18F-FDG PET/CT metabolic tumor parameters and radiomics features in aggressive non-Hodgkin’s lymphoma as predictors of treatment outcome and survival

Parvez, Aatif; Tau, Noam; Hussey, Doug; Maganti, Manjula; Metser, Ur

doi:10.1007/s12149-018-1260-1

“…Results of the classifier using the largest lesion were not satisfactory (accuracy = 60%), but the small sample size prevent any speculation about their reliability. Conversely, the RUBTE provided promising results when all lesions were used for the analysis, similarly to the previous investigations [14,15,32,33]. Unlike, in the study by Milgrom et al the authors found the mediastinal lesion-derived features could predict patient outcome, while features extracted from all lymphoma sites did not predict refractory disease [13].…”

Section: Discussionsupporting

confidence: 76%

Methodological framework for radiomics applications in Hodgkin Lymphoma

Sollini

¹

,

Kirienko

²

,

Cavinato

³

et al. 2020

Preprint

0

View full text Add to dashboard Cite

Background According to published data radiomics features differ between lesions of refractory/relapsing HL patients from those of long-term responders. However, several methodological aspects have not been elucidated yet.Purpose The study aimed at setting up a methodological framework in radiomics applications in Hodgkin Lymphoma (HL), especially at: a) developing a novel feature selection approach; b) evaluating radiomic intra-patient lesions’ similarity; c) classifying relapsing refractory (R/R) vs non-(R/R) patients.Methods We retrospectively included 85 patients (male:female=52:33; median age 35 years, range 19-74). LIFEx (www.lifexsoft.org) was used for [18F]FDG-PET/CT segmentation and feature extraction. Features were a-priori selected if they were highly correlated or uncorrelated to the volume. Principal component analysis-transformed features were used to build the fingerprints, that were tested to assess lesions’ similarity, using the silhouette. For intra-patient similarity analysis, we used patients having multiple lesions only. To classify patients as non-R/R and R/R, the fingerprint considering one single lesion (fingerprint_One) and all lesions (fingerprint_All) was tested using Random Undersamplig Boosting of Tree Ensemble (RUBTE). Results HL fingerprints included up to 15 features. Intra-patient lesion similarity analysis resulted in mean/median silhouette values below 0.5 (low similarity especially in the non-R/R group). In the test set, the fingerprint_One classification accuracy was 62% (78% sensitivity and 53% specificity); the classification by RUBTE using fingerprint_All resulted in 82% accuracy (70% sensitivity and 88% specificity). Conclusions Lesion similarity analysis was developed, and it allowed to demonstrate that HL lesions were not homogeneous within patients in terms of radiomics signature. Therefore, a random target lesion selection should not be adopted for radiomics applications. Moreover, the classifier to predict R/R vs non-R/R, performed the best when all the lesions were used.

show abstract

“…Results of the classifier using the largest lesion were not satisfactory (accuracy=60%), but the small sample size prevent any speculation about their reliability. Conversely, the RUBTE provided promising results when all lesions were used for the analysis, similarly to the previous investigations [10,11,30,31]. Unlike, in the study by Milgrom et al the authors found the mediastinal lesion-derived features could predict patient outcome, while features extracted from all lymphoma sites did not predict refractory disease [9].…”

Section: Discussionsupporting

confidence: 72%

Methodological framework for radiomics applications in Hodgkin Lymphoma

Sollini

¹

,

Kirienko

²

,

Cavinato

³

et al. 2020

Preprint

0

View full text Add to dashboard Cite

Background According to published data radiomics features differ between lesions of refractory/relapsing HL patients from those of long-term responders. However, several methodological aspects have not been elucidated yet.Purpose The study aimed at setting up a methodological framework in radiomics applications in Hodgkin Lymphoma (HL), especially at: a) developing a novel feature selection approach; b) evaluating radiomic intra-patient lesions’ similarity; c) classifying relapsing refractory (R/R) vs non-(R/R) patients.Methods We retrospectively included 85 patients (male:female=52:33; median age 35 years, range 19-74). LIFEx (www.lifexsoft.org) was used for [18F]FDG-PET/CT segmentation and feature extraction. Features were a-priori selected if they were highly correlated or uncorrelated to the volume. Principal component analysis-transformed features were used to build the fingerprints, that were tested to assess lesions’ similarity, using the silhouette. For intra-patient similarity analysis, we used patients having multiple lesions only. To classify patients as non-R/R and R/R, the fingerprint considering one single lesion (fingerprint_One) and all lesions (fingerprint_All) was tested using Random Undersamplig Boosting of Tree Ensemble (RUBTE). Results HL fingerprints included up to 15 features. Intra-patient lesion similarity analysis resulted in mean/median silhouette values below 0.5 (low similarity especially in the non-R/R group). In the test set, the fingerprint_One classification accuracy was 62% (78% sensitivity and 53% specificity); the classification by RUBTE using fingerprint_All resulted in 82% accuracy (70% sensitivity and 88% specificity). Conclusions Lesion similarity analysis was developed, and it allowed to demonstrate that HL lesions were not homogeneous within patients in terms of radiomics signature. Therefore, a random target lesion selection should not be adopted for radiomics applications. Moreover, the classifier to predict R/R vs non-R/R, performed the best when all the lesions were used.

show abstract

“…Most studies acknowledge the difficulties in the assessment of multiple lymphoma lesions. Some used boxes or VOIs to constrain individual tumors [ 6 , 8 , 12 ], or limited segmentation to a representative maximum of 5 lymphoma lesions [ 26 ], but none of these studies compared such strategies with another workflow.…”

Section: Discussionmentioning

confidence: 99%

Optimizing Workflows for Fast and Reliable Metabolic Tumor Volume Measurements in Diffuse Large B Cell Lymphoma

Burggraaff

¹

,

Rahman

²

,

Kaßner

³

et al. 2020

View full text Add to dashboard Cite

Purpose: This pilot study aimed to determine interobserver reliability and ease of use of three workflows for measuring metabolic tumor volume (MTV) and total lesion glycolysis (TLG) in diffuse large B cell lymphoma (DLBCL). Procedures: Twelve baseline [ 18 F]FDG PET/CT scans from DLBCL patients with wide variation in number and size of involved organs and lymph nodes were selected from the international PETRA consortium database. Three observers analyzed scans using three workflows. Workflow A: user-defined selection of individual lesions followed by four automated segmentations (41%SUVmax, A50%SUVpeak, SUV≥2.5, SUV≥4.0). For each lesion, observers indicated their "preferred segmentation." Individually selected lesions were summed to yield total MTV and TLG. Workflow B: fully automated preselection of [ 18 F]FDG-avid structures (SUV≥4.0 and volume≥3ml), followed by removing non-tumor regions with single mouse clicks. Workflow C: preselected volumes based on Workflow B modified by manually adding lesions or removing physiological uptake, subsequently checked by experienced nuclear medicine physicians. Workflow C was performed 3 months later to avoid recall bias from the initial Workflow B analysis. Interobserver reliability was expressed as intraclass correlation coefficients (ICC).

show abstract

18F-FDG PET/CT metabolic tumor parameters and radiomics features in aggressive non-Hodgkin’s lymphoma as predictors of treatment outcome and survival

Cited by 73 publications

References 15 publications

Methodological framework for radiomics applications in Hodgkin Lymphoma

Methodological framework for radiomics applications in Hodgkin Lymphoma

Methodological framework for radiomics applications in Hodgkin Lymphoma

Optimizing Workflows for Fast and Reliable Metabolic Tumor Volume Measurements in Diffuse Large B Cell Lymphoma

Contact Info

Product

Resources

About