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

A cost-effective predictive tool for AFP-negative focal hepatic lesions of retrospective study: enhancing clinical triage and decision-making

Background

Identifying alpha-fetal protein (AFP)-negative focal hepatic lesions presents a significant challenge, particularly in China. We sought to develop an economically portable tool for the diagnosis of benign and malignant liver lesions with AFP-negative status, and explore its clinical diagnostic efficiency.

Methods

A retrospective study was conducted at Peking University Shenzhen Hospital from January 2017 to February 2023, including a total of 348 inpatients with AFP-negative liver space-occupying lesions. The study used a training set of 252 inpatients from January 2017 to September 2021 to establish a diagnostic model for differentiating benign and malignant AFP-negative liver space-occupying lesions. Additionally, a validation cohort of 96 inpatients from October 2021 to February 2023 was used to confirm the diagnostic performance of the model. From January 2017 to February 2023, patients at JingNing People's Hospital, Gansu Province were assigned to the external cohort (n = 78).

Results

A predictive tool was established by screening age, gender, hepatitis B virus (HBV)/hepatitis C virus (HCV) infected, single lesion, alanine amino transferase (ALT), and lymphocyte-to-monocyte ratio (LMR) using multivariate logistic regression analysis and clinical practice. The area under the curve (AUC) of the model was 0.911 (95% CI [0.873-0.949]) in the training set and 0.882 (95% CI [0.815-0.949]) in the validation cohort. In addition, the model achieved an area under the curve of 0.811 (95% CI [0.687-0.935]) in the external validation cohort.

Conclusion

Our results demonstrated that the predictive tool has the characteristics of good diagnostic efficiency, economy and convenience, which is helpful for the clinical triage and decision-making of AFP-negative liver space-occupying lesions.

Role of Imaging in Screening for Hepatocellular Carcinoma

Primary liver cancer is among the most common cancers globally. It is the sixth-most common malignancy encountered and the third-most common cause of cancer-related death. Hepatocellular carcinoma (HCC) is the most common primary liver malignancy, accounting for about 90% of primary liver cancers. The majority of HCCs occur in patients with underlying cirrhosis, which results from chronic liver diseases such as fatty liver, hepatitis B and hepatitis C infections, and chronic alcohol use, which are the leading causes. The obesity pandemic has led to an increased prevalence of nonalcoholic fatty liver disease (NAFLD), which leads to nonalcoholic steatohepatitis and could progress to cirrhosis. As HCC is among the most common cancers and occurs in the setting of chronic liver disease in most patients, screening the population at risk could help in early diagnosis and management, leading to improved survival. Screening for HCC is performed using biochemical marker testing such as α-fetoprotein (AFP) and cross-sectional imaging. It is critical to emphasize that HCC could potentially occur in patients without cirrhosis (non-cirrhotic HCC), which can account for almost 20% of all HCCs. The lack of cirrhosis can cause a delay in surveillance, which could potentially lead to diagnosis at a later stage, worsening the prognosis for such patients. In this article, we discuss the diagnosis of cirrhosis in at-risk populations with details on the different modalities available for screening HCC in patients with cirrhosis, emphasizing the role of abdominal ultrasounds, the primary imaging modality in HCC screening.

Artificial intelligence techniques in liver cancer

Hepatocellular Carcinoma (HCC), the most common primary liver cancer, is a significant contributor to worldwide cancer-related deaths. Various medical imaging techniques, including computed tomography, magnetic resonance imaging, and ultrasound, play a crucial role in accurately evaluating HCC and formulating effective treatment plans. Artificial Intelligence (AI) technologies have demonstrated potential in supporting physicians by providing more accurate and consistent medical diagnoses. Recent advancements have led to the development of AI-based multi-modal prediction systems. These systems integrate medical imaging with other modalities, such as electronic health record reports and clinical parameters, to enhance the accuracy of predicting biological characteristics and prognosis, including those associated with HCC. These multi-modal prediction systems pave the way for predicting the response to transarterial chemoembolization and microvascular invasion treatments and can assist clinicians in identifying the optimal patients with HCC who could benefit from interventional therapy. This paper provides an overview of the latest AI-based medical imaging models developed for diagnosing and predicting HCC. It also explores the challenges and potential future directions related to the clinical application of AI techniques.

Diagnostic Performance of Artificial Intelligence in Detection of Hepatocellular Carcinoma: A Meta-analysis

Due to the increasing interest in the use of artificial intelligence (AI) algorithms in hepatocellular carcinoma detection, we performed a systematic review and meta-analysis to pool the data on diagnostic performance metrics of AI and to compare them with clinicians' performance. A search in PubMed and Scopus was performed in January 2024 to find studies that evaluated and/or validated an AI algorithm for the detection of HCC. We performed a meta-analysis to pool the data on the metrics of diagnostic performance. Subgroup analysis based on the modality of imaging and meta-regression based on multiple parameters were performed to find potential sources of heterogeneity. The risk of bias was assessed using Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) and Prediction Model Study Risk of Bias Assessment Tool (PROBAST) reporting guidelines. Out of 3177 studies screened, 44 eligible studies were included. The pooled sensitivity and specificity for internally validated AI algorithms were 84% (95% CI: 81,87) and 92% (95% CI: 90,94), respectively. Externally validated AI algorithms had a pooled sensitivity of 85% (95% CI: 78,89) and specificity of 84% (95% CI: 72,91). When clinicians were internally validated, their pooled sensitivity was 70% (95% CI: 60,78), while their pooled specificity was 85% (95% CI: 77,90). This study implies that AI can perform as a diagnostic supplement for clinicians and radiologists by screening images and highlighting regions of interest, thus improving workflow.

