Home›Oncology› Radiomics-based artificial intelligence models predict VETC presence in patients with hepatocellular carcinoma
Radiomics-based artificial intelligence models predict VETC presence in patients with hepatocellular carcinomaAI helps predict aggressive liver cancer growth patterns
Journal of medical Internet researchPublished July 4, 2026Study authors: Hua Xuefeng, Fu Rongdang, Yin ZiweiPubMed ↗DOI ↗Editorial oversight: Dr. Julia Lee, PhD · Oncology, Genomics & Drug Development
AI-generated summary of the cited source, checked by automated accuracy review.
How we work
Share
Key Takeaway
Recognize radiomics-based AI as a potential noninvasive tool to identify aggressive tumor features in hepatocellular carcinoma.
The study evaluated the diagnostic accuracy of radiomics-based artificial intelligence (AI) models across multiple imaging modalities, including CEMRI, CECT, CEUS, and PET/CT. The primary objective was to determine if these AI models could noninvasively predict the presence of vessel-encapsulated tumor clusters (VETCs) in patients with hepatocellular carcinoma.
The analysis reported that radiomics-based AI demonstrated favorable sensitivity and specificity for identifying VETCs. Furthermore, the results indicated a significant association between AI-predicted VETC positivity and an increased risk of early recurrence for these patients. These findings suggest that radiomics may serve as a valuable bridge between imaging phenotypes and biologically aggressive tumor characteristics.
Several limitations were noted by the authors, including the use of retrospective designs, geographically concentrated cohorts, and limited external validation. The researchers also highlighted heterogeneity in results and a risk of bias, leading to low-to-moderate certainty for some findings. Clinically, these AI models should be viewed as decision-support tools rather than replacements for histopathology or multidisciplinary clinical judgment.
Doctors often struggle to tell just by looking at a scan how aggressively a liver tumor will behave. A new study looked at using artificial intelligence (AI) to spot 'VETCs'—specific clusters of blood vessels that indicate a more aggressive form of liver cancer. By analyzing various imaging types, the AI showed high accuracy in identifying these risky patterns.
The analysis included over 1,300 patients and several different types of scans. The results showed that when the AI detected these specific markers, patients faced a significantly higher risk of their cancer returning quickly. This suggests that AI can act as a powerful tool to help doctors identify which patients might need more intensive monitoring or treatment.
While the technology shows great promise, there are important notes for its use today. Because the data came from mostly older records and specific regions, the results aren't perfect yet. Currently, these AI tools should be used to support a doctor's decision-making process rather than replacing the expert judgment of a medical team.
What this means for you:
AI can help identify aggressive liver cancer markers on scans to better predict how quickly a tumor might return.
Common questions
How accurate is the AI at finding these cancer markers?
The study found that AI using MRI scans was quite effective. It showed a sensitivity of 0.84 and a specificity of 0.79 in identifying the high-risk vessel clusters. This means it can reliably pick up these specific patterns on imaging to help doctors assess the tumor's behavior.
What does finding these markers mean for a patient?
When the AI identifies these specific markers (known as VETCs), it is linked to a higher risk of early cancer recurrence. Specifically, the study reported a hazard ratio of 2.34 for patients with positive results, meaning those tumors are more likely to return quickly.
Will AI replace doctors in diagnosing liver cancer?
No, these tools are not meant to replace human experts. The study suggests that AI should currently be used as a decision-support tool. It helps provide extra information for the medical team to use alongside their own judgment and traditional tests.
BACKGROUND: Vessels encapsulating tumor clusters (VETCs), a CD34-positive vascular pattern in hepatocellular carcinoma (HCC), are linked to aggressive biology, early recurrence, and poor survival. Because pathologic VETC assessment requires postoperative immunohistochemistry and may be affected by sampling, preoperative noninvasive prediction remains clinically important. Radiomics-based artificial intelligence (AI) applied to routine contrast-enhanced imaging may provide a surrogate marker, but evidence across has not been comprehensively appraised.
OBJECTIVE: This study aimed to evaluate the diagnostic accuracy and prognostic value of radiomics-based AI models for noninvasive VETC prediction in HCC using PICOTS (patient population, intervention, comparator, outcomes, timing, and setting) and PRISMA-DTA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses for Diagnostic Test Accuracy) frameworks.
METHODS: We searched PubMed, Embase, Web of Science, and the Cochrane Library, gray literature, and citations (original search July 11, 2025; updated April 17, 2026). The original search was completed on July 11, 2025, and the reconstructed strategy was rerun on April 17, 2026. Eligible retrospective cohort studies developed or validated radiomics or deep-learning models using contrast-enhanced magnetic resonance imaging (CEMRI), contrast-enhanced computed tomography (CECT), contrast-enhanced ultrasound (CEUS), or 2-[¹⁸F]fluoro-2-deoxy-D-glucose positron emission tomography or computed tomography ([18F]FDG PET/CT) to predict CD34-confirmed VETC and reported 2×2 diagnostic data and/or hazard ratios (HRs) for early recurrence. Mutually exclusive cohorts were treated as separate datasets only when patient overlap was absent. Risk of bias was assessed with the Prediction model Risk Of Bias Assessment Tool+AI, and certainty with GRADE (Grading of Recommendations, Assessment, Development, and Evaluation). Diagnostic accuracy was synthesized with bivariate random-effects models; prognostic HRs were pooled with restricted maximum likelihood+Hartung-Knapp-Sidik-Jonkman random-effects models.
RESULTS: In total, 15 studies (729 internal-validation and 613 external-validation patients) were included; 14 were from China and 1 from Japan. Moreover, 10 studies used CEMRI, 3 CECT, 1 CEUS, and 1 [18F]FDG PET/CT. CEMRI-based AI showed the best performance: sensitivity=0.84 (95% CI 0.73-0.93; 95% prediction interval [PI] 0.45-1.00), specificity=0.79 (95% CI 0.70-0.86; 95% PI 0.50-0.97), and area under the curve=0.87. Meta-regression suggested that center type, validation design, algorithm class, and magnetic resonance imaging field strength contributed to specificity heterogeneity. CECT, CEUS, and [18F]FDG PET/CT evidence was limited. AI-predicted VETC positivity was associated with early recurrence (HR 2.34, 95% CI 1.93-2.84). GRADE certainty ranged from low to moderate, mainly due to imprecision, risk of bias, and heterogeneity.
CONCLUSIONS: This review is innovative because it integrates diagnostic accuracy, modality comparison, algorithm performance, and recurrence prognosis for AI-based VETC prediction. Unlike previous modality-specific reviews, it clarifies what AI brings to the field: a potential preoperative bridge between imaging phenotypes and biologically aggressive HCC. Real-world use should remain cautious and decision-supportive, given retrospective designs, geographically concentrated cohorts, limited external validation, heterogeneity, risk of bias, and low-to-moderate GRADE certainty, rather than replacing histopathology or multidisciplinary clinical judgment in practice.
