Abstract
Radiogenomics is a specific application of radiomics where imaging features are linked to genomic profiles. We aim to develop a radiogenomics model based on ovarian US images for predicting germline BRCA1/2 gene status in women with healthy ovaries. From January 2013 to December 2017 a total of 255 patients addressed to germline BRCA1/2 testing and pelvic US documenting normal ovaries, were retrospectively included. Feature selection for univariate analysis was carried out via correlation analysis. Multivariable analysis for classification of germline BRCA1/2 status was then carried out via logistic regression, support vector machine, ensemble of decision trees and automated machine learning pipelines. Data were split into a training (75%) and a testing (25%) set. The four strategies obtained a similar performance in terms of accuracy on the testing set (from 0.54 of logistic regression to 0.64 of the auto-machine learning pipeline). Data coming from one of the tested US machine showed generally higher performances, particularly with the auto-machine learning pipeline (testing set specificity 0.87, negative predictive value 0.73, accuracy value 0.72 and 0.79 on training set). The study shows that a radiogenomics model on machine learning techniques is feasible and potentially useful for predicting gBRCA1/2 status in women with healthy ovaries.
Highlights
Radiogenomics is a specific application of radiomics where imaging features are linked to genomic profiles
Around 10–20% of ovarian cancer and 6% breast cancer overall are caused by inheritable BRCA1/BRCA2 P Vs18
Considering the Voluson data only, performances were generally higher: in particular, testing set specificity and negative predictive value reached 0.77 and 0.81, respectively, with the radial kernel support vector machine of strategy B; 0.82 and 0.70 with the extreme gradient boosting of strategy C, and 0.87 and 0.73 with the automachine learning pipeline of strategy D. The latter combination, strategy D on Voluson data, showed the best consistency between training and testing set accuracy (0.79 and 0.72 respectively) suggesting that the adopted strategy performance was robust to avoid data overfitting, which might improve the classification performances on larger image numbers to obtain both a higher specificity and a higher negative predictive value. In this single center study, we developed an automated machine learning pipeline model with encouraging performances to identify gBRCA1/2 status based on US images of healthy ovaries, acquired on different US machines
Summary
Radiogenomics is a specific application of radiomics where imaging features are linked to genomic profiles. We aim to develop a radiogenomics model based on ovarian US images for predicting germline BRCA1/2 gene status in women with healthy ovaries. Multivariable analysis for classification of germline BRCA1/2 status was carried out via logistic regression, support vector machine, ensemble of decision trees and automated machine learning pipelines. The four strategies obtained a similar performance in terms of accuracy on the testing set (from 0.54 of logistic regression to 0.64 of the auto-machine learning pipeline). Data coming from one of the tested US machine showed generally higher performances, with the auto-machine learning pipeline (testing set specificity 0.87, negative predictive value 0.73, accuracy value 0.72 and 0.79 on training set). The study shows that a radiogenomics model on machine learning techniques is feasible and potentially useful for predicting gBRCA1/2 status in women with healthy ovaries. Around 10–20% of ovarian cancer and 6% breast cancer overall are caused by inheritable BRCA1/BRCA2 P Vs18
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
