Dataset Information

Deep learning radiomics model accurately predicts hepatocellular carcinoma occurrence in chronic hepatitis B patients: a five-year follow-up.

ABSTRACT: An early and accurate prediction of hepatocellular carcinoma (HCC) is beneficial for individualized treatment and follow-up of chronic hepatitis B (CHB) patients. We aimed to establish a prediction model for HCC by radiomics analysis in CHB patients and compare performance with liver stiffness measurement (LSM) and other clinical prognostic scores. Initially, 1215 patients were included and finally 434 CHB patients with 5-year follow-up were enrolled, 96.3% of them underwent liver biopsy. Deep learning radiomics analysis was performed on 2170 two-dimensional shear wave elastography (2D-SWE) and corresponding B-mode ultrasound (US) images. These high-throughput imaging features were also combined with low-dimensional serological clinical data by deep learning radiomics to establish different HCC prediction models and to overcome challenges of an unbalanced sample. The best model which is simple with high accuracy was selected. Prediction performance of the selected model was compared with LSM and other clinical prognostic scores. During 5-year follow-up, 32 (7.4%) of 434 patients developed HCC. The best prediction model was HCC-R, which included 2D-SWE and B-mode US images, sex and age. This model showed a high predictive value with areas under the receiver operating characteristic curve (AUCs) of 0.981, 0.942 and 0.900 in training, validation and testing cohorts for predicting 5-year prognosis of HCC. These predictive values were significantly higher than that of LSM (AUC: 0.676~0.784, p < 0.05) and better than that of other clinical prognostic scores (AUC: 0.544~0.869). HCC-R radiomics model based on 2D-SWE and B-mode US images, sex and age comprehensively reflected biomechanical and morphological information of patients and can accurately predict HCC occurrence; thus, this model has great value for treatment and follow-up of CHB patients.

SUBMITTER: Jin J

PROVIDER: S-EPMC7868753 | biostudies-literature | 2021

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep learning radiomics model accurately predicts hepatocellular carcinoma occurrence in chronic hepatitis B patients: a five-year follow-up.

Jin Jieyang J Yao Zhao Z Zhang Ting T Zeng Jie J Wu Lili L Wu Manli M Wang Jinfen J Wang Yuanyuan Y Yu Jinhua J Zheng Rongqin R

American journal of cancer research 20210201 2

An early and accurate prediction of hepatocellular carcinoma (HCC) is beneficial for individualized treatment and follow-up of chronic hepatitis B (CHB) patients. We aimed to establish a prediction model for HCC by radiomics analysis in CHB patients and compare performance with liver stiffness measurement (LSM) and other clinical prognostic scores. Initially, 1215 patients were included and finally 434 CHB patients with 5-year follow-up were enrolled, 96.3% of them underwent liver biopsy. Deep l ...[more]

PMID: 33575088

Similar Datasets

Project description:PURPOSE: To assess longitudinal changes in parapapillary atrophy in the adult population of Greater Beijing. METHODS: The population-based Beijing Eye Study 2006 included 3251 subjects who had participated in the Beijing Eye Study 2001 and returned for re-examination. The mean age was 60.4 ± 10.1 years. Using optic disc photographs, we measured parapapillary atrophy which was divided into alpha zone and beta zone. RESULTS: Overall progression rate of alpha zone was seen in 0.6 ± 0.1% (95% confidence interval (CI):0.3,0.9) of the subjects and of beta zone in 8.2 ± 0.5% (95%CI:7.2,9.1) of the subjects. In binary regression analysis, rate of progression of alpha zone was significantly associated higher age (P = 0.04) and the co-progression of zone Beta (P<0.001). Rate of progression of beta zone was significantly associated with higher age (P<0.001; odds ratio (OR):1.11;95%CI:1.10,1.14), higher intraocular pressure (P<0.001;OR:1.10;95%CI:1.05,1.14), higher myopic refractive error (P<0.001;OR:0.71; 95%CI:0.67,0.75), rural region of habitation (P = 0.002;OR: 0.58; 95%CI:0.41,0.82), presence of glaucomatous optic nerve damage (P<0.001;OR:2.89; 95%CI:1.62,5.14), co-progression of alpha zone (P<0.001;OR:7.13;95%CI:2.43,20.9), absence of arterial hypertension (P = 0.03;OR: 0.70; 95%CI:0.51,0.96), and thicker central corneal thickness (P = 0.02;OR:1.01;95%CI:1.00,1.01). Subjects with a non-glaucomatous optic nerve damage (n = 22) as compared to the remaining subjects did not vary in the progression rate of alpha zone (0.0% versus 0.6 ± 0.1%; P = 1.0) and beta zone (8.2 ± 0.5% versus 6.3 ± 0.6%;P = 1.0). CONCLUSIONS: In adult Chinese in Greater Beijing, the 5-year progression rate of beta zone of parapapillary atrophy (seen in 8.2 ± 0.5% of subjects) was significantly correlated with higher age, rural region of habitation, absence of arterial hypertension, higher intraocular pressure, higher myopic refractive error, thicker central corneal thickness, and presence of glaucoma. It was not associated with non-glaucomatous optic nerve damage.

Project description:In 2009, Mekele, the capital of the Tigray Region in Ethiopia, presented a mean prevalence of 44.7% of schistosomiasis (S. mansoni) in school children. Termed a public health problem, NALA, an international public health non-governmental organization, and their partners implemented a novel model of intervention, which aimed to compliment mass drug administration (MDA) campaigns with behavioral change (BC) and improved sanitation to achieve sustained elimination of schistosomiasis. The four-year intervention (2009-2012) covered 38 primary schools. The objective of this study was to examine factors associated with control or resurgence of the disease, and the association between the behavioral change program and disease prevalence, ten years after initiation. Eleven primary schools were selected for this follow-up study. All students provided a stool sample and filled in a knowledge, attitude and practice (KAP) questionnaire. In seven out of eleven schools (63.6%) the prevalence of schistosomiasis was maintained below 2% ten years after the initiation of the intervention. In four schools, prevalence returned to pre-intervention levels, defining them as persistent hot spots (PHS). Students from PHS schools scored lower on KAP questionnaires compared to students from responder schools; 3.9 ± 0.9 vs. 4.2 ± 0.9 (p-value < 0.001) for practice questions and 4.4 ± 1.4 vs. 4.6 ± 1.5 (p-value = 0.03) for attitude questions. The prevalence of schistosomiasis correlated positively with age, (p-value = 0.049), sex (relative risk = 1.7, p-value < 0.001), and location. Semi-urban locations (n = 382) had higher disease prevalence than urban locations (n = 242), (22.7% vs. 5.5%, p-value < 0.001). Students residing in semi-urban areas and close to a river (<500 m) were at higher risk of contracting schistosomiasis than those living in urban areas far from the river (RR = 5.95, p-value < 0.001). Finally, a correlation between prevalence and proximity of schools to rivers was found (semi-urban areas; RR = -0.91, p-value = 0.001 vs. urban areas; RR = -0.51, p-value = 0.001). Soil-transmitted-helminths prevalence in 2009 was 8.1% and declined during the intervention years to 0.5%. Prevalence in 2018 was found to be stable at 0.8%. These results demonstrate the long-term success of NALAs' comprehensive model of intervention for elimination of schistosomiasis in school children, combining behavioral change and improved sanitation with MDA.

Dataset Information

Deep learning radiomics model accurately predicts hepatocellular carcinoma occurrence in chronic hepatitis B patients: a five-year follow-up.

Publications

Deep learning radiomics model accurately predicts hepatocellular carcinoma occurrence in chronic hepatitis B patients: a five-year follow-up.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets