Refining Tumor Treatment in Sinonasal Cancer Using Delta Radiomics of Multi-Parametric MRI after the First Cycle of Induction Chemotherapy.

delta radiomics machine learning magnetic resonance imaging radiomics sinonasal cancer

Journal

Journal of imaging
ISSN: 2313-433X
Titre abrégé: J Imaging
Pays: Switzerland
ID NLM: 101698819

Informations de publication

Date de publication:
15 Feb 2022
Historique:
received: 16 12 2021
revised: 23 01 2022
accepted: 11 02 2022
entrez: 24 2 2022
pubmed: 25 2 2022
medline: 25 2 2022
Statut: epublish

Résumé

Response to induction chemotherapy (IC) has been predicted in patients with sinonasal cancer using early delta radiomics obtained from T1- and T2-weighted images and apparent diffusion coefficient (ADC) maps, comparing results with early radiological evaluation by RECIST. Fifty patients were included in the study. For each image (at baseline and after the first IC cycle), 536 radiomic features were extracted as follows: semi-supervised principal component analysis components, explaining 97% of the variance, were used together with a support vector machine (SVM) to develop a radiomic signature. One signature was developed for each sequence (T1-, T2-weighted and ADC). A multiagent decision-making algorithm was used to merge multiple signatures into one score. The area under the curve (AUC) for mono-modality signatures was 0.79 (CI: 0.65-0.88), 0.76 (CI: 0.62-0.87) and 0.93 (CI: 0.75-1) using T1-, T2-weighted and ADC images, respectively. The fuse signature improved the AUC when an ADC-based signature was added. Radiological prediction using RECIST criteria reached an accuracy of 0.78. These results suggest the importance of early delta radiomics and of ADC maps to predict the response to IC in sinonasal cancers.

Sections du résumé

BACKGROUND BACKGROUND
Response to induction chemotherapy (IC) has been predicted in patients with sinonasal cancer using early delta radiomics obtained from T1- and T2-weighted images and apparent diffusion coefficient (ADC) maps, comparing results with early radiological evaluation by RECIST.
METHODS METHODS
Fifty patients were included in the study. For each image (at baseline and after the first IC cycle), 536 radiomic features were extracted as follows: semi-supervised principal component analysis components, explaining 97% of the variance, were used together with a support vector machine (SVM) to develop a radiomic signature. One signature was developed for each sequence (T1-, T2-weighted and ADC). A multiagent decision-making algorithm was used to merge multiple signatures into one score.
RESULTS RESULTS
The area under the curve (AUC) for mono-modality signatures was 0.79 (CI: 0.65-0.88), 0.76 (CI: 0.62-0.87) and 0.93 (CI: 0.75-1) using T1-, T2-weighted and ADC images, respectively. The fuse signature improved the AUC when an ADC-based signature was added. Radiological prediction using RECIST criteria reached an accuracy of 0.78.
CONCLUSIONS CONCLUSIONS
These results suggest the importance of early delta radiomics and of ADC maps to predict the response to IC in sinonasal cancers.

Identifiants

pubmed: 35200748
pii: jimaging8020046
doi: 10.3390/jimaging8020046
pmc: PMC8877083
pii:
doi:

Types de publication

Journal Article

Langues

eng

Subventions

Organisme : Italian Association for Cancer Research
ID : 1732

Références

Magn Reson Med. 2016 Apr;75(4):1708-16
pubmed: 25995019
AJR Am J Roentgenol. 2007 Jun;188(6):1622-35
pubmed: 17515386
JAMA Otolaryngol Head Neck Surg. 2017 Apr 1;143(4):355-366
pubmed: 28152117
J Clin Oncol. 2006 Feb 1;24(4):593-8
pubmed: 16380415
Korean J Intern Med. 2016 May;31(3):570-8
pubmed: 26976150
Arch Otolaryngol Head Neck Surg. 2011 Jan;137(1):78-81
pubmed: 21242552
Med Phys. 2019 Nov;46(11):5116-5123
pubmed: 31539450
Ann Oncol. 2003 Mar;14(3):367-72
pubmed: 12598339
J Magn Reson Imaging. 2014 Dec;40(6):1375-81
pubmed: 24222019
Eur J Cancer. 2009 Jan;45(2):228-47
pubmed: 19097774
IEEE Access. 2018;6:77796-77806
pubmed: 30607311
Radiology. 2020 May;295(2):328-338
pubmed: 32154773
J Magn Reson Imaging. 2016 Jan;43(1):213-9
pubmed: 26013043
Magn Reson Imaging. 2012 Nov;30(9):1234-48
pubmed: 22898692
Laryngoscope. 2015 Nov;125(11):2491-7
pubmed: 26228792
NMR Biomed. 2020 Feb 3;:e4265
pubmed: 32009265
J Magn Reson Imaging. 2018 Mar;47(3):829-840
pubmed: 28653477
Phys Med Biol. 2020 Aug 27;65(17):175006
pubmed: 32554891
Radiother Oncol. 2014 Mar;110(3):429-34
pubmed: 24630535
Clin Radiol. 2017 Jan;72(1):74-83
pubmed: 27789026
Int Forum Allergy Rhinol. 2018 Sep;8(9):1065-1072
pubmed: 29935059
Radiat Oncol. 2019 Mar 12;14(1):43
pubmed: 30866965
J Magn Reson Imaging. 2019 Aug;50(2):497-510
pubmed: 30569552
Cancer Imaging. 2020 Jan 14;20(1):7
pubmed: 31937372

Auteurs

Valentina D A Corino (VDA)

Department of Electronics, Information and Bioengineering, Politecnico di Milano, 20133 Milan, Italy.

Marco Bologna (M)

Department of Electronics, Information and Bioengineering, Politecnico di Milano, 20133 Milan, Italy.

Giuseppina Calareso (G)

Radiology Department, Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, 20133 Milan, Italy.

Carlo Resteghini (C)

Head and Neck Medical Oncology Department, Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, 20133 Milan, Italy.

Silvana Sdao (S)

Radiology Department, Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, 20133 Milan, Italy.

Ester Orlandi (E)

Head and Neck Medical Oncology Department, Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, 20133 Milan, Italy.

Lisa Licitra (L)

Head and Neck Medical Oncology Department, Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, 20133 Milan, Italy.
Department of Oncology and Hemato-Oncology, University of Milan, 20122 Milan, Italy.

Luca Mainardi (L)

Department of Electronics, Information and Bioengineering, Politecnico di Milano, 20133 Milan, Italy.

Paolo Bossi (P)

Department of Medical and Surgical Specialties, Radiological Sciences and Public Health, University of Brescia, 25123 Brescia, Italy.

Classifications MeSH