MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

Assessment of Respiratory Excursion of Internal Mammary Nodes for ITV Determination

M Jermoumi1*, S James1 , C Ferguson1 , J Barrett1 , C Huh1 , A Al-Basheer1 , (1) Georgia Cancer Center, Augusta University, Augusta, GA,

Presentations

(Tuesday, 7/31/2018) 10:30 AM - 11:00 AM

Room: Exhibit Hall | Forum 9

Purpose: The purpose of this work is to identify the appropriate margin for the Internal Target Volume (ITV) and to assess the respiratory motion for internal mammary node (IMN) chain delineation in locally advanced breast cancer using 4DCT scan.

Methods: Two board certified radiation oncologists participated in this work to delineate the IMN target of 10 consecutive patients who underwent treatment for locally advanced breast cancer. The image data sets were acquired using 4DCT to take into account the respiratory motion. The ITV margin was identified by using all 10 respiratory phases and the AV-CT scan. The inter-observer variability was quantified by evaluating spheroid target volume, target shape, and dice coefficient index (DCI).

Results: The maximum ITV margin was 5 mm. High inter-observer variability was shown for target volume and eccentricity feature. For both clinicians, the mean of target volume (SD) were 4.29 cm³ (±1.53) and 2.14 cm³ (±0.53) with p value of 0.005 and the mean of eccentricity were 0.73 (±0.13) and 0.73 (±0.09) with p value of 0.04. The mean (SD) of the similarity agreement between the two manual delineations was 0.65 (±0.16). In seventy percent of cases, the DCI showed good agreement.

Conclusion: In this work, we investigated ITV determination and the inter-observer variability in contouring the clinical target volume of IMN. 5 mm was shown could be reasonable margin to be added to the clinical target volume. Although the dice coefficient index showed good agreement, significant differences for spheroid target volume and target shape were observed between two observers. Future work will include semi-automatic contouring to reduce this variability.

Keywords

Breast, Image Analysis

Taxonomy

IM/TH- image segmentation: CT

Contact Email