scholarly journals Predicting microvascular invasion in hepatocellular carcinoma: a deep learning model validated across hospitals

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Shu-Cheng Liu ◽  
Jesyin Lai ◽  
Jhao-Yu Huang ◽  
Chia-Fong Cho ◽  
Pei Hua Lee ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Deep Learning ◽  
Ct Images ◽  
Microvascular Invasion ◽  
University Hospital ◽  
Arterial Phase ◽  
Support Vector ◽  
Imaging Features ◽  
Clinical Factors ◽  
Validation Set

Abstract Background The accuracy of estimating microvascular invasion (MVI) preoperatively in hepatocellular carcinoma (HCC) by clinical observers is low. Most recent studies constructed MVI predictive models utilizing radiological and/or radiomics features extracted from computed tomography (CT) images. These methods, however, rely heavily on human experiences and require manual tumor contouring. We developed a deep learning-based framework for preoperative MVI prediction by using CT images of arterial phase (AP) with simple tumor labeling and without the need of manual feature extraction. The model was further validated on CT images that were originally scanned at multiple different hospitals. Methods CT images of AP were acquired for 309 patients from China Medical University Hospital (CMUH). Images of 164 patients, who took their CT scanning at 54 different hospitals but were referred to CMUH, were also collected. Deep learning (ResNet-18) and machine learning (support vector machine) models were constructed with AP images and/or patients’ clinical factors (CFs), and their performance was compared systematically. All models were independently evaluated on two patient cohorts: validation set (within CMUH) and external set (other hospitals). Subsequently, explainability of the best model was visualized using gradient-weighted class activation map (Grad-CAM). Results The ResNet-18 model built with AP images and patients’ clinical factors was superior than other models achieving a highest AUC of 0.845. When evaluating on the external set, the model produced an AUC of 0.777, approaching its performance on the validation set. Model interpretation with Grad-CAM revealed that MVI relevant imaging features on CT images were captured and learned by the ResNet-18 model. Conclusions This framework provide evidence showing the generalizability and robustness of ResNet-18 in predicting MVI using CT images of AP scanned at multiple different hospitals. Attention heatmaps obtained from model explainability further confirmed that ResNet-18 focused on imaging features on CT overlapping with the conditions used by radiologists to estimate MVI clinically.

scholarly journals Prediction of Microvascular Invasion in Hepatocellular Carcinoma: Preoperative Gd-EOB-DTPA-Dynamic Enhanced MRI and Histopathological Correlation

2018 ◽  
Vol 2018 ◽  
pp. 1-9 ◽  
Author(s):  
Mengqi Huang ◽  
Bing Liao ◽  
Ping Xu ◽  
Huasong Cai ◽  
Kun Huang ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Prediction Model ◽  
Tumor Size ◽  
Univariate Analysis ◽  
Microvascular Invasion ◽  
Arterial Phase ◽  
Hepatobiliary Phase ◽  
Imaging Features ◽  
Tumor Margin ◽  
Enhanced Mri

Objective. To investigate the imaging features observed in preoperative Gd-EOB-DTPA-dynamic enhanced MRI and correlated with the presence of microvascular invasion (MVI) in hepatocellular carcinoma (HCC) patients. Methods. 66 HCCs in 60 patients with preoperative Gd-EOB-DTPA-dynamic enhanced MRI were retrospectively analyzed. Features including tumor size, signal homogeneity, tumor capsule, tumor margin, peritumor enhancement during mid-arterial phase, peritumor hypointensity during hepatobiliary phase, signal intensity ratio on DWI and apparent diffusion coefficients (ADC), T1 relaxation times, and the reduction rate between pre- and postcontrast enhancement images were assessed. Correlation between these features and histopathological presence of MVI was analyzed to establish a prediction model. Results. Histopathology confirmed that MVI were observed in 17 of 66 HCCs. Univariate analysis showed tumor size (p=0.003), margin (p=0.013), peritumor enhancement (p=0.001), and hypointensity during hepatobiliary phase (p=0.004) were associated with MVI. A multiple logistic regression model was established, which showed tumor size, margin, and peritumor enhancement were combined predictors for the presence of MVI (α=0.1). R2 of this prediction model was 0.353, and the sensitivity and specificity were 52.9% and 93.0%, respectively. Conclusion. Large tumor size, irregular tumor margin, and peritumor enhancement in preoperative Gd-EOB-DTPA-dynamic enhanced MRI can predict the presence of MVI in HCC.

