Virtual biopsy in abdominal pathology: where do we stand?

Area of Science:

• Medical Imaging
• Pathology
• Artificial Intelligence

Background:

• Traditional biopsies are invasive and may not capture full tissue heterogeneity.
• Virtual biopsy (VB) emerges as a non-invasive alternative using radiological images.
• VB integrates radiology, pathology, genomics, and AI for quantitative tissue analysis.

Purpose of the Study:

• To review the concept and pipeline development of virtual biopsy in abdominal pathology.
• To explore the potential benefits of VB as a non-invasive diagnostic tool.

Main Methods:

• Radiomics extracts quantitative features from radiological images.
• Artificial intelligence models process radiomic features to predict pathological characteristics.
• Data correlation links imaging findings with phenotypic expression and molecular status.

Main Results:

• VB provides quantitative information from radiological images, inaccessible by visual inspection.
• Current VB output approximates traditional biopsy findings.
• Future advancements in imaging and processing promise enhanced VB capabilities.

Conclusions:

• Virtual biopsy, powered by radiomics and AI, offers a non-invasive approach to abdominal pathology.
• VB has the potential to partially replace traditional biopsies due to its non-invasive nature, cost-effectiveness, and ability to account for lesion heterogeneity.
• Continued development in imaging resolution and processing techniques will be crucial for realizing the full potential of VB.