Comparison of radiomics models and dual-energy material decomposition to decipher abdominal lymphoma in contrast-enhanced CT - Report - MDSpire

Comparison of radiomics models and dual-energy material decomposition to decipher abdominal lymphoma in contrast-enhanced CT

  • By

  • Simon Bernatz

  • Vitali Koch

  • Daniel Pinto Dos Santos

  • Jörg Ackermann

  • Leon D. Grünewald

  • Inga Weitkamp

  • Ibrahim Yel

  • Simon S. Martin

  • Lukas Lenga

  • Jan-Erik Scholtz

  • Thomas J. Vogl

  • Scherwin Mahmoudi

  • March 6, 2023

  • 0 min

Share

Radiomics and Dual-Energy CT for Abdominal Lymphoma Analysis

Overview

This study evaluated the use of radiomics features combined with dual-energy CT (DECT) material decomposition to stratify abdominal lymphoma in contrast-enhanced CT scans. The combined approach was hypothesized to improve differentiation of malignant lymph nodes compared to DECT features alone.

Background

Lymph node involvement in lymphoma is typically assessed by size and number on cross-sectional imaging, but borderline cases remain challenging. Positron emission tomography (PET) CT can aid diagnosis but involves high radiation and cost. Dual-energy CT (DECT) with iodine quantification offers a non-invasive alternative to characterize tissue vascularity. Radiomics extracts high-dimensional quantitative imaging features that describe tumor heterogeneity and phenotype, potentially improving tumor characterization and outcome prediction. The combined use of DECT and radiomics for lymphoma stratification has not been previously evaluated.

Data Highlights

Seventy-two patients with contrast-enhanced abdominal DECT scans were included, with visually unequivocal malignant lymph nodes >1.5 cm confirmed by histopathology. DECT iodine density (ID), normalized iodine uptake (ID%), and fat fraction were quantified. Radiomics features were extracted from single-energy images. Imaging was performed on a third-generation dual-source DECT scanner with standardized contrast injection and acquisition protocols.

Key Findings

  • DECT iodine quantification provides surrogate markers of tumor vascularity useful for lymph node characterization.
  • Radiomics features capture tissue heterogeneity and shape, offering additional quantitative data beyond standard imaging metrics.
  • The combined application of radiomics and DECT features has potential to improve stratification accuracy of abdominal lymphoma compared to DECT alone.
  • Quantitative imaging biomarkers may help focus radiologists’ attention on borderline or indistinct lymph nodes, potentially accelerating diagnosis.
  • Use of DECT and radiomics avoids additional radiation exposure associated with PET-CT and may reduce need for further invasive work-up.

Clinical Implications

Integrating radiomics with DECT iodine quantification could enhance non-invasive differentiation of malignant lymph nodes in lymphoma, aiding early diagnosis and treatment planning. This combined imaging biomarker approach may reduce reliance on costly or high-radiation modalities like PET-CT and streamline radiological workflows by prioritizing ambiguous cases.

Conclusion

The study supports the potential of combining radiomics and dual-energy CT features to improve abdominal lymphoma stratification in contrast-enhanced CT scans. This approach may serve as a valuable non-invasive imaging biomarker to assist clinical decision-making.

References

  1. 1 -- Criteria for lymph node involvement in lymphoma imaging
  2. 2 -- Size thresholds for pathological lymph nodes
  3. 3,4 -- Challenges in borderline lymph node diagnosis
  4. 5 -- PET-CT limitations in lymphoma
  5. 6-7 -- DECT material decomposition techniques
  6. 8-10 -- Iodine-selective imaging and material decomposition
  7. 11-12 -- Iodine quantification as tumor vascularity surrogate
  8. 13-15 -- DECT iodine quantification in lymph node differentiation
  9. 16-17 -- Radiomics in oncology
  10. 18-19 -- Radiomics and DECT in lymph node characterization
  11. 20 -- Imaging biomarkers to support radiologists
  12. 21 -- Definition of unequivocal malignant lymph nodes

Original Source(s)

Related Content