, Liliana - Savio, Eduardo - Engler, Henry William - González Sprinberg, Gabriel Adrián - Alonso, Omar

Resumen:

The aim of this study was quantitative comparison between 68Ga‑Gallgas positron emission tomography (PET) and 99mTc‑Technegas single photon emission computed tomography (SPECT) for lung ventilation function assessment in patients with moderate‑to‑severe obstructive pulmonary disease and to identify image‑derived texture features correlating to the physiologic parameters. Five patients with moderate‑to‑severe chronic obstructive pulmonary disease with PET and SPECT lung ventilation scans were selected for this study. Threshold‑based segmentations were used to compare ventilated regions between both imaging techniques. Histograms of both scans were compared to reveal main differences in distributions of radiotracers. Volumes of segmentation as well as 50 textural features measured in the pulmonary region were correlated to the forced expiratory volume in 1 s (FEV1) as the relevant physiological variable. A better peripheral distribution of the radiotracer was observed in PET scans for three out of five patients. A segmentation threshold of 27% and 31% for normalized scans, for PET and SPECT respectively, was found optimal for volume correlation with FEV1. A high correlation (Pearson correlation coefficient >0.9) was found between 16 texture features measured from SPECT and 7 features measured from PET and FEV1. Quantitative measurements revealed different tracer distribution in both techniques. These results suggest that tracer distribution patterns may depend on the cause of the pulmonary obstruction. We found several texture features measured from SPECT to correlate to FEV1.


Detalles Bibliográficos
2019
Gallgas
Quantification
Technegas
Texture features
Inglés
Universidad de la República
COLIBRI
https://hdl.handle.net/20.500.12008/30285
Acceso abierto
Licencia Creative Commons Atribución (CC - By 4.0)