Shanghai Institute of Medical Imaging, Shanghai, China.
Department of Radiology, Huashan Hospital, Fudan University, 12 Middle Urumqi Road, Jing'an District, Shanghai, 200040, China.
Eur Radiol. 2021 Jul;31(7):5012-5020. doi: 10.1007/s00330-020-07558-2. Epub 2021 Jan 6.
To evaluate for the first time the performance of a deep learning method based on no-new-Net for fully automated segmentation and volumetric measurements of intracerebral hemorrhage (ICH), intraventricular extension of intracerebral hemorrhage (IVH), and perihematomal edema (PHE) in primary ICH on CT.
Three hundred and eighty primary ICH patients who underwent CT at hospital arrival were divided into a training cohort (n = 300) and a validation cohort (n = 80). An independent cohort with 80 patients was used for testing. Ground truth (segmentation masks) was manually generated by radiologists. Model performance on lesion segmentation and volumetric measurement of ICH, IVH, and PHE were evaluated by comparing the model results with the segmentations performed by radiologists.
In the test cohort, the Dice scores of lesion segmentation were 0.92, 0.79, and 0.71 for ICH, IVH, and PHE, respectively. The sensitivities were 0.93 for ICH, 0.88 for IVH, and 0.81 for PHE. The positive predictive values were 0.92, 0.76, and 0.69 for ICH, IVH, and PHE, respectively. Excellent concordance (concordance correlation coefficients [CCCs] ≥ 0.98) of ICH and IVH and good concordance of PHE (CCCs ≥ 0.92) were demonstrated between manually and automatically measured volumes. The model took approximately 15 s to provide automatic segmentation and volume analysis for each patient.
Our model demonstrates good reliability for automatic segmentation and volume measurement of ICH, IVH, and PHE in primary ICH, which can be useful to reduce the effort and time of doctors to calculate volumes of ICH, IVH, and PHE.
• Deep learning algorithms can provide automatic and reliable assessment of intracerebral hemorrhage, intraventricular hemorrhage, and perihematomal edema on CT. • Non-contrast CT-based deep learning method can be helpful to provide efficient and accurate measurements of ICH, IVH, and PHE in primary ICH patients, thereby reducing the effort and time of doctors to segment and calculate volumes of ICH, IVH, and PHE in primary ICH patients.
首次评估一种基于 no-new-Net 的深度学习方法在原发性脑出血（ICH）的 CT 上对脑出血（ICH）、脑室内扩展（IVH）和血肿周围水肿（PHE）的全自动分割和容量测量的性能。
将 380 名在医院就诊时接受 CT 检查的原发性 ICH 患者分为训练队列（n=300）和验证队列（n=80）。一个由 80 名患者组成的独立队列用于测试。由放射科医生手动生成真实（分割掩模）。通过将模型结果与放射科医生的分割结果进行比较，评估病变分割和 ICH、IVH 和 PHE 容量测量的模型性能。
在测试队列中，病变分割的 Dice 评分分别为 0.92、0.79 和 0.71，用于 ICH、IVH 和 PHE。灵敏度分别为 0.93、0.88 和 0.81。阳性预测值分别为 0.92、0.76 和 0.69，用于 ICH、IVH 和 PHE。手动和自动测量体积之间显示出极好的一致性（一致性相关系数[CCC]≥0.98）ICH 和 IVH，以及 PHE 的良好一致性（CCC≥0.92）。模型为每位患者提供自动分割和体积分析大约需要 15 秒。
我们的模型在原发性 ICH 中对 ICH、IVH 和 PHE 的自动分割和体积测量具有良好的可靠性，这有助于减少医生计算 ICH、IVH 和 PHE 体积的工作量和时间。
深度学习算法可提供 CT 上脑出血、脑室内出血和血肿周围水肿的自动和可靠评估。
基于非对比 CT 的深度学习方法有助于为原发性 ICH 患者提供高效、准确的 ICH、IVH 和 PHE 测量，从而减少医生分割和计算原发性 ICH 患者 ICH、IVH 和 PHE 体积的工作量和时间。