Malignancy diagnosis of liver lesion in contrast enhanced ultrasound using an end-to-end method based on deep learning

BACKGROUND

Contrast-enhanced ultrasound (CEUS) is considered as an efficient tool for focal liver lesion characterization, given it allows real-time scanning and provides dynamic tissue perfusion information. An accurate diagnosis of liver lesions with CEUS requires a precise interpretation of CEUS images. However,it is a highly experience dependent task which requires amount of training and practice. To help improve the constrains, this study aims to develop an end-to-end method based on deep learning to make malignancy diagnosis of liver lesions using CEUS.

METHODS

A total of 420 focal liver lesions with 136 benign cases and 284 malignant cases were included. A deep learning model based on a two-dimensional convolution neural network, a long short-term memory (LSTM), and a linear classifier (with sigmoid) was developed to analyze the CEUS loops from different contrast imaging phases. For comparison, a 3D-CNN based method and a machine-learning (ML)-based time-intensity curve (TIC) method were also implemented for performance evaluation.

RESULTS

Results of the 4-fold validation demonstrate that the mean AUC is 0.91, 0.88, and 0.78 for the proposed method, the 3D-CNN based method, and the ML-based TIC method, respectively.

CONCLUSIONS

The proposed CNN-LSTM method is promising in making malignancy diagnosis of liver lesions in CEUS without any additional manual features selection.

Deep Learning Methods in Medical Image-Based Hepatocellular Carcinoma Diagnosis: A Systematic Review and Meta-Analysis

(1) Background: The aim of our research was to systematically review papers specifically focused on the hepatocellular carcinoma (HCC) diagnostic performance of DL methods based on medical images. (2) Materials: To identify related studies, a comprehensive search was conducted in prominent databases, including Embase, IEEE, PubMed, Web of Science, and the Cochrane Library. The search was limited to studies published before 3 July 2023. The inclusion criteria consisted of studies that either developed or utilized DL methods to diagnose HCC using medical images. To extract data, binary information on diagnostic accuracy was collected to determine the outcomes of interest, namely, the sensitivity, specificity, and area under the curve (AUC). (3) Results: Among the forty-eight initially identified eligible studies, thirty studies were included in the meta-analysis. The pooled sensitivity was 89% (95% CI: 87-91), the specificity was 90% (95% CI: 87-92), and the AUC was 0.95 (95% CI: 0.93-0.97). Analyses of subgroups based on medical image methods (contrast-enhanced and non-contrast-enhanced images), imaging modalities (ultrasound, magnetic resonance imaging, and computed tomography), and comparisons between DL methods and clinicians consistently showed the acceptable diagnostic performance of DL models. The publication bias and high heterogeneity observed between studies and subgroups can potentially result in an overestimation of the diagnostic accuracy of DL methods in medical imaging. (4) Conclusions: To improve future studies, it would be advantageous to establish more rigorous reporting standards that specifically address the challenges associated with DL research in this particular field.

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

Hepatitis B virus (HBV) is identified as a major risk factor for hepatocellular carcinoma (HCC), resulting in so-called hepatitis B virus-related hepatocellular carcinoma (HBV-related HCC). Hepatectomy for HCC is acknowledged as an efficient treatment strategy, especially for early HCC. Furthermore, patients with advanced HCC can still obtain survival benefits through surgical treatment combined with neoadjuvant therapy, adjuvant therapy, transcatheter arterial chemoembolization, and radiofrequency ablation. Therefore, preoperative and postoperative predictors of HBV-related HCC have crucial indicative functions for the follow-up treatment of patients with feasible hepatectomy. This review covers a variety of research results on preoperative and postoperative predictors of hepatectomy for HBV-related HCC over the past decade and in previous landmark studies. The relevant contents of Hepatitis C virus-related HCC, non-HBV non-HCV HCC, and the artificial intelligence application in this field are briefly addressed in the extended content. Through the integration of this review, a large number of preoperative and postoperative factors can predict the prognosis of HBV-related HCC, while most of the predictors have no standardized thresholds. According to the characteristics, detection methods, and application of predictors, the predictors can be divided into the following categories: 1. serological and hematological predictors, 2. genetic, pathological predictors, 3. imaging predictors, 4. other predictors, 5. analysis models and indexes. Similar results appear in HCV-related HCC, non-HBV non-HCV HCC. Predictions based on AI and big biological data are actively being applied. A reasonable prediction model should be established based on the economic, health, and other levels in specific countries and regions.