SMART LAB

Publication

Publication

Differentiation of recurrent glioblastoma from radiation necrosis using diffusion radiomics with machine learning model development and external validation
Journal
Scientific reports
Vol
11
Page
2913
Author
Yae Won Park , Dongmin Choi , Ji Eun Park , Sung SooAhn, Hwiyoung Kim , Jong Hee Chang , Se Hoon Kim , Ho Sung Kim, Seung‑Koo Lee
Year
Data-driven Clinical outcome Prediction
Date
Feb. 2021
File
HW Kim_2021_Scientific reports.pdf (1.3M) 3회 다운로드 DATE : 2021-03-14 09:09:05
File
s41598-021-82467-y.pdf (1.3M) 2회 다운로드 DATE : 2021-03-15 16:29:21
The purpose of this study was to establish a high-performing radiomics strategy with machine learning
from conventional and difusion MRI to diferentiate recurrent glioblastoma (GBM) from radiation
necrosis (RN) after concurrent chemoradiotherapy (CCRT) or radiotherapy. Eighty-six patients with
GBM were enrolled in the training set after they underwent CCRT or radiotherapy and presented
with new or enlarging contrast enhancement within the radiation feld on follow-up MRI. A diagnosis
was established either pathologically or clinicoradiologically (63 recurrent GBM and 23 RN). Another
41 patients (23 recurrent GBM and 18 RN) from a diferent institution were enrolled in the test set.
Conventional MRI sequences (T2-weighted and postcontrast T1-weighted images) and ADC were
analyzed to extract 263 radiomic features. After feature selection, various machine learning models
with oversampling methods were trained with combinations of MRI sequences and subsequently
validated in the test set. In the independent test set, the model using ADC sequence showed the best
diagnostic performance, with an AUC, accuracy, sensitivity, specifcity of 0.80, 78%, 66.7%, and 87%,
respectively. In conclusion, the radiomics models models using other MRI sequences showed AUCs
ranging from 0.65 to 0.66 in the test set. The difusion radiomics may be helpful in diferentiating
recurrent GBM from RN.