S Ramkumar1, S Ranjbar2, S Ning1, D Lal3, C M Zwart4, C P Wood5, S M Weindling6, T Wu1, J R Mitchell7, J Li1, J M Hoxworth8. 1. From the School of Computing, Informatics, and Decision Systems Engineering (S.Ramkumar, S.N., T.W., J.L.). 2. Department of Biomedical Informatics (S.Ranjbar), Arizona State University, Tempe, Arizona. 3. Departments of Otolaryngology (D.L.). 4. Radiology (C.M.Z., J.M.H.), Mayo Clinic, Phoenix, Arizona. 5. Department of Radiology (C.P.W.), Mayo Clinic, Rochester, Minnesota. 6. Department of Radiology (S.M.W.), Mayo Clinic, Jacksonville, Florida. 7. Department of Research (J.R.M.), Mayo Clinic, Scottsdale, Arizona. 8. Radiology (C.M.Z., J.M.H.), Mayo Clinic, Phoenix, Arizona hoxworth.joseph@mayo.edu.
Abstract
BACKGROUND AND PURPOSE: Because sinonasal inverted papilloma can harbor squamous cell carcinoma, differentiating these tumors is relevant. The objectives of this study were to determine whether MR imaging-based texture analysis can accurately classify cases of noncoexistent squamous cell carcinoma and inverted papilloma and to compare this classification performance with neuroradiologists' review. MATERIALS AND METHODS: Adult patients who had inverted papilloma or squamous cell carcinoma resected were eligible (coexistent inverted papilloma and squamous cell carcinoma were excluded). Inclusion required tumor size of >1.5 cm and preoperative MR imaging with axial T1, axial T2, and axial T1 postcontrast sequences. Five well-established texture analysis algorithms were applied to an ROI from the largest tumor cross-section. For a training dataset, machine-learning algorithms were used to identify the most accurate model, and performance was also evaluated in a validation dataset. On the basis of 3 separate blinded reviews of the ROI, isolated tumor, and entire images, 2 neuroradiologists predicted tumor type in consensus. RESULTS: The inverted papilloma (n = 24) and squamous cell carcinoma (n = 22) cohorts were matched for age and sex, while squamous cell carcinoma tumor volume was larger (P = .001). The best classification model achieved similar accuracies for training (17 squamous cell carcinomas, 16 inverted papillomas) and validation (7 squamous cell carcinomas, 6 inverted papillomas) datasets of 90.9% and 84.6%, respectively (P = .537). For the combined training and validation cohorts, the machine-learning accuracy (89.1%) was better than that of the neuroradiologists' ROI review (56.5%, P = .0004) but not significantly different from the neuroradiologists' review of the tumors (73.9%, P = .060) or entire images (87.0%, P = .748). CONCLUSIONS: MR imaging-based texture analysis has the potential to differentiate squamous cell carcinoma from inverted papilloma and may, in the future, provide incremental information to the neuroradiologist.
BACKGROUND AND PURPOSE: Because sinonasal inverted papilloma can harbor squamous cell carcinoma, differentiating these tumors is relevant. The objectives of this study were to determine whether MR imaging-based texture analysis can accurately classify cases of noncoexistent squamous cell carcinoma and inverted papilloma and to compare this classification performance with neuroradiologists' review. MATERIALS AND METHODS: Adult patients who had inverted papilloma or squamous cell carcinoma resected were eligible (coexistent inverted papilloma and squamous cell carcinoma were excluded). Inclusion required tumor size of >1.5 cm and preoperative MR imaging with axial T1, axial T2, and axial T1 postcontrast sequences. Five well-established texture analysis algorithms were applied to an ROI from the largest tumor cross-section. For a training dataset, machine-learning algorithms were used to identify the most accurate model, and performance was also evaluated in a validation dataset. On the basis of 3 separate blinded reviews of the ROI, isolated tumor, and entire images, 2 neuroradiologists predicted tumor type in consensus. RESULTS: The inverted papilloma (n = 24) and squamous cell carcinoma (n = 22) cohorts were matched for age and sex, while squamous cell carcinoma tumor volume was larger (P = .001). The best classification model achieved similar accuracies for training (17 squamous cell carcinomas, 16 inverted papillomas) and validation (7 squamous cell carcinomas, 6 inverted papillomas) datasets of 90.9% and 84.6%, respectively (P = .537). For the combined training and validation cohorts, the machine-learning accuracy (89.1%) was better than that of the neuroradiologists' ROI review (56.5%, P = .0004) but not significantly different from the neuroradiologists' review of the tumors (73.9%, P = .060) or entire images (87.0%, P = .748). CONCLUSIONS: MR imaging-based texture analysis has the potential to differentiate squamous cell carcinoma from inverted papilloma and may, in the future, provide incremental information to the neuroradiologist.
