Abstract

PurposeTo test the performances of native and tumour to liver ratio (TLR) radiomic features extracted from pre-treatment 2-[18F] fluoro-2-deoxy-D-glucose ([18F]FDG) PET/CT and combined with machine learning (ML) for predicting cancer recurrence in patients with locally advanced cervical cancer (LACC).MethodsOne hundred fifty-eight patients with LACC from multiple centers were retrospectively included in the study. Tumours were segmented using the Fuzzy Local Adaptive Bayesian (FLAB) algorithm. Radiomic features were extracted from the tumours and from regions drawn over the normal liver. Cox proportional hazard model was used to test statistical significance of clinical and radiomic features. Fivefold cross validation was used to tune the number of features. Seven different feature selection methods and four classifiers were tested. The models with the selected features were trained using bootstrapping and tested in data from each scanner independently. Reproducibility of radiomics features, clinical data added value and effect of ComBat-based harmonisation were evaluated across scanners.ResultsAfter a median follow-up of 23 months, 29% of the patients recurred. No individual radiomic or clinical features were significantly associated with cancer recurrence. The best model was obtained using 10 TLR features combined with clinical information. The area under the curve (AUC), F1-score, precision and recall were respectively 0.78 (0.67–0.88), 0.49 (0.25–0.67), 0.42 (0.25–0.60) and 0.63 (0.20–0.80). ComBat did not improve the predictive performance of the best models. Both the TLR and the native models performance varied across scanners used in the test set.Conclusion[18F]FDG PET radiomic features combined with ML add relevant information to the standard clinical parameters in terms of LACC patient’s outcome but remain subject to variability across PET/CT devices.

Highlights

  • Cervical cancer is the fourth most common cancer in women [1]

  • 2-[18F]fluoro-2-deoxy-D-glucose ([18F]FDG) positron emission tomography combined with computed tomography (PET/CT) imaging plays an important role in treatment stratification in oncology

  • We evaluated the predictive value of tumour to liver ratios (TLR) of radiomic features [27]

Read more

Summary

Introduction

Cervical cancer is the fourth most common cancer in women [1]. Currently, in clinical routine, the disease prognosis is based upon the FIGO/TNM staging system, with a particular emphasis on the lymph node involvement [2, 3]. Radiomic features have shown to predict treatment outcome in several cancer diseases including cervical cancer, and using various imaging modalities [8,9,10,11]. Most of radiomic features show high sensitivity to multiple factors, including the scanner manufacturer and specific properties, acquisition protocols and the reconstruction algorithm and settings of each clinical center [12,13,14,15,16,17,18]. Radiomics have increasingly been combined with machine learning (ML) techniques in order to predict a specific clinical outcome [8, 10, 11, 19,20,21,22,23,24,25,26]

Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call