Automated liver lesion detection in CT images based on multi-level geometric features

Abstract

The proposed method can efficiently detect liver lesions irrespective of their size, shape, density and heterogeneity within half a minute. According to the evaluation, its accuracy is competitive with the actual state-of-the-art approaches, and the contour of the detected findings is acceptable in most of the cases. Future work shall focus on more precise lesion contouring so that the proposed method can be a solid basis for fully automated liver tumour burden estimation.

Automated segmentation methods for liver analysis in oncology applications

“…In this chapter the author presented an automated approach that can detect all types of liver lesions with high sensitivity and low false positive rate within a short run time. The related results were published in a journal paper [3]. The author developed a novel technique for automated liver lesion detection in contrastenhanced CT images.…”

“…When the resected part is not visible (h) one can see the cutting surface is smooth and the location of the input traces is not remarkable. In addition to the visualization the volume parts were quantified: the remnant liver (red) was 1461 cm 3 , and the resected part (green) was 716 cm 3 . …”

“…In addition to the characteristic intensity, the expected volume (ca. 1500 cm 3 ) and the location (right abdomen) of the liver can be exploit to localize significant part of the organ automatically. The image to be segmented is acquired in the portal-phase of the multi-phase CT examination.…”

“…This chapter presents a novel technique for automated liver lesion detection in portal-phase contrastenhanced CT images. The related work of the author was published in a journal paper [3].…”

“…The author proposed an automated approach to solve this problem. The related results were published in a journal paper [3].…”