Multi-Scale deep learning framework for cochlea localization, segmentation and analysis on clinical ultra-high-resolution CT images
Multi-Scale deep learning framework for cochlea localization, segmentation and analysis on clinical ultra-high-resolution CT images
复制标题
DOI：
10.1016/j.cmpb.2020.105387
复制
发表时间：
2020-07-01
影响因子：
6.1
通讯作者：
Caballo, Marco
中科院分区：
文献类型：
--
作者：
Heutink, Floris;Koch, Valentin;Caballo, Marco
背景和目的：进行患者特定的、基于术前耳蜗CT的测量可能有助于在耳蜗内创伤和残余听力丧失方面积极影响耳蜗手术的结果。因此，我们提出了一种方法来自动分割和测量临床超高分辨率（UHR）CT图像中的人类耳蜗，并研究个性化植入planning.Methods的耳蜗大小的差异：123颞骨CT扫描采集两个UHR-CT扫描仪，并用于开发和验证基于深度学习的系统，用于自动耳蜗分割和测量。该分割算法由两个主要步骤（检测和逐像素分类）级联组成，旨在将多尺度计算机辅助检测方案的结果与用于逐像素分类的U-Net结构相结合。分割结果被用作测量算法的输入，该算法通过卷积神经网络和细化算法的组合使用提供自动耳蜗测量（体积、基底直径和耳蜗管长度（CDL））。通过Dice相似性、Boundary-F1（BF）评分以及最大和平均Hausdorff距离对手动注释进行自动分割验证，同时计算自动结果与相应手动获得的地面真实值之间的测量误差。最后，开发的系统用于研究我们的患者队列中耳蜗大小的差异，将测量误差与不同患者耳蜗大小的实际变化联系起来。结果：自动分割导致Dice为0.90 +/- 0.03，BF评分为0.95 +/- 0.03，并且相对于手动注释的最大和平均Hausdorff距离为3.05 +/-0.39和0.32 +/-0.07。自动耳蜗测量的误差分别为8.4%（容积）、5.5%（CDL）、7.8%（基底直径）。耳蜗的大小变化很大，范围在0.10和0.28 ml（体积），1.3和2.5 mm（基底直径），27.7和40.1 mm（CDL）。结论：该算法可以成功地分割和分析UHR-CT图像上的耳蜗，从而准确测量耳蜗解剖结构。鉴于我们的患者队列中发现的耳蜗大小差异很大，它可能会在耳蜗植入手术中作为术前工具应用，可能有助于基于患者特定的基于图像的解剖测量制定个性化治疗策略。(C)2020爱思唯尔B. V.保留所有权利。
Background and objective: Performing patient-specific, pre-operative cochlea CT-based measurements could be helpful to positively affect the outcome of cochlear surgery in terms of intracochlear trauma and loss of residual hearing. Therefore, we propose a method to automatically segment and measure the human cochlea in clinical ultra-high-resolution (UHR) CT images, and investigate differences in cochlea size for personalized implant planning.Methods: 123 temporal bone CT scans were acquired with two UHR-CT scanners, and used to develop and validate a deep learning-based system for automated cochlea segmentation and measurement. The segmentation algorithm is composed of two major steps (detection and pixel-wise classification) in cascade, and aims at combining the results of a multi-scale computer-aided detection scheme with a U-Net-like architecture for pixelwise classification. The segmentation results were used as an input to the measurement algorithm, which provides automatic cochlear measurements (volume, basal diameter, and cochlear duct length (CDL)) through the combined use of convolutional neural networks and thinning algorithms. Automatic segmentation was validated against manual annotation, by the means of Dice similarity, Boundary-F1 (BF) score, and maximum and average Hausdorff distances, while measurement errors were calculated between the automatic results and the corresponding manually obtained ground truth on a per-patient basis. Finally, the developed system was used to investigate the differences in cochlea size within our patient cohort, to relate the measurement errors to the actual variation in cochlear size across different patients.Results: Automatic segmentation resulted in a Dice of 0.90 +/- 0.03, BF score of 0.95 +/- 0.03, and maximum and average Hausdorff distance of 3.05 +/- 0.39 and 0.32 +/- 0.07 against manual annotation. Automatic cochlear measurements resulted in errors of 8.4% (volume), 5.5% (CDL), 7.8% (basal diameter). The cochlea size varied broadly, ranging between 0.10 and 0.28 ml (volume), 1.3 and 2.5 mm (basal diameter), and 27.7 and 40.1 mm (CDL).Conclusions: The proposed algorithm could successfully segment and analyze the cochlea on UHR-CT images, resulting in accurate measurements of cochlear anatomy. Given the wide variation in cochlear size found in our patient cohort, it may find application as a pre-operative tool in cochlear implant surgery, potentially helping elaborate personalized treatment strategies based on patient-specific, image-based anatomical measurements. (C) 2020 Elsevier B.V. All rights reserved.