Radiomics prognostic analysis of PET/CT images in a multicenter head and neck cancer cohort: investigating ComBat strategies, sub-volume characterization, and automatic segmentation

Abstract

Purpose This study aimed to investigate the impact of several ComBat harmonization strategies, intra-tumoral sub-volume characterization, and automatic segmentations for progression-free survival (PFS) prediction through radiomics modeling for patients with head and neck cancer (HNC) in PET/CT images. Methods The HECKTOR MICCAI 2021 challenge set containing PET/CT images and clinical data of 325 oropharynx HNC patients was exploited. A total of 346 IBSI-compliant radiomic features were extracted for each patient's primary tumor volume defined by the reference manual contours. Modeling relied on least absolute shrinkage Cox regression (Lasso-Cox) for feature selection (FS) and Cox proportional-hazards (CoxPH) models were built to predict PFS. Within this methodological framework, 8 different strategies for ComBat harmonization were compared, including before or after FS, in feature groups separately or all features directly, and with center or clustering-determined labels. Features extracted from tumor sub-volume clustering were also investigated for their prognostic additional value. Finally, 3 automatic segmentations (2 threshold-based and a 3D U-Net) were also compared. All results were evaluated with the concordance index (C-index). Results Radiomics features without harmonization, combined with clinical factors, led to models with C-index values of 0.69 in the testing set. The best version of ComBat harmonization, i.e., after FS, for feature groups separately and relying on clustering-determined labels, achieved a C-index of 0.71. The use of features extracted from tumor sub-volumes further improved the C-index to 0.72. Models that relied on the automatic segmentations yielded close but slightly lower prognostic performance (0.67-0.70) compared to reference contours. Conclusion A standard radiomics pipeline allowed for prediction of PFS in a multicenter HNC cohort. Applying a specific strategy of ComBat harmonization improved the performance. The extraction of intra-tumoral sub-volume features and automatic segmentation could contribute to the improvement and automation of prognosis modeling, respectively.

An Innovative and Efficient Diagnostic Prediction Flow for Head and Neck Cancer: A Deep Learning Approach for Multi-Modal Survival Analysis Prediction Based on Text and Multi-Center PET/CT Images

“…By combining these modalities, clinicians can obtain a more comprehensive understanding of the patient's condition, leading to improved prognostic predictions and treatment planning. previous work [34,35] also fully explored the PET/CT fusion modality and the diagnosis and prognosis of head and neck cancer in the radiomics dimension, respectively.…”

A novel loss function to reproduce texture features for deep learning‐based MRI‐to‐CT synthesis

“…The reproducibility of the radiomics features is an essential aspect of radiomic studies since the nonreproducibility may add noise to the data and enhance the confounding factors in statistical models. 35,36 The deep learning method was expected to improve the reproducibility of radiomic features and thus the accuracy and reliability of clinical studies. [37][38][39] However, as mentioned by Ke Nie et al, 40 the deep learning algorithm widened the scope of imaging but also can generate unrealistic artifacts into images for radiomics studies.…”

“…The reproducibility of the radiomics features is an essential aspect of radiomic studies since the non‐reproducibility may add noise to the data and enhance the confounding factors in statistical models 35,36 . The deep learning method was expected to improve the reproducibility of radiomic features and thus the accuracy and reliability of clinical studies 37–39 .…”

Performance of baseline FDG-PET/CT radiomics for prediction of bone marrow minimal residual disease status in the LyMa-101 trial

“…The intuition reported in the current work needs to be confirmed by an external dataset using the same therapy strategy. Finally, the small number of patients per center involved in this multicentric study prevented the use of harmonisation techniques such as ComBat-like approaches 15 . Therefore, it was not considered in this study.…”