Hepatic Xanthogranuloma that Originated from a Liver Cyst and Mimicked a Malignant Tumor

Surgical treatment of hepatic xanthogranuloma arising from a giant hepatic cyst causing gastrointestinal obstruction: a case report

Hepatic cysts are benign liver lesions and are often asymptomatic. Large hepatic cysts may cause jaundice and portal hypertension; however, they rarely cause gastrointestinal obstruction. Symptomatic cysts require treatment, and when malignancy is suspected, cyst puncture for pathological examination of the fluid may pose a risk of dissemination. Herein, we describe a case of xanthogranuloma arising from a large hepatic cyst that was causing duodenal obstruction. Thus, cyst puncture was performed for emergency decompression. Cytological examination of the puncture fluid revealed no malignant findings. Hence, laparoscopic deroofing was performed to treat the hepatic cyst. As the cyst and duodenal wall were firmly adherent, the cyst wall was left behind without dissection from the duodenum. A two-stage approach of cyst puncture followed by surgery may be an option for patients requiring urgent treatment for potentially malignant hepatic cysts.

Automated Knowledge Extraction of Liver Cysts From CT Images Using Modified Whale Optimization and Fuzzy C Means Clustering Algorithm

In this study, the integrated modified whale optimization and modified fuzzy c-means clustering algorithm using morphological operations are developed and implemented for appropriate knowledge extraction of a cyst from computer tomography (CT) images of the liver to facilitate modern intelligent healthcare systems. The proposed approach plays an efficient role in diagnosing the liver cyst. To evaluate the efficiency, the outcomes of the proposed approach have been compared with the minimum cross entropy based modified whale optimization algorithm (MCE and MWOA), teaching-learning optimization algorithm based upon minimum cross entropy (MCE and TLBO), particle swarm intelligence algorithm (PSO), genetic algorithm (GA), differential evolution (DE) algorithm, and k-means clustering algorithm. For this, various parameters such as uniformity (U), mean structured similarity index (MSSIM), structured similarity index (SSIM), random index (RI), and peak signal-to-noise ratio (PSNR) have been considered. The experimental results show that the proposed approach is more efficient and accurate than others.