Abstract：
OBJECTIVE: To validate how an automated model for vestibular schwannoma (VS) segmentation developed on an external homogeneous dataset performs when applied to internal heterogeneous data.
METHODS: The external dataset comprised 242 patients with previously untreated, sporadic unilateral VS undergoing Gamma Knife radiosurgery, with homogeneous magnetic resonance imaging (MRI) scans. The internal dataset comprised 10 patients from our institution, with heterogeneous MRI scans.
METHODS: An automated VS segmentation model was developed on the external dataset. The model was tested on the internal dataset.
METHODS: Dice score, which measures agreement between ground truth and predicted segmentations.
RESULTS: When applied to the internal patient scans, the automated model achieved a mean Dice score of 61% across all 10 images. There were three tumors that were not detected. These tumors were 0.01 ml on average (SD = 0.00 ml). The mean Dice score for the seven tumors that were detected was 87% (SD = 14%). There was one outlier with Dice of 55%-on further review of this scan, it was discovered that hyperintense petrous bone had been included in the tumor segmentation.
CONCLUSIONS: We show that an automated segmentation model developed using a restrictive set of siloed institutional data can be successfully adapted for data from different imaging systems and patient populations. This is an important step toward the validation of automated VS segmentation. However, there are significant shortcomings that likely reflect limitations of the data used to train the model. Further validation is needed to make automated segmentation for VS generalizable.
METHODS: The external dataset comprised 242 patients with previously untreated, sporadic unilateral VS undergoing Gamma Knife radiosurgery, with homogeneous magnetic resonance imaging (MRI) scans. The internal dataset comprised 10 patients from our institution, with heterogeneous MRI scans.
METHODS: An automated VS segmentation model was developed on the external dataset. The model was tested on the internal dataset.
METHODS: Dice score, which measures agreement between ground truth and predicted segmentations.
RESULTS: When applied to the internal patient scans, the automated model achieved a mean Dice score of 61% across all 10 images. There were three tumors that were not detected. These tumors were 0.01 ml on average (SD = 0.00 ml). The mean Dice score for the seven tumors that were detected was 87% (SD = 14%). There was one outlier with Dice of 55%-on further review of this scan, it was discovered that hyperintense petrous bone had been included in the tumor segmentation.
CONCLUSIONS: We show that an automated segmentation model developed using a restrictive set of siloed institutional data can be successfully adapted for data from different imaging systems and patient populations. This is an important step toward the validation of automated VS segmentation. However, there are significant shortcomings that likely reflect limitations of the data used to train the model. Further validation is needed to make automated segmentation for VS generalizable.
摘要：
目的：验证在外部同质数据集上开发的前庭神经鞘瘤（VS）分割自动模型在应用于内部异构数据时的性能。
方法：外部数据集包括242名先前未经治疗的患者，接受伽玛刀放射外科治疗的零星单侧VS，使用均匀磁共振成像（MRI）扫描。内部数据集包括我们机构的10名患者，不同类型的MRI扫描。
方法：在外部数据集上开发了自动VS分割模型。在内部数据集上对模型进行了测试。
方法：骰子得分，衡量地面真相和预测分割之间的一致性。
结果：应用于患者内部扫描时，自动化模型在所有10张图像中获得了61%的平均骰子得分。有三个未检测到的肿瘤。这些肿瘤平均为0.01ml(SD=0.00ml)。检测到的七个肿瘤的平均Dice评分为87%(SD=14%)。在进一步审查这次扫描时，有一个异常值，骰子为55%，发现高强度的岩骨已包括在肿瘤分割中。
结论：我们表明，使用限制性的孤立机构数据集开发的自动分割模型可以成功地适应来自不同成像系统和患者人群的数据。这是朝着验证自动VS分割迈出的重要一步。然而,有明显的缺陷，可能反映了用于训练模型的数据的局限性。需要进一步验证以使VS的自动分割变得可概括。
方法：外部数据集包括242名先前未经治疗的患者，接受伽玛刀放射外科治疗的零星单侧VS，使用均匀磁共振成像（MRI）扫描。内部数据集包括我们机构的10名患者，不同类型的MRI扫描。
方法：在外部数据集上开发了自动VS分割模型。在内部数据集上对模型进行了测试。
方法：骰子得分，衡量地面真相和预测分割之间的一致性。
结果：应用于患者内部扫描时，自动化模型在所有10张图像中获得了61%的平均骰子得分。有三个未检测到的肿瘤。这些肿瘤平均为0.01ml(SD=0.00ml)。检测到的七个肿瘤的平均Dice评分为87%(SD=14%)。在进一步审查这次扫描时，有一个异常值，骰子为55%，发现高强度的岩骨已包括在肿瘤分割中。
结论：我们表明，使用限制性的孤立机构数据集开发的自动分割模型可以成功地适应来自不同成像系统和患者人群的数据。这是朝着验证自动VS分割迈出的重要一步。然而,有明显的缺陷，可能反映了用于训练模型的数据的局限性。需要进一步验证以使VS的自动分割变得可概括。
