The purpose of this work was to characterize expert variation in segmentation of intracranial structures pertinent to radiation therapy, and to assess a registration-driven atlas-based segmentation algorithm in that context. Eight experts were recruited to segment the brainstem, optic chiasm, optic nerves, and eyes, of 20 patients who underwent therapy for large space-occupying tumors. Performance variability was assessed through three geometric measures: volume, Dice similarity coefficient, and Euclidean distance. In addition, two simulated ground truth segmentations were calculated via the simultaneous truth and performance level estimation (STAPLE) algorithm and a novel application of probability maps. The experts and automatic system were found to generate structures of similar volume, though the experts exhibited higher variation with respect to tubular structures. No difference was found between the mean Dice coefficient (DSC) of the automatic and expert delineations as a group at a 5% significance level over all cases and organs. The larger structures of the brainstem and eyes exhibited mean DSC of approximately 0.8–0.9, whereas the tubular chiasm and nerves were lower, approximately 0.4–0.5. Similarly low DSC have been reported previously without the context of several experts and patient volumes. This study, however, provides evidence that experts are similarly challenged. The average maximum distances (maximum inside, maximum outside) from a simulated ground truth ranged from (−4.3, +5.4) mm for the automatic system to (−3.9, +7.5) mm for the experts considered as a group. Over all the structures in a rank of true positive rates at a 2 mm threshold from the simulated ground truth, the automatic system ranked second of the nine raters. This work underscores the need for large scale studies utilizing statistically robust numbers of patients and experts in evaluating quality of automatic algorithms.

Segmenting the thyroid gland in head and neck CT images is of vital clinical significance in designing intensity-modulated radiation therapy (IMRT) treatment plans. In this work, we evaluate and compare several multiple-atlas-based methods to segment this structure. Using the most robust method, we generate automatic segmentations for the thyroid gland and study their clinical applicability. The various methods we evaluate range from selecting one single atlas based on one of three similarity measures, to combining the segmentation results obtained with several atlases and weighting their contribution using techniques including a simple majority vote rule, a technique called STAPLE that is widely used in the medical imaging literature, and the similarity between the atlas and the volume to be segmented. We show that the best results are obtained when several atlases are combined and their contributions are weighted with a measure of similarity between each atlas and the volume to be segmented. We also show that with our data set, STAPLE does not always lead to the best results. Automatic segmentations generated by the combination method using the correlation coefficient (CC) between the deformed atlas and the patient volume, which is the most accurate and robust method we evaluated, are presented to a physician as 2D contours and modified to meet clinical requirements. It is shown that about 40% of the contours of the left thyroid and about 42% of the right thyroid can be used directly. An additional 21% on the left and 24% on the right require only minimal modification. The amount and the location of the modifications are qualitatively and quantitatively assessed. We demonstrate that, although challenged by large inter-subject anatomical discrepancy, atlas-based segmentation of the thyroid gland in IMRT CT images is feasible by involving multiple atlases. The results show that a weighted combination of segmentations by atlases using the CC as the similarity measure slightly outperforms standard combination methods, e.g., the majority vote rule and STAPLE, as well as methods selecting one single most similar atlas. Results we have obtained suggest that using our contours as initial contours to be edited has clinical value.

Purpose: We have developed a novel method for automatic segmentation of critical structures in the brain. The purpose of this study is to test the feasibility of this method as an alternative to manually‐derived physician contours. We test feasibility by evaluating the dosimetric consequences of auto‐segmentation versus physician‐drawn contours. Method and Materials: Brainstem, eyes, optic nerves and chiasm were segmented through non‐rigid registration of CT and MR‐based atlases to two patients. Patient A presents a challenging case in which a base of skull chondrosarcoma distorts normal brainstem anatomy. Patient B suffers from parotid disease and presents normal critical structure anatomy. Intensity‐modulated radiosurgery treatment plans were derived from physician contours and applied to the automatic contours. Results: For patient B (tumor far from critical structures) calculated doses for manual and automatic contours were within 2% of tumor dose for a given volume. Dose to the eyes, optic nerves, and chiasm of patient A were similar in agreement to those of patient B. The maximum dose to the brainstem of patient A, however, was 13% higher for the automatic contour. These dose differences were clinically negligible for all structures except the brainstem of patient A, in which case the difference was significant but acceptable. Conclusion: Clinical incorporation of our automated method is shown to be feasible dosimetrically. For the tumor lying far from the critical structures, the dose differences between automatically and manually‐derived contours were insignificant. The differences increased for the case in which a critical structure lay directly adjacent to a large tumor. These cases illustrate the system is accurate for critical structures far from the lesion but sensitive to local disturbance and inherently steeper dose gradients when the critical structures lie near the lesion.

In the first paragraph on p 95, '(Wells et al 1996)' should read '(Wells et al 1996, Maes et al 1997)'.In the last paragraph on p 96, the citation 'Maes et al 1997' should be removed.