Rationale and Objectives
To develop and validate a combined model integrating clinical and radiomic features
to non-invasive discriminate between the benign and malignant solid ovarian tumors.
Materials and Methods
A total of 148 patients with 156 solid ovarian tumors (86 benign and 70 malignant
tumors) were included in this study. The dataset was split into the training and the
test set with a ratio of 8:2 using stratified random sampling. 12 clinical features
and 1612 radiomic features were extracted from each tumor. These features were selected
by least absolute shrinkage and selection operator (Lasso). Three classification models
were built using extreme gradient boosting (XGB) algorithm: clinical model, radiomic
model, combined model. The area under the receiver operating characteristic curve
(AUC), accuracy, precision and sensitivity were analyzed to evaluate the performance
of these models.
Results
All of the three models obtained good performances in differentiating benign with
malignant solid ovarian tumors in both training and test sets. The AUC, accuracy,
precision, sensitivity of clinical model and radiomic model in test set were 0.847
(95% confidence interval (CI), 0.707-0.986, p <0.01), 0.774, 0.769, 0.714, and 0.807 (95%CI, 0.652-0.961, p <0.05), 0.677, 0.643, 0.643, respectively. Combined model had the best prediction
results, the AUC, accuracy, precision and sensitivity were 0.954 (95%CI, 0.862-1.0,
p <0.01), 0.839, 0.909 and 0.714 in test set.
Conclusion
Radiomics based on machine learning can be helpful for radiologists in differentiating
the benign and malignant solid ovarian tumors.
Key Words
To read this article in full you will need to make a payment
Purchase one-time access:
Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online accessOne-time access price info
- For academic or personal research use, select 'Academic and Personal'
- For corporate R&D use, select 'Corporate R&D Professionals'
Subscribe:
Subscribe to Academic RadiologyAlready a print subscriber? Claim online access
Already an online subscriber? Sign in
Register: Create an account
Institutional Access: Sign in to ScienceDirect
References
- Developing an MR imaging strategy for diagnosis of ovarian masses.Radiographics. 2006; 26: 1431-1448https://doi.org/10.1148/rg.265045206
- Clinical characteristics and surgical management options for ovarian fibroma/fibrothecoma: a study of 97 cases.Gynecol Obstet Invest. 2013; 76: 182-187https://doi.org/10.1159/000354555
- Laparoscopic surgical management and clinical characteristics of ovarian fibromas.Jsls. 2011; 15: 16-20https://doi.org/10.4293/108680810x12924466009087
- Magnetic resonance imaging features of ovarian fibroma, fibrothecoma, and thecoma.Abdom Imaging. 2015; 40: 1263-1272https://doi.org/10.1007/s00261-014-0257-z
- Primary fallopian tube carcinoma.Eur J Obstet Gynecol Reprod Biol. 2013; 169: 155-161https://doi.org/10.1016/j.ejogrb.2013.03.023
- Fallopian tube carcinoma.J Oncol Pract. 2019; 15: 375-382https://doi.org/10.1200/jop.18.00662
- Validity and reproducibility of the ADNEX MR scoring system in the diagnosis of sonographically indeterminate adnexal masses.J Magn Reson Imaging. 2021; 53: 292-304https://doi.org/10.1002/jmri.27285
- Ovarian-Adnexal Reporting Data System Magnetic Resonance Imaging (O-RADS MRI) score for risk stratification of sonographically indeterminate adnexal masses.JAMA Netw Open. 2020; 3e1919896https://doi.org/10.1001/jamanetworkopen.2019.19896
- MR imaging of ovarian masses: classification and differential diagnosis.Insights Imaging. 2016; 7: 21-41https://doi.org/10.1007/s13244-015-0455-4
- Artificial intelligence in radiology.Nat Rev Cancer. 2018; 18: 500-510https://doi.org/10.1038/s41568-018-0016-5
- Artificial intelligence in medical practice: the question to the answer?.Am J Med. 2018; 131: 129-133https://doi.org/10.1016/j.amjmed.2017.10.035
- High-performance medicine: the convergence of human and artificial intelligence.Nat Med. 2019; 25: 44-56https://doi.org/10.1038/s41591-018-0300-7
- Thirty years of artificial intelligence in medicine (AIME) conferences: a review of research themes.Artif Intell Med. 2015; 65: 61-73https://doi.org/10.1016/j.artmed.2015.07.003
- Machine learning in oncology: a clinical appraisal.Cancer Lett. 2020; 481: 55-62https://doi.org/10.1016/j.canlet.2020.03.032
- Machine learning in medicine.Circulation. 2015; 132: 1920-1930https://doi.org/10.1161/circulationaha.115.001593
- Machine learning for medical imaging.Radiographics. 2017; 37: 505-515https://doi.org/10.1148/rg.2017160130
- Introduction to radiomics.J Nucl Med. 2020; 61: 488-495https://doi.org/10.2967/jnumed.118.222893
- Radiomics: images are more than pictures, they are data.Radiology. 2016; 278: 563-577https://doi.org/10.1148/radiol.2015151169
