当前位置： SCI文献检索 > PHYSICS IN MEDICINE AND BIOLOGY期刊下所有文献 > Regional deep hyperthermia: impact of observer variability in CT-based manual tissue segmentation on simulated temperature distribution.

Regional deep hyperthermia: impact of observer variability in CT-based manual tissue segmentation on simulated temperature distribution.

Abstract：

:The aim of this study was to systematically investigate the influence of the inter- and intra-observer segmentation variation of tumors and organs at risk on the simulated temperature coverage of the target. CT scans of six patients with tumors in the pelvic region acquired for radiotherapy treatment planning were used for hyperthermia treatment planning. To study the effect of inter-observer variation, three observers manually segmented in the CT images of each patient the following structures: fat, muscle, bone and the bladder. The gross tumor volumes (GTV) were contoured by three radiation oncology residents and used as the hyperthermia target volumes. For intra-observer variation, one of the observers of each group contoured the structures of each patient three times with a time span of one week between the segmentations. Moreover, the impact of segmentation variations in organs at risk (OARs) between the three inter-observers was investigated on simulated temperature distributions using only one GTV. The spatial overlap between individual segmentations was assessed by the Dice similarity coefficient (DSC) and the mean surface distance (MSD). Additionally, the temperatures T90/T10 delivered to 90%/10% of the GTV, respectively, were assessed for each observer combination. The results of the segmentation similarity evaluation showed that the DSC of the inter-observer variation of fat, muscle, the bladder, bone and the target was 0.68 ± 0.12, 0.88 ± 0.05, 0.73 ± 0.14, 0.91 ± 0.04 and 0.64 ± 0.11, respectively. Similar results were found for the intra-observer variation. The MSD results were similar to the DSCs for both observer variations. A statistically significant difference (p < 0.05) was found for T90 and T10 in the predicted target temperature due to the observer variability. The conclusion is that intra- and inter-observer variations have a significant impact on the temperature coverage of the target. Furthermore, OARs, such as bone and the bladder, may essentially influence the homogeneity of the simulated target temperature distribution.

journal_name

Phys Med Biol

journal_title

Physics in medicine and biology

authors

Aklan B,Hartmann J,Zink D,Siavooshhaghighi H,Merten R,Putz F,Ott O,Fietkau R,Bert C

doi

10.1088/1361-6560/aa685b

subject

Has Abstract

pub_date

2017-05-08 00:00:00

pages

4479-4495

issue

eissn

0031-9155

issn

1361-6560

journal_volume

pub_type

杂志文章

在线工具

Regional deep hyperthermia: impact of observer variability in CT-based manual tissue segmentation on simulated temperature distribution.