Virtual Library

Start Your Search

SeulGi You



Author of

  • +

    MA10 - Emerging Technologies for Lung Cancer Detection (ID 129)

    • Event: WCLC 2019
    • Type: Mini Oral Session
    • Track: Screening and Early Detection
    • Presentations: 1
    • +

      MA10.03 - Quantitative Computed Tomography (CT) Based Texture Analysis; Could We Predict the Future Growth of the Pure Ground Glass Nodules? (ID 229)

      15:15 - 16:45  |  Author(s): SeulGi You

      • Abstract
      • Slides

      Background

      To evaluate whether the quantitative computed tomography based texture analysis (QTA) could predict the future growth of the pure ground glass opacity nodule (GGN) or not.

      Method

      We retrieved CT images of 9284 patients who underwent chest CT in 2013 from the picture archiving and communication system (PACS). We queried the database of PACS to filter reports of chest CT containing one of these key words, as follows; “ground-glass”, ground glass”, or “GGO(s)”. 78 patients were finally included [5 patients (5GGNs) who underwent operation due to growth of GGN during follow-up and 73 patients who had 3-year-follow-up CT]. Total 90 GGNs from 78 patients were analyzed by QTA. The parameters of QTA were mean HU value, standard deviation (SD), entropy, mean positive pixels (MPP), skewness, and kurtosis. QTA was performed with image filtration step to remove photon noise, filtration technique enhanced features of different sizes based on the spatial scale filter (SSF) value varying from fine-texture (SSF2), medium-texture (SSF3), and coarse-texture (SSF4). We focused on the change of volume% of GGNs [(follow-up volume of GGN/initial volume of GGN)*100%], and assessed the differences of QTA parameters’ value according to the change of volume % for three cut-off levels (150%, 170%, and 200%); group 1a (≤130%), group 1b (>130%);group 2a (≤150%), group 2b (>150%);group 3a (≤170%), group 3b (>170%).

      patients.png

      Result

      r.png

      Only entropy was a variable that showed statistically significant difference between group 3a and 3b with all the filtrations (SSF2, 3, 4) applied or without filtration (SSF0). The mean, SD, MPP, kurtosis and skewness, showed no significant difference according to the cut-off value of volume % change (130%, 150%). There was no significant difference in QTA parameters in group2a vs 2b, group3a vs 3b.

      Conclusion

      The initial entropy parameter of texture analysis for GGNs may have the potential to predict the GGNs growth.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.