Abstract
Objective To determine whether machine learning can accurately classify human papillomavirus (HPV) status of oropharyngeal squamous cell carcinoma (OPSCC) using computed tomography (CT)-based texture analysis. Methods Texture analyses were retrospectively applied to regions of interest from OPSCC primary tumors on contrast-enhanced neck CT, and machine learning was used to create a model that classified HPV status with the highest accuracy. Results were compared against the blinded review of 2 neuroradiologists. Results The HPV-positive (n = 92) and-negative (n = 15) cohorts were well matched clinically. Neuroradiologist classification accuracies for HPV status (44.9%, 55.1%) were not significantly different (P = 0.13), and there was a lack of agreement between the 2 neuroradiologists (κ =-0.145). The best machine learning model had an accuracy of 75.7%, which was greater than either neuroradiologist (P < 0.001, P = 0.002). Conclusions Useful diagnostic information regarding HPV infection can be extracted from the CT appearance of OPSCC beyond what is apparent to the trained human eye.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 299-305 |
| Number of pages | 7 |
| Journal | Journal of Computer Assisted Tomography |
| Volume | 42 |
| Issue number | 2 |
| DOIs | |
| State | Published - Mar 1 2018 |
Fingerprint
Keywords
- human papillomavirus
- machine learning
- oropharyngeal cancer
- oropharynx
- radiomics
- squamous cell carcinoma
- texture analysis
ASJC Scopus subject areas
- Radiology Nuclear Medicine and imaging
Cite this
Computed Tomography-Based Texture Analysis to Determine Human Papillomavirus Status of Oropharyngeal Squamous Cell Carcinoma. / Ranjbar, Sara; Ning, Shuluo; Zwart, Christine M.; Wood, Christopher P.; Weindling, Steven M.; Wu, Teresa; Ross Mitchell, J.; Li, Jing; Hoxworth, Joseph M.
In: Journal of Computer Assisted Tomography, Vol. 42, No. 2, 01.03.2018, p. 299-305.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Computed Tomography-Based Texture Analysis to Determine Human Papillomavirus Status of Oropharyngeal Squamous Cell Carcinoma
AU - Ranjbar, Sara
AU - Ning, Shuluo
AU - Zwart, Christine M.
AU - Wood, Christopher P.
AU - Weindling, Steven M.
AU - Wu, Teresa
AU - Ross Mitchell, J.
AU - Li, Jing
AU - Hoxworth, Joseph M.
PY - 2018/3/1
Y1 - 2018/3/1
N2 - Objective To determine whether machine learning can accurately classify human papillomavirus (HPV) status of oropharyngeal squamous cell carcinoma (OPSCC) using computed tomography (CT)-based texture analysis. Methods Texture analyses were retrospectively applied to regions of interest from OPSCC primary tumors on contrast-enhanced neck CT, and machine learning was used to create a model that classified HPV status with the highest accuracy. Results were compared against the blinded review of 2 neuroradiologists. Results The HPV-positive (n = 92) and-negative (n = 15) cohorts were well matched clinically. Neuroradiologist classification accuracies for HPV status (44.9%, 55.1%) were not significantly different (P = 0.13), and there was a lack of agreement between the 2 neuroradiologists (κ =-0.145). The best machine learning model had an accuracy of 75.7%, which was greater than either neuroradiologist (P < 0.001, P = 0.002). Conclusions Useful diagnostic information regarding HPV infection can be extracted from the CT appearance of OPSCC beyond what is apparent to the trained human eye.
AB - Objective To determine whether machine learning can accurately classify human papillomavirus (HPV) status of oropharyngeal squamous cell carcinoma (OPSCC) using computed tomography (CT)-based texture analysis. Methods Texture analyses were retrospectively applied to regions of interest from OPSCC primary tumors on contrast-enhanced neck CT, and machine learning was used to create a model that classified HPV status with the highest accuracy. Results were compared against the blinded review of 2 neuroradiologists. Results The HPV-positive (n = 92) and-negative (n = 15) cohorts were well matched clinically. Neuroradiologist classification accuracies for HPV status (44.9%, 55.1%) were not significantly different (P = 0.13), and there was a lack of agreement between the 2 neuroradiologists (κ =-0.145). The best machine learning model had an accuracy of 75.7%, which was greater than either neuroradiologist (P < 0.001, P = 0.002). Conclusions Useful diagnostic information regarding HPV infection can be extracted from the CT appearance of OPSCC beyond what is apparent to the trained human eye.
KW - human papillomavirus
KW - machine learning
KW - oropharyngeal cancer
KW - oropharynx
KW - radiomics
KW - squamous cell carcinoma
KW - texture analysis
UR - http://www.scopus.com/inward/record.url?scp=85044400483&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85044400483&partnerID=8YFLogxK
U2 - 10.1097/RCT.0000000000000682
DO - 10.1097/RCT.0000000000000682
M3 - Article
C2 - 29189396
AN - SCOPUS:85044400483
VL - 42
SP - 299
EP - 305
JO - Journal of Computer Assisted Tomography
JF - Journal of Computer Assisted Tomography
SN - 0363-8715
IS - 2
ER -