No abstract

Traumatic lymphorrhea is a rare but potentially life-threatening complication. Postoperative lymphorrhea is the leading cause of traumatic lymphorrhea and can arise anywhere within the lymphatic system. Leaks arising from the aortoiliac region to the thoracic duct (TD) and from hepatic lymphatics can be identified with intranodal lymphangiography and transhepatic lymphangiography, respectively. Therefore, an appropriate lymphangiography technique is essential for identifying the sources of leaks. Chylothorax resulting from damage to the TD can be serious because the TD transports large amounts of lymphatic fluid from the gastrointestinal, hepatic, and aortoiliac regions. Percutaneous TD embolization-comprising access to the TD followed by embolization-has recently become a minimally invasive alternative to surgical TD ligation for high-output chylothorax. The selection of access routes to the TD depends on its anatomy. If the TD cannot be approached by such means, other options include TD needle interruption or drainage of lymphatic fluid adjacent to the leakage point followed by sclerotherapy. Most cases of abdominal lymphorrhea arise from the aorta-iliac lymphatic system, and lymphangiography alone or computed tomography-guided sclerotherapy might be useful. Rarely, leakage may arise from hepatic lymphatics due to a damaged gastroduodenal ligament and can be visualized and embolized transhepatically. This article comprehensively reviews clinically relevant anatomic TD variations, lymphangiography techniques and criteria for their selection, and treatment strategies for lymphorrhea. RSNA, 2016.

Objective: This study aimed to use convolutional neural network (CNN), a deep learning software, to assist in cT1b diagnosis. Methods: This retrospective study used 190 colon lesion images from 41 cases of colon endoscopies performed between February 2015 and October 2016. Unenhanced colon endoscopy images (520 × 520 pixels) with white light were used. Images included 14 cTis cases with endoscopic resection and 14 cT1a and 13 cT1b cases with surgical resection. Protruding, flat, and recessed lesions were analyzed. AlexNet and Caffe were used for machine learning. Fine tuning of data to increase image numbers was performed. Oversampling for the training images was conducted to avoid impartiality in image numbers, and learning was carried out. The 3-fold cross-validation method was used. Sensitivity, specificity, accuracy, and area under the curve (AUC) values in the receiver operating characteristic curve were calculated for each group. Results: The results were the average of obtained values. With CNN learning, cT1b sensitivity, specificity, and accuracy were 67.5, 89.0, and 81.2%, respectively, and AUC was 0.871. Conclusion: Quantitative diagnosis is possible using an endoscopic diagnostic support system with machine learning, without relying on the skill and experience of endoscopists. Moreover, this system could be used to objectively evaluate endoscopic diagnoses.

No abstract

Background: Cirrhotic patients with hepatocellular carcinoma (HCC) frequently have impaired glucose metabolism. Aims: To investigate whether impaired glucose metabolism affects the growth rate of the tumour. Patients and methods: Tumour doubling time (DT), assessed by ultrasound imaging analysis, was measured in 60 patients with single small HCC (diameter <30 mm). DT was compared with plasma insulin and glucose concentrations following the oral glucose tolerance test (OGTT). The effect of continuous infusion of octreotide (a somatostatin analogue 200 µg/day) for three months on DT in five cases was assessed. Results: The 60 patients were divided into two groups because the median DT was 140 days: rapid growth group (DT <140 days, n=30) and slow growth group (DT >140 days, n=30). Fasting plasma insulin concentration and area under the plasma insulin curve (AUC ins ) of the OGTT (10.4 (6.2) µU/ml and 262 (152) µU/ml/h, respectively; mean (SD)) in the rapid growth group were significantly higher than those in the slow growth group (7.6 (4.3) and 146 (140), respectively) (p=0.041 and p=0.0006, respectively). In contrast, fasting plasma glucose concentration and area under the plasma glucose curve (AUC gluc ) in the rapid growth group were significantly lower than those in the slow growth group (p=0.0003 and p=0.0012, respectively). Univariate and multivariate analyses of logistic regression models demonstrated that AUC ins was a significant factor contributing to the growth rate of HCC (p=0.001 and p=0.016, respectively). AUC ins significantly decreased after octreotide treatment (p<0.02) but AUC gluc did not significantly change. DT after treatment increased in three of the five patients and could not be calculated in the remaining two patients because of no change in the diameter of the tumour. Conclusions: These data suggest that postprandial hyperinsulinaemia is associated with accelerated HCC growth.