Automated Extraction of Lymph Nodes from 3-D Abdominal CT Images Using 3-D Minimum Directional Difference Filter

Abstract

Abstract. This paper presents a method for extracting lymph node regions from 3-D abdominal CT images using 3-D minimum directional difference filter. In the case of surgery of colonic cancer, resection of metastasis lesions is performed with resection of a primary lesion. Lymph nodes are main route of metastasis and are quite important for deciding resection area. Diagnosis of enlarged lymph nodes is quite important process for surgical planning. However, manual detection of enlarged lymph nodes on CT images is quite burden task. Thus, development of lymph node detection process is very helpful for assisting such surgical planning task. Although there are several report that present lymph node detection, these methods detect lymph nodes primary from PET images or detect in 2-D image processing way. There is no method that detects lymph nodes directly from 3-D images. The purpose of this paper is to show an automated method for detecting lymph nodes from 3-D abdominal CT images. This method employs a 3-D minimum directional difference filter for enhancing blob structures with suppressing line structures. After that, false positive regions caused by residua and vein are eliminated using several kinds of information such as size, blood vessels, air in the colon. We applied the proposed method to three cases of 3-D abdominal CT images. The experimental results showed that the proposed method could detect 57.0 % of enlarged lymph nodes with 58 FPs per case.

Segmentation Based Features for Lymph Node Detection from 3-D Chest CT

“…The comparability is however limited because of different data, different criterions for a detection, different body regions and different minimum lymph node sizes used for evaluation. Both [9] and [5] report a very high number of false alarms and also consider a lymph node already as detected if there is just overlap with the automatic segmentation. In [3], very good results are reported, but the method was evaluated on a single dataset.…”

“…In [9,5], lymph nodes are detected by a cascade of filters (Hessian based, morphological operations and a so-called 3-D Min-DD filter). In [3], lymph nodes are detected and segmented by fitting a mass-spring model at different positions.…”

Graph-Based Segmentation of Lymph Nodes in CT Data

“…Their volumetric segmentation error ranged between 39% and 52%. Kitasaka et al [20] utilized a 3-D minimum directional difference filter for extracting abdominal lymph nodes in CT data. A series of processing steps like region growing and several false positive reduction strategies were applied for segmentation.…”

Mediastinal lymph node detection and station mapping on chest CT using spatial priors and random forest

“…In addition to mediastinal lymph node detection, there have been several preceding works which have investigated lymph node detection in the abdomen, pelvic, and axillary. [13][14][15][16][17][18] For example, Kitasaka et al 16 presented a 3D minimum directional difference filter for abdominal lymph nodes detection and applied Hessian-based vesselness to reduce the false positives. Barbu et al…”