Methodological framework for radiomics applications in Hodgkin’s lymphoma
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..
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2020 |
---|---|
Erschienen: |
2020 |
Enthalten in: |
Zur Gesamtaufnahme - volume:4 |
---|---|
Enthalten in: |
European journal of hybrid imaging - 4(2020), 1 vom: 01. Juni |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Sollini, Martina [VerfasserIn] |
---|
Links: |
Volltext [kostenfrei] |
---|
Themen: |
Feature selection |
---|
doi: |
10.1186/s41824-020-00078-8 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
SPR039893812 |
---|
LEADER | 01000caa a22002652 4500 | ||
---|---|---|---|
001 | SPR039893812 | ||
003 | DE-627 | ||
005 | 20230519221152.0 | ||
007 | cr uuu---uuuuu | ||
008 | 201007s2020 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1186/s41824-020-00078-8 |2 doi | |
035 | |a (DE-627)SPR039893812 | ||
035 | |a (SPR)s41824-020-00078-8-e | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
082 | 0 | 4 | |a 610 |q ASE |
082 | 0 | 4 | |a 610 |q ASE |
100 | 1 | |a Sollini, Martina |e verfasserin |4 aut | |
245 | 1 | 0 | |a Methodological framework for radiomics applications in Hodgkin’s lymphoma |
264 | 1 | |c 2020 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a Computermedien |b c |2 rdamedia | ||
338 | |a Online-Ressource |b cr |2 rdacarrier | ||
520 | |a 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. | ||
650 | 4 | |a Lymphoma |7 (dpeaa)DE-He213 | |
650 | 4 | |a PET/CT |7 (dpeaa)DE-He213 | |
650 | 4 | |a Radiomics |7 (dpeaa)DE-He213 | |
650 | 4 | |a Similarity |7 (dpeaa)DE-He213 | |
650 | 4 | |a Feature selection |7 (dpeaa)DE-He213 | |
650 | 4 | |a Silhouette |7 (dpeaa)DE-He213 | |
650 | 4 | |a Response prediction |7 (dpeaa)DE-He213 | |
650 | 4 | |a Outcome prediction |7 (dpeaa)DE-He213 | |
700 | 1 | |a Kirienko, Margarita |e verfasserin |4 aut | |
700 | 1 | |a Cavinato, Lara |e verfasserin |4 aut | |
700 | 1 | |a Ricci, Francesca |e verfasserin |4 aut | |
700 | 1 | |a Biroli, Matteo |e verfasserin |4 aut | |
700 | 1 | |a Ieva, Francesca |e verfasserin |4 aut | |
700 | 1 | |a Calderoni, Letizia |e verfasserin |4 aut | |
700 | 1 | |a Tabacchi, Elena |e verfasserin |4 aut | |
700 | 1 | |a Nanni, Cristina |e verfasserin |4 aut | |
700 | 1 | |a Zinzani, Pier Luigi |e verfasserin |4 aut | |
700 | 1 | |a Fanti, Stefano |e verfasserin |4 aut | |
700 | 1 | |a Guidetti, Anna |e verfasserin |4 aut | |
700 | 1 | |a Alessi, Alessandra |e verfasserin |4 aut | |
700 | 1 | |a Corradini, Paolo |e verfasserin |4 aut | |
700 | 1 | |a Seregni, Ettore |e verfasserin |4 aut | |
700 | 1 | |a Carlo-Stella, Carmelo |e verfasserin |4 aut | |
700 | 1 | |a Chiti, Arturo |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t European journal of hybrid imaging |d [Cham] : Springer International Publishers, 2017 |g 4(2020), 1 vom: 01. Juni |w (DE-627)SPR038321564 |w (DE-600)2886512-1 |x 2510-3636 |7 nnns |
773 | 1 | 8 | |g volume:4 |g year:2020 |g number:1 |g day:01 |g month:06 |
856 | 4 | 0 | |u https://dx.doi.org/10.1186/s41824-020-00078-8 |z kostenfrei |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_SPRINGER | ||
912 | |a SSG-OLC-PHA | ||
951 | |a AR | ||
952 | |d 4 |j 2020 |e 1 |b 01 |c 06 |