Developments of Image Processing Methods for Detecting Small Lesions at Early Stage of Cancer on X-RAY Images

本文从医学x射线图像形成的各个物理机制出发，研究了具有光学模糊和量子斑驳的x射线图像的恢复方法。具体地说，首先利用激光扫描仪的传递特性、薄膜的特性曲线、光密度的对数变换和基于点扩散函数的伪逆滤波器，确定性地消除了光学模糊。其次，根据贝叶斯定理推导了两种剩余量子斑的恢复方法。一种是利用量子斑驳的统计特性的随机型广义数字滤波器。另一种是可能缺乏一些信息的新的扩展回归模型和一个简单的实验回归模型。因此，量子斑点在统计上被去除。研究结果为具有光学模糊和量子斑点的医学x射线图像的恢复提供了一种新的方法[j]。和基于扩展回归分析方法的医学x射线图像恢复方法j （Denshi Joho Tsushin Gakkai Ronbunshi D-II, Vol. J74-D-II, No. 3, pp. 442-452(1991)）。此外，通过将多变量贝叶斯定理扩展为与各种波形的实际复杂程度相匹配的层次形式，建立了随机测量噪声下退化图像的动态恢复方法。该结果在1991年10月于日本京都召开的“物理系统与1/f波动中的噪声”国际会议上，作为一种利用广义数字滤波器对光学模糊和量子斑驳的医学x射线图像进行动态自适应恢复的方法被提出。最后，将本研究提出的x线图像复原方法应用于牙科实际图像，由牙医直观评价并确认处理后图像的有效性。少

In this study, some restoration methods for the x-ray images with optical blurs and quantum mottles were developed by considering each physical mechanism of the medical x-ray image formation. More specifically, the optical blurs were first deterministically cleared away by using the transfer characteristic of the laser scanner, the characteristic curve of the film, the logarithmic transformation of the optical density and a pseudo-inverse filter based on the point spread function. Next, two restoration methods for remaining quantum mottles were derived from the basis of a Bayes' theorem. One was a wide sense digital filter of a stochastic type using the statistical properties of quantum mottles. Another was a new extended regression model that might have a little information lack and a simple experimental regression model. Thus, the quantum mottles were statistically removed. The results were presented as iA new restoration method for medical x-ray images with optical blurs and quantum … More mottles j (Trans. of IEICE, Vol. E73, No. 5, pp. 670-678 (1990) ) and as iA restoration method of medical x-ray images based on an extended regression analysis method j (Denshi Joho Tsushin Gakkai Ronbunshi D-II, Vol. J74-D-II, No. 3, pp. 442-452 (1991) ), respectively. Moreover, by expanding the multi-variate Bayes' theorem into the hierarchical form matched to the actual complexity of various wave forms, a dynamical method of restoring the degraded images under the random measurement noises was newly established. This result was presented as iA dynamical method of adaptive restoration for the medical x-ray images with optical blurs and quantum mottles by use of wide sense digital filter j in The International Conference on Noise in Physical Systems and 1/f Fluctuations in Kyoto in Oct. 1991.Finally, the restoration method of x-ray images proposed in this study was applied to those of the actual dental images, and dentists visually evaluated and confirmed the effectiveness of these processed images. Less