南方医科大学学报 ›› 2022, Vol. 42 ›› Issue (2): 223-231.doi: 10.12122/j.issn.1673-4254.2022.02.08

• • 上一篇    下一篇

低剂量CT图像重建算法对脑出血检测性能的影响

符 帅,李明强,边兆英,马建华   

  1. 南方医科大学生物医学工程学院,广东 广州 510515;广州市医用放射成像与检测技术重点实验室,广东 广州 510515;琶洲实验室,广东 广州 510515
  • 出版日期:2022-02-20 发布日期:2022-03-16

Performance of low-dose CT image reconstruction for detecting intracerebral hemorrhage: selection of dose, algorithms and their combinations

FU Shuai, LI Mingqiang, BIAN Zhaoying, MA Jianhua   

  1. School of Biomedical Engineering, Southern Medical University, Guangzhou 510515, China; Guangzhou Key Laboratory of Medical Radiation Imaging and Detection Technology, Guangzhou 510515, China; Pazhou Lab, Guangzhou 510515, China
  • Online:2022-02-20 Published:2022-03-16

摘要: 目的 探究低剂量CT图像重建算法对脑出血检测性能的影响。方法 对正常剂量(定义为100% dose)脑出血CT图像进行低剂量CT成像仿真,仿真剂量包括30%、25%和20% dose。采用7种CT图像重建算法进行图像重建,以实现抑制低剂量CT图像噪声,包括滤波反投影算法(FBP)、惩罚加权最小二乘的全变分(PWLS-TV)、非局部均值滤波(NLM)、3维块匹配(BM3D)、残差编码解码卷积神经网络(REDCNN)、FBP卷积神经网络(FBPConvNet)和图像恢复迭代残差卷积网络(IRLNet)。基于深度学习方法的脑出血检测模型(CNN-LSTM)对正常剂量图像和7种重建算法得到的图像进行脑出血检测。对7种重建算法与正常剂量图像的脑出血检测结果进行比较,评估不同重建算法对脑出血检测性能的影响。结果(1)对于同一算法,剂量对脑出血检测性能的影响:在30%、25%和20% dose下,FBP算法脑出血检测正确率分别为82.21%、74.61%和65.55%。(2)在相同剂量(30% dose)下,不同图像重建算法对脑出血检测性能的影响:FBP、PWLS-TV、NLM、BM3D、REDCNN、FBPConvNet和IRLNet算法的脑出血检测正确率分别为82.21%、86.80%、89.37%、81.43%、90.05%、90.72%和93.51%。(3)IRLNet算法在30%、25%和20% dose下的脑出血检测正确率分别为93.51%、93.51%和93.06%。结论 剂量和重建算法的选择对脑出血检测性能有显著影响,临床中选择合适的剂量和算法组合能在大幅度降低辐射剂量同时保证脑出血检测性能。

关键词: 低剂量CT成像;脑出血检测;深度学习

Abstract: Objective To investigate the performance of different low-dose CT image reconstruction algorithms for detecting intracerebral hemorrhage. Methods Low-dose CT imaging simulation was performed on CT images of intracerebral hemorrhage at 30%, 25% and 20% of normal dose level (defined as 100% dose). Seven algorithms were tested to reconstruct low-dose CT images for noise suppression, including filtered back projection algorithm (FBP), penalized weighted least squares-total variation (PWLS-TV), non-local mean filter (NLM), block matching 3D (BM3D), residual encoding-decoding convolutional neural network (REDCNN) , the FBP convolutional neural network (FBPConvNet) and image restoration iterative residual convolutional network (IRLNet). A deep learning-based model (CNN-LSTM) was used to detect intracerebral hemorrhage on normal dose CT images and low-dose CT images reconstructed using the 7 algorithms. The performance of different reconstruction algorithms for detecting intracerebral hemorrhage was evaluated by comparing the results between normal dose CT images and low-dose CT images. Results At different dose levels, the low-dose CT images reconstructed by FBP had accuracies of detecting intracerebral hemorrhage of 82.21% , 74.61% and 65.55% at 30% , 25% and 20% dose levels, respectively. At the same dose level (30% dose), the images reconstructed by FBP, PWLS-TV, NLM, BM3D, REDCNN, FBPConvNet and IRLNet algorithms had accuracies for detecting intracerebral hemorrhage of 82.21%, 86.80%, 89.37%, 81.43%, 90.05%, 90.72% and 93.51%, respectively. The images reconstructed by IRLNet at 30%, 25% and 20% dose levels had accuracies for detecting intracerebral hemorrhage of 93.51% , 93.51% and 93.06% , respectively. Conclusion The performance of reconstructed low-dose CT images for detecting intracerebral hemorrhage is significantly affected by both dose and reconstruction algorithms. In clinical practice, choosing appropriate dose level and reconstruction algorithm can greatly reduce the radiation dose and ensure the detection performance of CT imaging for intracerebral hemorrhage.

Key words: low-dose CT imaging; intracranial hemorrhage detection; deep learning