AUTHOR=Liu Xue-Fei , Lu Jing-Jing , Li Meng-Die , Li Ying , Zeng An-Rong , Qiang Jin-Wei TITLE=Prediction of pre-eclampsia by using radiomics nomogram from gestational hypertension patients JOURNAL=Frontiers in Neuroscience VOLUME=16 YEAR=2022 URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2022.961348 DOI=10.3389/fnins.2022.961348 ISSN=1662-453X ABSTRACT=Background

Pre-eclampsia (PE) is the main cause of death in maternal and prenatal morbidity. No effective clinical tools could be used for the prediction of PE. A radiomics nomogram based on diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) maps was established to predict PE from gestational hypertension (GH).

Materials and methods

A total of 138 patients with hypertensive disorders of pregnancy were continuously enrolled in the study prospectively, namely, 58 patients with PE and 80 patients with GH. The patients were randomly divided into a training cohort (n = 97) and a test cohort (n = 41). Radiomics features were extracted from DWI and ADC maps. The radiomics signature was constructed using a least absolute shrinkage and selection operator (LASSO) algorithm in the training cohort. A radiomics nomogram was developed by combining the radiomics signature with the selected clinical risk factors. The area under the receiver operating characteristic (ROC) curves (AUC), specificity, sensitivity, accuracy, positive predictive value, and negative predictive values of the radiomics signature, clinical risk factors, and radiomics nomogram were calculated. Decision curve analysis (DCA) was performed to determine the clinical usefulness of the radiomics nomogram.

Results

The LASSO analysis finally included 11 radiomics features, which were defined as the radiomics signature. The individualized prediction nomogram was constructed by integrating the radiomics signature, maternal age, and body mass index (BMI). The nomogram exhibited a good performance both in the training cohort [AUC of 0.89 (95% CI, 0.82–0.95)] and test cohort [AUC of 0.85 (95% CI, 0.73–0.97)] for predicting PE from GH. The DCA indicated that clinicians and patients could benefit from the use of radiomics nomogram.

Conclusion

The radiomics nomogram could individually predict PE from GH. The nomogram could be conveniently used to facilitate the treatment decision for clinicians and patients.