Research On The Pulmonary Nodules Detection Based On Fractional Differential

In the process of acquiring the lung image,due to medical equipment or environmental factors,the detailed information of the lung image is blurred,the contrast is low,the boundary between the lung tissue structure and the background is unclear and contains noise.These characteristics make it easy to ignore some knot nodes with lower luminance values during the detection of pulmonary nodules,which will result in missed detection or false detection.Therefore,it is necessary to enhance the pulmonary nodules.However,most of the traditional image enhancement methods are based on the whole lung image.No special consideration is given to the enhancement of a specific lesion,such as the enhancement of pulmonary nodules,enhancement of blood vessels and so on.The detection of lung cancer in clinical application is mainly to detect pulmonary nodules.Therefore,the detection of nodules has important implications for studying how to enhance and improve the image quality of pulmonary nodules,and in particular,the enhancement of the contrast between the nodes of the lung node and its surrounding tissues.Fractional differential can also remove the noise while enhancing the texture details of the image.Therefore,the fractional differential technique is introduced in the paper to enhance the lung image.In order to improve the contrast between pulmonary nodules and other surrounding tissues,an adaptive fractional differential algorithm for enhancing the details of the lung texture was proposed in this paper.Also it was applied to the detection of pulmonary nodules to assist early diagnosis of lung cancer diagnosis.The main research content and innovative work of this article are as follows.Firstly,the fractional differential theory was analyzed and studied.According to the characteristics of lung nodule detection,a fractional differential mask for enhancing the texture information of lung image was constructed,and the fractional differential was introduced into the processing of the lung image.Comparing the fractional differential algorithm with the traditional image enhancement algorithm for simulation experiments,and the advantages and disadvantages of various methods were analyzed.The experimental results show that fractional derivative improves the contrast between pulmonary nodules and other surrounding tissues,and has better performance in enhancing image texture details.Secondly,aiming at the problem of selecting the best fractional differential order for multiple trials,an adaptive fractional differential algorithm for enhancing the details of the lung texture was proposed.Compared with the traditional fractional differential algorithm,this algorithm has obvious advantages in processing time,and can achieve the enhancement effect of the best order differential.Finally,the adaptive fractional order enhancement algorithm was applied to the preprocessing stage of the lung image to reduce the influence of noise.After the pretreated lung image,a threshold method was used to initially segment the lung parenchyma.After the segmentation,there is a substantial portion of the lung in the recessed region.The mathematical morphology is used to fill in the treatment so that it can cover all lung nodule regions,in order to identify all possible pulmonary nodule candidates,an algorithm based on fractional differential for initial identification of pulmonary nodules is proposed in this paper.The experimental results show that the proposed method candetect pulmonary nodule candidate regions effectively,and then pulmonary nodules were detected according to the shape of pulmonary nodules characteristics.