Yajun Li

Objectives: To investigate the potential of dual-energy computed tomography (DECT) parameters in identifying metastatic cervical lymph nodes in oral squamous cell carcinoma (OSCC) patients and to explore the relationships between DECT and pathological features. Methods: Clinical and DECT data were collected from patients who underwent radical resection of OSCC and cervical lymph node dissection between November 2019 and June 2021. Microvascular density was assessed using the Weidner counting method. The electron density (ED) and effective atomic number (Z_eff) in non - contrast phase and iodine concentration (IC), normalized IC, slope of the energy spectrum curve (λ_HU), and dual-energy index (DEI) in parenchymal phase were compared between metastatic and non - metastatic lymph nodes. Student’s t-test, Pearson’s rank correlation, and receiver operating characteristic curves were performed. Results: The inclusion criteria were met in 399 lymph nodes from 103 patients. Metastatic nodes (n = 158) displayed significantly decreased ED, IC, normalized IC, λ_HU, and DEI values compared with non-metastatic nodes (n = 241) (all p < 0.01). Strong correlations were found between IC (r = 0.776), normalized IC (r = 0.779), λ_HU (r = 0.738), DEI (r = 0.734), and microvascular density. Area under the curve (AUC) for normalized IC performed the highest (0.875) in diagnosing metastatic nodes. When combined with the width of nodes, AUC increased to 0.918. Conclusion: DECT parameters IC, normalized IC, λ_HU, and DEI reflect pathologic changes in lymph nodes to a certain extent, and aid for detection of metastatic cervical lymph nodes from OSCC. Key Points: • Electron density, iodine concentration, normalized iodine concentration, λ_HU, and dual-energy index values showed significant differences between metastatic and non-metastatic nodes. • Strong correlations were found between iodine concentration, normalized iodine concentration, slope of the spectral Hounsfield unit curve, dual-energy index, and microvascular density. • DECT qualitative parameters reflect the pathologic changes in lymph nodes to a certain extent, and aid for the detection of metastatic cervical lymph nodes from oral squamous cell carcinoma.

Achieving high feature reproducibility while preserving biological information is one of the main challenges for the generalizability of current radiomics studies. Non-clinical imaging variables, such as reconstruction kernels, have shown to significantly impact radiomics features. In this study, we retrain an open-source convolutional neural network (CNN) to harmonize computerized tomography (CT) images with various reconstruction kernels to improve feature reproducibility and radiomic model performance using epidermal growth factor receptor (EGFR) mutation prediction in lung cancer as a paradigm. In the training phase, the CNN was retrained and tested on 32 lung cancer patients’ CT images between two different groups of reconstruction kernels (smooth and sharp). In the validation phase, the retrained CNN was validated on an external cohort of 223 lung cancer patients’ CT images acquired using different CT scanners and kernels. The results showed that the retrained CNN could be successfully applied to external datasets with different CT scanner parameters, and harmonization of reconstruction kernels from sharp to smooth could significantly improve the performance of radiomics model in predicting EGFR mutation status in lung cancer. In conclusion, the CNN based method showed great potential in improving feature reproducibility and generalizability by harmonizing medical images with heterogeneous reconstruction kernels.

We evaluated whether the optimal selection of CT reconstruction settings enables the construction of a radiomics model to predict epidermal growth factor receptor (EGFR) mutation status in primary lung adenocarcinoma (LAC) using standard of care CT images. Fifty-one patients (EGFR:wildtype = 23:28) with LACs of clinical stage I/II/IIIA were included in the analysis. The LACs were segmented in four conditions, two slice thicknesses (Thin: 1 mm; Thick: 5 mm) and two convolution kernels (Sharp: B70f/B70s; Smooth: B30f/B31f/B31s), which constituted four groups: (1) Thin-Sharp, (2) Thin-Smooth, (3) Thick-Sharp, and (4) Thick-Smooth. Machine learning algorithms selected and combined 1,695 quantitative image features to build prediction models. The performance of prediction models was assessed by calculating the area under the curve (AUC). The best prediction model yielded AUC (95%CI) = 0.83 (0.68, 0.92) using the Thin-Smooth reconstruction setting. The AUC of models using thick slices was significantly lower than that of thin slices (P < 10 ⁻³ ), whereas the impact of reconstruction kernel was not significant. Our study showed that the optimal prediction of EGFR mutational status in early stage LACs was achieved by using thin CT-scan slices, independently of convolution kernels. Results from the prediction model suggest that tumor heterogeneity is associated with EGFR mutation.