scholarly journals Prediction of Microvascular Invasion of Hepatocellular Carcinoma Based on Contrast-Enhanced MR and 3D Convolutional Neural Networks

2021 ◽  
Vol 11 ◽  
Author(s):  
Wu Zhou ◽  
Wanwei Jian ◽  
Xiaoping Cen ◽  
Lijuan Zhang ◽  
Hui Guo ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Neural Networks ◽  
Deep Learning ◽  
Convolutional Neural Networks ◽  
Loss Function ◽  
Microvascular Invasion ◽  
Arterial Phase ◽  
Contrast Enhanced ◽  
Auc Value ◽  
3D Cnn

Background and PurposeIt is extremely important to predict the microvascular invasion (MVI) of hepatocellular carcinoma (HCC) before surgery, which is a key predictor of recurrence and helps determine the treatment strategy before liver resection or liver transplantation. In this study, we demonstrate that a deep learning approach based on contrast-enhanced MR and 3D convolutional neural networks (CNN) can be applied to better predict MVI in HCC patients.Materials and MethodsThis retrospective study included 114 consecutive patients who were surgically resected from October 2012 to October 2018 with 117 histologically confirmed HCC. MR sequences including 3.0T/LAVA (liver acquisition with volume acceleration) and 3.0T/e-THRIVE (enhanced T1 high resolution isotropic volume excitation) were used in image acquisition of each patient. First, numerous 3D patches were separately extracted from the region of each lesion for data augmentation. Then, 3D CNN was utilized to extract the discriminant deep features of HCC from contrast-enhanced MR separately. Furthermore, loss function for deep supervision was designed to integrate deep features from multiple phases of contrast-enhanced MR. The dataset was divided into two parts, in which 77 HCCs were used as the training set, while the remaining 40 HCCs were used for independent testing. Receiver operating characteristic curve (ROC) analysis was adopted to assess the performance of MVI prediction. The output probability of the model was assessed by the independent student’s t-test or Mann-Whitney U test.ResultsThe mean AUC values of MVI prediction of HCC were 0.793 (p=0.001) in the pre-contrast phase, 0.855 (p=0.000) in arterial phase, and 0.817 (p=0.000) in the portal vein phase. Simple concatenation of deep features using 3D CNN derived from all the three phases improved the performance with the AUC value of 0.906 (p=0.000). By comparison, the proposed deep learning model with deep supervision loss function produced the best results with the AUC value of 0.926 (p=0.000).ConclusionA deep learning framework based on 3D CNN and deeply supervised net with contrast-enhanced MR could be effective for MVI prediction.

COVID-19 pneumonia diagnosis using a simple 2D deep learning framework with a single chest CT image (Preprint)

2020 ◽  
Author(s):  
Jinseok Lee
Keyword(s):  
Deep Learning ◽  
Diagnostic Performance ◽  
Ct Images ◽  
Chest Ct ◽  
University Hospital ◽  
Detection Accuracy ◽  
Ct Image ◽  
Test Dataset ◽  
Learning Framework ◽  
Testing Dataset

