Predictive value of initial FDG-PET features for treatment response and survival in esophageal cancer patients treated with chemo-radiation therapy using a random forest classifier

Abstract

PurposeIn oncology, texture features extracted from positron emission tomography with 18-fluorodeoxyglucose images (FDG-PET) are of increasing interest for predictive and prognostic studies, leading to several tens of features per tumor. To select the best features, the use of a random forest (RF) classifier was investigated.MethodsSixty-five patients with an esophageal cancer treated with a combined chemo-radiation therapy were retrospectively included. All patients underwent a pretreatment whole-body FDG-PET. The patients were followed for 3 years after the end of the treatment. The response assessment was performed 1 month after the end of the therapy. Patients were classified as complete responders and non-complete responders. Sixty-one features were extracted from medical records and PET images. First, Spearman’s analysis was performed to eliminate correlated features. Then, the best predictive and prognostic subsets of features were selected using a RF algorithm. These results were compared to those obtained by a Mann-Whitney U test (predictive study) and a univariate Kaplan-Meier analysis (prognostic study).ResultsAmong the 61 initial features, 28 were not correlated. From these 28 features, the best subset of complementary features found using the RF classifier to predict response was composed of 2 features: metabolic tumor volume (MTV) and homogeneity from the co-occurrence matrix. The corresponding predictive value (AUC = 0.836 ± 0.105, Se = 82 ± 9%, Sp = 91 ± 12%) was higher than the best predictive results found using the Mann-Whitney test: busyness from the gray level difference matrix (P < 0.0001, AUC = 0.810, Se = 66%, Sp = 88%). The best prognostic subset found using RF was composed of 3 features: MTV and 2 clinical features (WHO status and nutritional risk index) (AUC = 0.822 ± 0.059, Se = 79 ± 9%, Sp = 95 ± 6%), while no feature was significantly prognostic according to the Kaplan-Meier analysis.ConclusionsThe RF classifier can improve predictive and prognostic values compared to the Mann-Whitney U test and the univariate Kaplan-Meier survival analysis when applied to several tens of features in a limited patient database.

Prediction of Response to Neoadjuvant Chemotherapy and Radiation Therapy with Baseline and Restaging 18F-FDG PET Imaging Biomarkers in Patients with Esophageal Cancer

“…To validate whether these radiomic features also provide complementary clinical value in external cohorts, pooling those results is required. In a systematic Medline database search, we found 10 relevant studies that investigated the value of PET/ CT radiomic features in the prediction of therapy response in esophageal cancer (11)(12)(13)(14)(15)(16)(17)(18)25,26). The study by Van Rossum et al (11) was, to our knowledge, the only study that performed quantitative 18 F-FDG PET from pre-NCRT and post-NCRT 18 F-FDG PET, classified response dissimilarity, and zone percentage; however, they assessed a limited number of radiomic features.…”

Improving Sea Ice Characterization in Dry Ice Winter Conditions Using Polarimetric Parameters from C- and L-Band SAR Data

“…The 0.7 and 0.9 are reasonable threshold values to define the three aforementioned categories and were successfully used in different image processing and feature selection studies, e.g., [34,35]. Thus, using the calculated K-S distances, polarimetric parameters with discrimination capability between FYI and OI are identified and groups based on the obtained K-S distance values are created.…”

Pre-Launch Assessment of RADARSAT Constellation Mission Medium Resolution Modes for Sea Oil Slicks and Lookalike Discrimination

“…We assume a CP feature has very good separability between 2 classes if the K-S distance !0.9. The selected 0.7 and 0.9 values are reasonable threshold values to define the 3 aforementioned groups and were successfully used in different image processing and feature selection studies (Desbordes et al 2017;Dabboor et al 2017Dabboor et al , 2018b. In each RCM mode, the analysis of the separability using the K-S distance leads to the identification of CP features with potential separability between LA and mineral oil slicks and the exclusion of those CP features with no separation capability.…”