B Ultrasound Image Based Fatty Liver Detection System Exploitation

Nowdays we check the fatty liver according to the B-ultrasound imaging,but theB-ultrasonic imaging exist shortcomings and deficiencies, and need the help of computerto achieve better recognition results. Existing computer-aided for fatty liver detected havea lot of weak point, for example, the recognition rate is not high and the process needingto calculate is very big, can not meet the requirements of modern digital hospital building.This article presents a new solution for the above, played a role in promoting theidentification on fatty liver and construction of hospital information.In this paper, two different types of fatty liver have different solutions.Heterogeneity of fatty liver using the method of image enhancement, combined with theadvantages of histogram equalization and homomorphic filtering, enhance the region ofthe suspected fat to highlight the lesion, reminding users to pay attention. For diffuse fattyliver, analysis and comparison of existing computer-aided method. A novel method forfatty liver B-scan image recognition with multi-grade gray level difference is proposed. Atrapeziform area is first extracted from B-scan image. And then the trapeziform area isdivided into n×m regular trapeziform areas. After computed every small areas averagegray level, the multi-grade gray level difference axis matrix is gained from near area to fararea. Finally the image is recognized with Neural Network based on multi-grade graylevel difference. The experiment shows that new method gets a good effect in fatty liverB-scan images recognition.In hospital information construction, this article use different image acquisitionmethods on support DICOM3.0digital interface and supports S-Video analog signaloutput of the B-ultrasound equipment. Connection fatty liver detection systems andhospital image database. To achieve the hospital information sharing. Propose a method touse semantics to generate the report card template. For the writing of the hospital medicalrecords. Improve the efficiency of doctor’s work.