Authors: Erez Eyal; B Nicolas Bloch; Neil M Rofsky; Edna Furman-Haran; Elizabeth M Genega; Robert E Lenkinski; Hadassa Degani Journal: Invest Radiol Date: 2010-04 Impact factor: 6.016
Authors: Haowei Zhang; Caleb M Graham; Okan Elci; Michael E Griswold; Xu Zhang; Majid A Khan; Karen Pitman; Jimmy J Caudell; Robert D Hamilton; Balaji Ganeshan; Andrew Dennis Smith Journal: Radiology Date: 2013-10-28 Impact factor: 11.105
Authors: Evangelia I Zacharaki; Sumei Wang; Sanjeev Chawla; Dong Soo Yoo; Ronald Wolf; Elias R Melhem; Christos Davatzikos Journal: Magn Reson Med Date: 2009-12 Impact factor: 4.668
Authors: Sylvia Drabycz; Gloria Roldán; Paula de Robles; Daniel Adler; John B McIntyre; Anthony M Magliocco; J Gregory Cairncross; J Ross Mitchell Journal: Neuroimage Date: 2009-09-28 Impact factor: 6.556
Authors: Tae Yeon Jeon; Hyung-Jin Kim; Joon Young Choi; In Ho Lee; Sung Tae Kim; Pyoung Jeon; Keon Ha Kim; Hong Sik Byun Journal: Neuroradiology Date: 2009-03-04 Impact factor: 2.804
Authors: Anna M Brown; Sidhartha Nagala; Mary A McLean; Yonggang Lu; Daniel Scoffings; Aditya Apte; Mithat Gonen; Hilda E Stambuk; Ashok R Shaha; R Michael Tuttle; Joseph O Deasy; Andrew N Priest; Piyush Jani; Amita Shukla-Dave; John Griffiths Journal: Magn Reson Med Date: 2015-05-20 Impact factor: 4.668
Authors: Amanda Drumstas Nussi; Sérgio Lucio Pereira de Castro Lopes; Catharina Simioni De Rosa; João Pedro Perez Gomes; Celso Massahiro Ogawa; Paulo Henrique Braz-Silva; Andre Luiz Ferreira Costa Journal: Oral Radiol Date: 2022-05-18 Impact factor: 1.852
Authors: Antonio Mario Bulfamante; Francesco Ferella; Austin Michael Miller; Cecilia Rosso; Carlotta Pipolo; Emanuela Fuccillo; Giovanni Felisati; Alberto Maria Saibene Journal: Eur Arch Otorhinolaryngol Date: 2022-10-19 Impact factor: 3.236
Authors: Sara Ranjbar; Stefanie N Velgos; Amylou C Dueck; Yonas E Geda; J Ross Mitchell Journal: J Neuropsychiatry Clin Neurosci Date: 2019-01-14 Impact factor: 2.198
Authors: Amit Jethanandani; Timothy A Lin; Stefania Volpe; Hesham Elhalawani; Abdallah S R Mohamed; Pei Yang; Clifton D Fuller Journal: Front Oncol Date: 2018-05-14 Impact factor: 6.244