BACKGROUND The coronavirus disease (COVID-19) has explosively spread worldwide since the beginning of 2020. According to a multinational consensus statement from the Fleischner Society, computed tomography (CT) can be used as a relevant screening tool owing to its higher sensitivity for detecting early pneumonic changes. However, physicians are extremely busy fighting COVID-19 in this era of worldwide crisis. Thus, it is crucial to accelerate the development of an artificial intelligence (AI) diagnostic tool to support physicians. OBJECTIVE We aimed to quickly develop an AI technique to diagnose COVID-19 pneumonia and differentiate it from non-COVID pneumonia and non-pneumonia diseases on CT. METHODS A simple 2D deep learning framework, named fast-track COVID-19 classification network (FCONet), was developed to diagnose COVID-19 pneumonia based on a single chest CT image. FCONet was developed by transfer learning, using one of the four state-of-art pre-trained deep learning models (VGG16, ResNet50, InceptionV3, or Xception) as a backbone. For training and testing of FCONet, we collected 3,993 chest CT images of patients with COVID-19 pneumonia, other pneumonia, and non-pneumonia diseases from Wonkwang University Hospital, Chonnam National University Hospital, and the Italian Society of Medical and Interventional Radiology public database. These CT images were split into a training and a testing set at a ratio of 8:2. For the test dataset, the diagnostic performance to diagnose COVID-19 pneumonia was compared among the four pre-trained FCONet models. In addition, we tested the FCONet models on an additional external testing dataset extracted from the embedded low-quality chest CT images of COVID-19 pneumonia in recently published papers. RESULTS Of the four pre-trained models of FCONet, the ResNet50 showed excellent diagnostic performance (sensitivity 99.58%, specificity 100%, and accuracy 99.87%) and outperformed the other three pre-trained models in testing dataset. In additional external test dataset using low-quality CT images, the detection accuracy of the ResNet50 model was the highest (96.97%), followed by Xception, InceptionV3, and VGG16 (90.71%, 89.38%, and 87.12%, respectively). CONCLUSIONS The FCONet, a simple 2D deep learning framework based on a single chest CT image, provides excellent diagnostic performance in detecting COVID-19 pneumonia. Based on our testing dataset, the ResNet50-based FCONet might be the best model, as it outperformed other FCONet models based on VGG16, Xception, and InceptionV3.

Diagnostic performance of imaging features in the HBP of gadoxetate disodium-enhanced MRI for microvascular invasion in hepatocellular carcinoma: a meta-analysis

2021 ◽  
pp. 028418512110388
Author(s):  
Yuhui Deng ◽  
Dawei Yang ◽  
Hui Xu ◽  
Ahong Ren ◽  
Zhenghan Yang
Keyword(s):  
Hepatocellular Carcinoma ◽  
Diagnostic Performance ◽  
Meta Analysis ◽  
Microvascular Invasion ◽  
Index Test ◽  
Imaging Features ◽  
Gadoxetate Disodium ◽  
Reference Test ◽  
Tumor Margin ◽  
Meta Regression

Background Microvascular invasion (MVI) is a major risk factor for early recurrence in patients with hepatocellular carcinoma (HCC). Preoperative accurate evaluation of the presence of MVI could enormously benefit its treatment and prognosis. Purpose To evaluate and compare the diagnostic performance of two imaging features (non-smooth tumor margin and peritumor hypointensity) in the hepatobiliary phase (HBP) to preoperatively diagnose the presence of MVI in HCC. Material and Methods Original articles were collected from Medline/PubMed, Web of Science, EMBASE, and the Cochrane Library up to 17 January 2021 linked to gadoxetate disodium–enhanced magnetic resonance imaging (MRI) on 1.5 or 3.0 T. The pooled sensitivity, specificity, and summary area under the receiver operating characteristic curve (AUC) were calculated and meta-regression analyses were performed. Results A total of 14 original articles involving 2193 HCCs were included. The pooled sensitivity and specificity of non-smooth tumor margin and peritumor hypointensity were 73% and 61%, and 43% and 90%, respectively, for the diagnosis of MVI in HCC. The summary AUC of non-smooth tumor margin (0.74) was comparable to that of peritumor hypointensity (0.76) ( z = 0.693, P = 0.488). The meta-regression analysis identified four covariates as possible sources of heterogeneity: average size; time interval between index test and reference test; blindness to index test during reference test; and risk of bias score. Conclusion This meta-analysis showed moderate and comparable accuracy for predicting MVI in HCC using either non-smooth tumor margin or peritumor hypointensity in HBP. Four discovered covariates accounted for the heterogeneity.

scholarly journals A CT-based radiomics nomogram for predicting prognosis of coronavirus disease 2019 (COVID-19) radiomics nomogram predicting COVID-19

