Deep Learning for Approaching Hepatocellular Carcinoma Ultrasound Screening Dilemma: Identification of α-Fetoprotein-Negative Hepatocellular Carcinoma From Focal Liver Lesion Found in High-Risk Patients

Abstract

BackgroundFirst-line surveillance on hepatitis B virus (HBV)-infected populations with B-mode ultrasound is relatively limited to identifying hepatocellular carcinoma (HCC) without elevated α-fetoprotein (AFP). To improve the present HCC surveillance strategy, the state of the art of artificial intelligence (AI), a deep learning (DL) approach, is proposed to assist in the diagnosis of a focal liver lesion (FLL) in HBV-infected liver background.MethodsOur proposed deep learning model was based on B-mode ultrasound images of surgery that proved 209 HCC and 198 focal nodular hyperplasia (FNH) cases with 413 lesions. The model cohort and test cohort were set at a ratio of 3:1, in which the test cohort was composed of AFP-negative HBV-infected cases. Four additional deep learning models (MobileNet, Resnet50, DenseNet121, and InceptionV3) were also constructed as comparative baselines. To evaluate the models in terms of diagnostic power, sensitivity, specificity, accuracy, confusion matrix, F1-score, and area under the receiver operating characteristic curve (AUC) were calculated in the test cohort.ResultsThe AUC of our model, Xception, achieved 93.68% in the test cohort, superior to other baselines (89.06%, 85.67%, 83.94%, and 78.13% respectively for MobileNet, Resnet50, DenseNet121, and InceptionV3). In terms of diagnostic power, our model showed sensitivity, specificity, accuracy, and F1-score of 96.08%, 76.92%, 86.41%, and 87.50%, respectively, and PPV, NPV, FPR, and FNR calculated from the confusion matrix were respectively 80.33%, 95.24%, 23.08%, and 3.92% in identifying AFP-negative HCC from HBV-infected FLL cases. Satisfactory robustness of our proposed model was shown based on 5-fold cross-validation performed among the models above.ConclusionsOur DL approach has great potential to assist B-mode ultrasound in identifying AFP-negative HCC from FLL found in surveillance of HBV-infected patients.

Artificial Intelligence and Machine Learning in Hepatocellular Carcinoma Screening, Diagnosis and Treatment - A Comprehensive Systematic Review

“…Zhang et al, identified hepatocellular carcinoma (HCC) cases testing negative for alpha-fetoprotein (AFP) using deep learning techniques in ultrasonic screening. Xception, an AI-based deep learning algorithm, demonstrated great sensitivity and specificity in identifying AFP-negative HCC in patients at high risk [34]. Researchers employed XGBoost and 3D-CNN models to accurately detect hepatocellular carcinoma (HCC) microvascular invasion (MVI) before surgery.…”

“…Alpha-Fetoprotein (AFP)-negative Hepatocellular Carcinoma (HCC) using B-mode ultrasonography. The strong diagnostic capability of the Xception model indicates its potential as a helpful aid for patients afflicted with HBV [34]. In a separate study, XGBoost and 3D-Convolutional Neural Network (3D-CNN) models were employed, utilizing CT data to provide preoperative predictions of microvascular invasion (MVI).…”

Perioperative predictors of outcome of hepatectomy for HBV-related hepatocellular carcinoma

“…For example, using omics data, machine learning (ML) and deep learning (DL) are used to predict metastasis ( 157 ). AI has been widely reported in the early diagnosis of HCC, such as the prediction model of HBV reverse transcriptase sequence combined with ML, establishing genome-wide interrogation of somatic copy number aberrations by ML as a non-invasive HCC detection, combining B-mode ultrasound with DL to identify AFP negative HCC ( 158 – 160 ). In terms of prognosis, artificial neural network model is helpful to preoperatively evaluate the HCC patients’ risk of liver failure and predict life quality after hepatectomy ( 161 , 162 ).…”

Application of artificial intelligence in the diagnosis of hepatocellular carcinoma

“…48 A DL model detailed by Zhang et al highlighted the potential of AI as a screening tool for AFP-negative HCCs. 54 The CNN model (Xception 55 ) was trained using a total of 305 images of HCC and focal nodular hyperplasia taken using B-mode ultrasound and model testing was done using 102 B-mode ultrasound images. HCC staging, lesion size, echogenicity and liver function were heterogenous in both training and testing data sets.…”