3D Fractal Dimension Analysis in CT Predicts Microvascular Invasion in HCC

Overview

This study demonstrates that fractal dimension (FD) analysis of contrast-enhanced CT images can objectively predict microvascular invasion (MVI) in hepatocellular carcinoma (HCC). The fractal approach quantifies tumor morphological complexity, correlating with MVI status and potentially informing prognosis and surgical planning.

Background

Hepatocellular carcinoma is a leading cause of cancer mortality, with microvascular invasion serving as a critical prognostic factor linked to tumor recurrence and poor survival. Preoperative identification of MVI is challenging since definitive diagnosis requires histopathology from surgical specimens. Conventional imaging features and serum markers have limitations due to subjectivity and inconsistent predictive value. Fractal analysis offers a quantitative method to assess tumor vascular complexity and heterogeneity, which may reflect MVI presence.

Data Highlights

The study retrospectively included HCC patients from two centers, divided into training, internal test, and external test cohorts. CT images were standardized and tumors segmented in 3D. Fractal dimensions were computed using box-counting methods and curve fitting to characterize tumor morphology. Clinical and pathological data including AFP, tumor size, and histologic grade were collected. Inter- and intra-reader reproducibility of segmentation was assessed in 100 patients.

Key Findings

• Fractal dimension analysis of portal venous phase CT images quantitatively characterizes tumor complexity associated with MVI in HCC.
• FD metrics demonstrated potential as objective imaging biomarkers to predict MVI preoperatively, overcoming limitations of subjective radiological features.
• Higher fractal dimension values correlated with the presence of MVI and adverse histopathological features.
• Reproducibility of tumor segmentation and fractal analysis was confirmed, supporting clinical applicability.
• Fractal analysis may aid in risk stratification and guide surgical margin decisions and adjuvant therapy planning.

Clinical Implications

Incorporating fractal dimension analysis into preoperative CT evaluation can enhance the prediction of microvascular invasion in HCC patients, enabling more tailored surgical and therapeutic strategies. This objective quantitative method may reduce interobserver variability inherent in traditional imaging assessments and improve prognostic accuracy.

Conclusion

Fractal dimension analysis of contrast-enhanced CT images is a feasible and promising tool for noninvasive prediction of microvascular invasion in hepatocellular carcinoma, with potential to impact clinical decision-making and patient outcomes.

References

1. 1,2 -- MVI as a prognostic factor in HCC
2. 3,4,5 -- Surgical management considerations in HCC with MVI
3. 6,7 -- MVI impact on liver transplantation outcomes
4. 8,9,10 -- Radiological features associated with MVI
5. 11 -- Interobserver variability in imaging MVI
6. 12,13 -- Fractal analysis methodology and applications
7. 14,15 -- Fractal analysis in cancer imaging
8. 16,17 -- FD quantification of tumor margins and prognosis
9. 18,19 -- FD in assessing tumor heterogeneity and microvascular complexity
10. 20 -- Histopathological definition of MVI