2021 ◽  
Vol 94 (1117) ◽  
pp. 20200634
Author(s):  
Hang Chen ◽  
Ming Zeng ◽  
Xinglan Wang ◽  
Liping Su ◽  
Yuwei Xia ◽  
...  
Keyword(s):  
Clinical Decision Making ◽  
Predictive Performance ◽  
Clinical Decision ◽  
Ct Images ◽  
Prediction Performance ◽  
Clinical Factors ◽  
Training Set ◽  
Clinical Model ◽  
Potential Biomarker ◽  
Validation Set

Objectives: To identify the value of radiomics method derived from CT images to predict prognosis in patients with COVID-19. Methods: A total of 40 patients with COVID-19 were enrolled in the study. Baseline clinical data, CT images, and laboratory testing results were collected from all patients. We defined that ROIs in the absorption group decreased in the density and scope in GGO, and ROIs in the progress group progressed to consolidation. A total of 180 ROIs from absorption group (n = 118) and consolidation group (n = 62) were randomly divided into a training set (n = 145) and a validation set (n = 35) (8:2). Radiomics features were extracted from CT images, and the radiomics-based models were built with three classifiers. A radiomics score (Rad-score) was calculated by a linear combination of selected features. The Rad-score and clinical factors were incorporated into the radiomics nomogram construction. The prediction performance of the clinical factors model and the radiomics nomogram for prognosis was estimated. Results: A total of 15 radiomics features with respective coefficients were calculated. The AUC values of radiomics models (kNN, SVM, and LR) were 0.88, 0.88, and 0.84, respectively, showing a good performance. The C-index of the clinical factors model was 0.82 [95% CI (0.75–0.88)] in the training set and 0.77 [95% CI (0.59–0.90)] in the validation set. The radiomics nomogram showed optimal prediction performance. In the training set, the C-index was 0.91 [95% CI (0.85–0.95)], and in the validation set, the C-index was 0.85 [95% CI (0.69–0.95)]. For the training set, the C-index of the radiomics nomogram was significantly higher than the clinical factors model (p = 0.0021). Decision curve analysis showed that radiomics nomogram outperformed the clinical model in terms of clinical usefulness. Conclusions: The radiomics nomogram based on CT images showed favorable prediction performance in the prognosis of COVID-19. The radiomics nomogram could be used as a potential biomarker for more accurate categorization of patients into different stages for clinical decision-making process. Advances in knowledge: Radiomics features based on chest CT images help clinicians to categorize the patients of COVID-19 into different stages. Radiomics nomogram based on CT images has favorable predictive performance in the prognosis of COVID-19. Radiomics act as a potential modality to supplement conventional medical examinations.

Genomic characterization of HBV-infected hepatocellular carcinoma patients using circulating tumor DNA.

2020 ◽  
Vol 38 (4_suppl) ◽  
pp. 575-575
Author(s):  
Daniel Lin ◽  
Zhong Ye ◽  
Chun Wang ◽  
Qiang Wei ◽  
Li Bingshan ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Cancer Patients ◽  
Early Stage ◽  
Circulating Tumor Dna ◽  
University Hospital ◽  
Clinical Factors ◽  
Characteristic Analysis ◽  
Genomic Aberrations ◽  
Mutant Genes ◽  
Tumor Dna

