Martina Sollini1,2, Margarita Kirienko3, Lara Cavinato2,4, Francesca Ricci2, Matteo Biroli1, Francesca Ieva4,5, Letizia Calderoni6, Elena Tabacchi6, Cristina Nanni6, Pier Luigi Zinzani7, Stefano Fanti6, Anna Guidetti8,9, Alessandra Alessi8, Paolo Corradini8,9, Ettore Seregni8, Carmelo Carlo-Stella1,2, Arturo Chiti1,2. 1. Humanitas University, Via Rita Levi Montalcini 4, MI 20090, Pieve Emanuele, Italy. 2. Humanitas Clinical and Research Center - IRCCS -, via Manzoni 56, 20089, Rozzano, MI, Italy. 3. Humanitas University, Via Rita Levi Montalcini 4, MI 20090, Pieve Emanuele, Italy. margarita.kirienko@icloud.com. 4. MOX-Modelling and Scientific Computing lab., Department of Mathematics, Politecnico di Milano, Milano, Italy. 5. CADS-Center for Analysis, Decision, and Society, Human Technopole, Milano, Italy. 6. Nuclear Medicine, AOU S.Orsola-Malpighi, Bologna, Italy. 7. Institute of Hematology "Seràgnoli", University of Bologna, Bologna, Italy. 8. Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy. 9. University of Milan, Milan, Italy.
Abstract
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's lymphoma (HL), especially at (a) developing a novel feature selection approach, (b) evaluating radiomic intra-patient lesions' similarity, and (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 Undersampling 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.
BACKGROUND: According to published data, radiomics features differ between lesions of refractory/relapsing HLpatients 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's lymphoma (HL), especially at (a) developing a novel feature selection approach, (b) evaluating radiomic intra-patient lesions' similarity, and (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 Undersampling 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.
Authors: M Carles; I Torres-Espallardo; A Alberich-Bayarri; C Olivas; P Bello; U Nestle; L Martí-Bonmatí Journal: Phys Med Biol Date: 2016-12-29 Impact factor: 3.609
Authors: B Ganeshan; K A Miles; S Babikir; R Shortman; A Afaq; K M Ardeshna; A M Groves; I Kayani Journal: Eur Radiol Date: 2016-07-05 Impact factor: 5.315