Radiomics Based on Dual-Layer Detector Spectral Computed Tomography for Predicting Histologic Subtypes and Ki-67...

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Non-small cell lung cancer (NSCLC) diagnosis and prognosis are critical.
• Accurate prediction of pathologic subtypes and Ki-67 index is essential for treatment planning.
• Dual-layer detector spectral computed tomography (DLCT) offers advanced imaging capabilities.

Purpose of the Study:

• To develop and validate combined radiomics and clinical nomogram models using DLCT.
• To predict pathologic subtypes in NSCLC patients.
• To predict Ki-67 index status in NSCLC patients.

Main Methods:

• A total of 405 NSCLC patients from two medical centers were included.
• Radiomics features were extracted from DLCT images (dual-phase enhanced and iodine images).
• Clinical factors were analyzed, and nomograms integrating radiomics scores and clinical factors were constructed and validated.

Main Results:

• The combined nomogram achieved high AUCs for type classification (0.931 internal, 0.910 external validation).
• The nomogram also showed strong performance for Ki-67 classification (0.808 internal, 0.791 external validation).
• Decision curve analysis indicated superior net benefits of the nomogram compared to individual models.

Conclusions:

• Nomograms integrating DLCT-based radiomics and clinical data effectively predict NSCLC subtypes and Ki-67 status.
• These models show promising potential to aid clinical decision-making in NSCLC management.
• The study highlights the value of spectral CT radiomics in improving NSCLC characterization.