575 Background: Hepatocellular carcinoma (HCC) is a leading cause of mortality, with Hepatitis B virus (HBV) infection as a dominant etiology. Surgery or ablation may be curative for early-stage HCC. Thus, effective detection strategies are needed. We investigated genomic aberrations in circulating tumor DNA (ctDNA) as a potential diagnostic marker of HCC in HBV-infected patients. Methods: We identified early stage (BCLC 0-A) HCC cases (n = 21) and cancer-free controls (n = 15) from a cohort of Asian patients with HBV, undergoing surveillance at Thomas Jefferson University Hospital between 2013-2017. Blood samples were collected. Circulating cell-free DNA was isolated from plasma and assayed by capture-based next-generation sequencing of a targeted panel of 23 genes implicated in HCC pathogenesis. Sequencing data analysis and somatic mutation identification were conducted using a computational pipeline. Using area under the curve (AUC) in receiver operating characteristic analysis, we evaluated gene alterations and clinical factors (age, gender, cirrhosis) in an exploratory early detection HCC model. Results: Mutant ARID1A, ATM, CDKN2A, CTNNB1, ERBB2, TP53 genes were increased in HCC cases relative to non-cancer patients (85.7% vs 53.3%, P = 0.058; 42.9% vs 6.7%, P = 0.025; 38.1% vs 6.7%, P = 0.051; 42.9% vs 0%, P = 0.005; 52.4% vs 13.3%, P = 0.016; 100% vs 66.7%, P = 0.008, respectively). HCC patients had higher prevalence of cirrhosis than controls (90.5% vs. 60%, P = 0.046). Using the 6 mutant genes alone, the AUC for discriminating HCC from non-cancer patients was 0.827 (95% confidence interval [CI]: 0.701-0.953), which was greater than the AUC for discriminating cirrhosis from non-cirrhosis (0.531). When the 6 mutant genes were combined with clinical factors, the AUC of the exploratory HCC detection model increased to 0.914 ( P= 0.045). Conclusions: We identified 6 genomic aberrations in ctDNA that were more prevalent in HCC patients compared with non-cancer patients. Combining these alterations with clinical factors may identify HCC in HBV-infected patients at an early stage. These findings warrant further validation in future studies.

scholarly journals Prediction of 5-year Progression-Free Survival in Advanced Nasopharyngeal Carcinoma with Pretreatment PET/CT using Multi-Modality Deep Learning-based Radiomics

2021 ◽  
Author(s):  
Bingxin Gu ◽  
Mingyuan Meng ◽  
Lei Bi ◽  
Jinman Kim ◽  
David Dagan Feng ◽  
...  
Keyword(s):  
Deep Learning ◽  
Medical Image Analysis ◽  
Progression Free Survival ◽  
Ct Images ◽  
Tnm Stage ◽  
Support Vector ◽  
Free Survival ◽  
Prognostic Performance ◽  
Pet Ct ◽  
Testing Error

Abstract Purpose Deep Learning-based Radiomics (DLR) has achieved great success on medical image analysis. In this study, we aimed to explore the capability of our proposed end-to-end multi-modality DLR model using pretreatment PET/CT images to predict 5-year Progression-Free Survival (PFS) in advanced NPC.Methods A total of 170 patients with pathological confirmed advanced NPC (TNM stage III or IVa) were enrolled in this study. A 3D Convolutional Neural Network (CNN), with two branches to process PET and CT separately, was optimized to extract deep features from pretreatment multi-modality PET/CT images and use the derived features to predict the probability of 5-year PFS. Optionally, TNM stage, as a high-level clinical feature, can be integrated into our DLR model to further improve prognostic performance. Results For a comparison between Conventional Radiomic (CR) and DLR, 1456 handcrafted features were extracted, and three top CR methods, Random Forest (RF) + RF (AUC = 0.796 ± 0.009, testing error = 0.267 ± 0.007), RF + Adaptive Boosting (AdaBoost) (AUC = 0.783 ± 0.011, testing error = 0.286 ± 0.009), and L1-Logistic Regression (L1-LOG) + Kernel Support Vector Machines (KSVM) (AUC = 0.769 ± 0.008, testing error = 0.298 ± 0.006), were selected as benchmarks from 54 combinations of 6 feature selection methods and 9 classification methods. Compared to the three CR methods, our multi-modality DLR models using both PET and CT, with or without TNM stage (named PCT or PC model), resulted in the highest prognostic performance (PCT model: AUC = 0.842 ± 0.034, testing error = 0.194 ± 0.029; PC model: AUC = 0.825 ± 0.041, testing error = 0.223 ± 0.035). Furthermore, the multi-modality PCT model outperformed single-modality DLR models using only PET and TNM stage (named PT model: AUC = 0.818 ± 0.029, testing error = 0.218 ± 0.024) or only CT and TNM stage (named CT model: AUC = 0.657 ± 0.055, testing error = 0.375 ± 0.048). Conclusion Our study identified potential radiomics-based prognostic model for survival prediction in advanced NPC, and suggests that DLR could serve as a tool for aiding in cancer management.

Sign in / Sign up

Export Citation Format

Share Document