Department of Biomedical Engineering, The George Washington University, Washington, DC, 20052, USA.
Med Phys. 2020 Apr;47(4):1786-1795. doi: 10.1002/mp.14069. Epub 2020 Mar 5.
To use machine-learning algorithms and blur measure (BM) operators to automatically detect motion blur in mammograms. Motion blur has been reported to reduce lesion detection performance and mask small abnormalities, resulting in failure to detect them until they reach more advanced stages. Automatic detection of blur could support the clinical decision-making process during the mammography exam by allowing for an immediate retake, thereby preventing unnecessary expense, time, and patient anxiety.
Blur was simulated mathematically to mimic the real blur seen in clinical practice. The blur point-spread-function (PSF) mask is generated by distributing pixel intensity of an image pixel moving under random motion within the range of blur effect (the maximum amount of tissue motion allowed). The random motion trajectory vector is generated on a super-sampled image frame to accommodate smaller substeps; the vector was then sampled on a regular pixel grid using subpixel linear interpolation to generate the blur PSF mask. This randomly generated motion trajectory is constrained by several factors: the effects of variations in tissue elasticity, imaging exposure time, and size of blur effect (motion boundary in millimeters) were examined. The blur mask is convolved with a mammogram to create blur. Five motion blur magnitudes (0.1, 0.25, 0.5, 1.0, and 1.5 mm) were simulated on 244 and 434 mammograms from the INbreast and DDSM databases, respectively. Blur was quantified using nine BM operators for each mammogram and at each blur level. The mammograms were assigned to training (70%) and testing (30%) datasets to train three machine-learning classifiers: Ensemble Bagged Trees, fine Gaussian SVM, and weighted KNN, to distinguish five levels of blurred from unblurred mammograms, using six-way classification.
For the INbreast mammograms, the average classification accuracies were 87.7%, 85.7%, and 85.7% for Ensemble Bagged Trees, fine Gaussian SVM, and weighted KNN, respectively, and the average classification accuracies for DDSM were 93.5%, 93.6%, and 92.7% for Ensemble Bagged Trees, fine Gaussian SVM, and weighted KNN, respectively.
Preliminary results show the potential to detect simulated blur automatically using those methods. Although limited work has been done to quantify the effects of motion blur on radiologists' performance, there is evidence that motion blur might not be detected visually by a human observer and could negatively affect radiologists' lesion detection performance. As of this date, no other study has investigated the ability of machine-learning classifiers and BM operators to detect motion blur in mammograms.
使用机器学习算法和模糊度量（BM）算子自动检测乳腺 X 光片中的运动模糊。运动模糊据报道会降低病变检测性能并掩盖小的异常，从而导致在病变发展到更晚期之前无法检测到。自动检测模糊可以通过允许立即重拍来支持乳腺 X 光检查的临床决策过程，从而避免不必要的费用、时间和患者焦虑。
通过数学模拟来模拟临床实践中看到的真实模糊。模糊点扩散函数（PSF）掩模通过在模糊效果范围内（允许的最大组织运动量）分布图像像素的像素强度来生成。随机运动轨迹向量在超采样图像帧上生成，以适应较小的子步骤；然后使用子像素线性插值在规则像素网格上对向量进行采样，以生成模糊 PSF 掩模。这种随机生成的运动轨迹受到多个因素的限制：检查了组织弹性变化、成像曝光时间和模糊效果大小（毫米处的运动边界）的影响。模糊掩模与乳腺 X 光片卷积以创建模糊。在 INbreast 和 DDSM 数据库中的 244 和 434 张乳腺 X 光片中分别模拟了 5 种运动模糊程度（0.1、0.25、0.5、1.0 和 1.5 毫米）。使用 9 种 BM 算子对每张乳腺 X 光片和每个模糊级别进行模糊量度。将乳腺 X 光片分配到训练（70%）和测试（30%）数据集，以使用六类分类训练三种机器学习分类器：集成袋装树、精细高斯 SVM 和加权 KNN，以区分 5 级模糊与非模糊乳腺 X 光片。
对于 INbreast 乳腺 X 光片，集成袋装树、精细高斯 SVM 和加权 KNN 的平均分类准确率分别为 87.7%、85.7%和 85.7%，而对于 DDSM，集成袋装树、精细高斯 SVM 和加权 KNN 的平均分类准确率分别为 93.5%、93.6%和 92.7%。
初步结果表明，使用这些方法自动检测模拟模糊具有潜力。尽管已经进行了有限的工作来量化运动模糊对放射科医生性能的影响，但有证据表明运动模糊可能无法被人类观察者视觉检测到，并可能对放射科医生的病变检测性能产生负面影响。截至目前，尚无其他研究调查机器学习分类器和 BM 算子检测乳腺 X 光片中运动模糊的能力。
