Microvascular invasion in hepatocellular carcinoma: is it predictable with quantitative computed tomography parameters?

Martins, Daniel Lahan; Perales, Simone Reges; Gallani, Stephanie Kilaris; Costa, Larissa Bastos Eloy da; Lago, Eduardo Andreazza Dal; Boin, I.F.S.F.; Caserta, Nelson Marcio Gomes; Ataíde, Elaine Cristina de

doi:10.1590/0100-3984.2018.0123

Cited by 10 publications

(7 citation statements)

References 30 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Evaluation of the liver by imaging methods has been the subject of various recent studies in the radiology literature of Brazil (28)(29)(30)(31)(32) . The availability of affordable, good quality ultrasound machines has allowed the development of focused point-of-care ultrasound protocols applicable to tropical diseases.…”

Section: Discussionmentioning

confidence: 99%

Ultrasound evaluation of schistosomiasis-related morbidity among the Xakriabá people in the state of Minas Gerais, Brazil

et al. 2020

View full text Add to dashboard Cite

Objective: To use ultrasound to investigate the morbidity related to schistosomiasis in the Xakriabá indigenous population. Materials and Methods: This was a field-based census study conducted in the territory of the Xakriabá people. A total of 166 individuals were invited, and 148 (≤ 77 years of age) agreed to participate. Most participants underwent abdominal ultrasound, physical examination, and stool examination. Mann-Whitney U and chi-square tests were used for comparisons. We determined risk by calculating odds ratio (OR) and performed logistic regression analysis. Results: Schistosoma mansoni eggs were found in 31 (26.7%) of the 116 stool samples examined, 22 (70.9%) of the 31 being from individuals 4-16 years of age. The median count was 144 eggs/g of feces (interquartile range, 264). Of the 105 participants examined with ultrasound, 68 (64.8%) had hepatomegaly (left lobe), 6 (5.7%) had splenomegaly, and 4 (3.8%) had portal hypertension. Egg-positive stool samples were more common in those with an enlarged left lobe (OR = 3.4; 95% confidence interval (CI): 1.1-11.2; p = 0.043). Periportal fibrosis was found in 30 participants (28.6%), of whom 9 (30%) had pattern C, 10 (33.3%) had pattern D, and 11 (36.7%) had pattern Dc. Age was the only independent risk factor for fibrosis (p = 0.007). Fibrosis was up to nine-fold more common in alcohol drinkers than in nondrinkers (OR = 9.28; 95% CI: 2.60-33.06; p < 0.001). Among the 138 participants in whom the clinical form was classified, the chronic hepatic form was identified in 54 (39.1%), of whom 32 (59.2%) were under 30 years of age and one (1.8%) was hepatosplenic. Conclusion: Schistosomiasis in the Xakriabá population is characterized by a high frequency of egg-positive stool samples, predominantly in children/adolescents, and by chronic hepatic form in the young, especially among alcohol drinkers.

show abstract

Section: Discussionmentioning

confidence: 99%

Ultrasound evaluation of schistosomiasis-related morbidity among the Xakriabá people in the state of Minas Gerais, Brazil

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Recent studies in the radiology literature of Brazil have emphasized the importance of imaging examinations in the evaluation of hepatic neoplasms (18)(19)(20)(21)(22)(23) . The present study used the LI-RADS classification system to analyze 38 patients infected with hepatitis viruses (HBV, HCV, HDV, or combinations thereof) who underwent CT. We found that approximately 32% of the patients presented with LI-RADS 5 observations (i.e., lesions diagnosed by imaging as classic HCC).…”

Section: Discussionmentioning

confidence: 99%

Use of the LI-RADS classification in patients with cirrhosis due to infection with hepatitis B, C, or D, or infected with hepatitis B and D

et al. 2020

View full text Add to dashboard Cite

Objective: To evaluate liver lesions, in accordance with the LI-RADS classification, using contrast-enhanced multiphase dynamic computed tomography in patients with hepatitis B, coinfected or not with hepatitis D, or with chronic hepatitis C, as well as to determine the level of agreement between radiologists. Materials and Methods: We evaluated 38 patients with hepatitis B, coinfected or not with hepatitis D, or with chronic hepatitis C, all of whom underwent contrast-enhanced multiphase dynamic computed tomography. For each examination, two radiologists selected up to three hepatic lesions, categorizing them in accordance with the LI-RADS classification and evaluating signs of chronic liver disease and portal hypertension. To determine the level of agreement between radiologists, we calculated the kappa statistic (κ) . Results: Radiologist 1 and radiologist 2 selected 56 and 48 liver lesions, respectively. According to radiologist 1 and radiologist 2, respectively, 27 (71%) and 23 (61%) of the 38 patients had at least one liver lesion; 13 (34%) and 12 (32%) had a LI-RADS 5 lesion (κ = 0.821); 19 (50%) and 16 (42%) had a hypervascular lesion (κ = 0.668); and 30 (79%) and 24 (63%) had splenomegaly (κ = 0.503). Both radiologists identified chronic liver disease in 31 (82%) of the patients (κ = 1.00). Conclusion: Lesions categorized as LI-RADS 5 were detected in approximately 32% of the patients, with almost perfect agreement between the radiologists. The level of agreement was substantial or moderate for the other LI-RADS categories.

show abstract

“…Classifier chains [11], [12] is one of the standard MLL methods, in which, conditional dependence between a label and its predecessors is modeled. Other approaches like As a demonstration, the relative locations of different gene types on the left-most chromosome may not be in the exact order on DNA string, or can even be on different chromosomes [6].…”

Section: Related Workmentioning

confidence: 99%

Multi-Instance Multi-Label Learning for Gene Mutation Prediction in Hepatocellular Carcinoma

Xu,

Zhao,

et al. 2020

Preprint

View full text Add to dashboard Cite

Gene mutation prediction in hepatocellular carcinoma (HCC) is of great diagnostic and prognostic value for personalized treatments and precision medicine. In this paper, we tackle this problem with multi-instance multi-label learning to address the difficulties on label correlations, label representations, etc. Furthermore, an effective oversampling strategy is applied for data imbalance. Experimental results have shown the superiority of the proposed approach.Clinical relevance The proposed framework applies multiinstance multi-label learning for gene mutation detection in HCC, which can be implemented to provide clinical diagnostic and predictive oncology services.

show abstract

Microvascular invasion in hepatocellular carcinoma: is it predictable with quantitative computed tomography parameters?

Cited by 10 publications

References 30 publications

Ultrasound evaluation of schistosomiasis-related morbidity among the Xakriabá people in the state of Minas Gerais, Brazil

Ultrasound evaluation of schistosomiasis-related morbidity among the Xakriabá people in the state of Minas Gerais, Brazil

Use of the LI-RADS classification in patients with cirrhosis due to infection with hepatitis B, C, or D, or infected with hepatitis B and D

Multi-Instance Multi-Label Learning for Gene Mutation Prediction in Hepatocellular Carcinoma

Contact Info

Product

Resources

About