- CT-based radiomics to predict the pathological grade of bladder cancer.Eur Radiol. 2020; 30: 6749-6756https://doi.org/10.1007/s00330-020-06893-8
- Radiomics analysis for evaluation of pathological complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer.Clin Cancer Res. 2017; 23: 7253-7262https://doi.org/10.1158/1078-0432.Ccr-17-1038
- Deep learning-based radiomics predicts response to chemotherapy in colorectal liver metastases.Med Phys. 2021; 48: 513-522https://doi.org/10.1002/mp.14563
- Radiomics derived from dynamic contrast-enhanced MRI pharmacokinetic protocol features: the value of precision diagnosis ovarian neoplasms.Eur Radiol. 2021; 31: 368-378https://doi.org/10.1007/s00330-020-07112-0
- MDCT-based radiomics features for the differentiation of serous borderline ovarian tumors and serous malignant ovarian tumors.Cancer Manag Res. 2021; 13: 329-336https://doi.org/10.2147/cmar.S284220
- Magnetic resonance imaging radiomics in categorizing ovarian masses and predicting clinical outcome: a preliminary study.Eur Radiol. 2019; 29: 3358-3371https://doi.org/10.1007/s00330-019-06124-9
- Ovarian-adnexal reporting lexicon for mri: a white paper of the acr ovarian-adnexal reporting and data systems mri committee.J Am Coll Radiol. 2021; 18: 713-729https://doi.org/10.1016/j.jacr.2020.12.022
- The accuracy of multi-detector computed tomography and laparoscopy in the prediction of peritoneal carcinomatosis index score in primary ovarian cancer.Acad Radiol. 2019; 26: 1650-1658https://doi.org/10.1016/j.acra.2019.04.005
Cui L, Xu H, Zhang YJAR Diagnostic accuracies of the ultrasound and magnetic resonance imaging ADNEX scoring systems for ovarian adnexal mass: systematic review and meta-analysis. Acad Radiol 2022; 29:897–908
- Apparent diffusion coefficient (ADC) measurement in ovarian tumor: effect of region-of-interest methods on ADC values and diagnostic ability.J Magn Reson Imaging. 2016; 43: 720-725https://doi.org/10.1002/jmri.25011
- Diagnostic accuracy of diffusion-weighted imaging with conventional MR imaging for differentiating complex solid and cystic ovarian tumors at 1.5T.World J Surg Oncol. 2012; 10: 237https://doi.org/10.1186/1477-7819-10-237
- Complete tumor encapsulation on magnetic resonance imaging: a potentially useful imaging biomarker for better survival in solitary large hepatocellular carcinoma.Liver Transpl. 2013; 19: 283-291https://doi.org/10.1002/lt.23597
- Tumor border pattern and size help predict lymph node status in papillary microcarcinoma: a clinicopathologic study.Ann Diagn Pathol. 2020; 48151592https://doi.org/10.1016/j.anndiagpath.2020.151592
- Differentiation between benign and malignant ovarian masses using multiparametric MRI.Diagn Interv Imaging. 2020; 101: 147-155https://doi.org/10.1016/j.diii.2020.01.006
- Correlation between apparent diffusion coefficient (ADC) and cellularity is different in several tumors: a meta-analysis.Oncotarget. 2017; 8: 59492-59499https://doi.org/10.18632/oncotarget.17752
- Transvaginal color doppler ultrasonography and CA-125 in suspicious adnexal masses.Int J Gynaecol Obstet. 1999; 66: 255-261https://doi.org/10.1016/s0020-7292(99)00104-6
- Radiologists make more errors interpreting off-hours body ct studies during overnight assignments as compared with daytime assignments.Radiology. 2020; 297: 374-379https://doi.org/10.1148/radiol.2020201558
- MR imaging of epithelial ovarian cancer: a combined model to predict histologic subtypes.Eur Radiol. 2020; 30: 5815-5825https://doi.org/10.1007/s00330-020-06993-5
- MR image-based radiomics to differentiate type Ι and type ΙΙ epithelial ovarian cancers.Eur Radiol. 2021; 31: 403-410https://doi.org/10.1007/s00330-020-07091-2
- Bosniak IIF and III renal cysts: can apparent diffusion coefficient-derived texture features discriminate between malignant and benign IIF and III cysts?.J Comput Assist Tomogr. 2019; 43: 485-492https://doi.org/10.1097/rct.0000000000000851
- The development and validation of a ct-based radiomics nomogram to preoperatively predict lymph node metastasis in high-grade serous ovarian cancer.Front Oncol. 2021; 11711648https://doi.org/10.3389/fonc.2021.711648
- Multiparametric MRI radiomic model for preoperative predicting who/isup nuclear grade of clear cell renal cell carcinoma.J Magn Reson Imaging. 2020; 52: 1557-1566https://doi.org/10.1002/jmri.27182
- Radiomics of high-grade serous ovarian cancer: association between quantitative CT features, residual tumour and disease progression within 12 months.Eur Radiol. 2018; 28: 4849-4859https://doi.org/10.1007/s00330-018-5389-z
- A combined radiomics and clinical variables model for prediction of malignancy in T2 hyperintense uterine mesenchymal tumors on MRI.Eur Radiol. 2021; 31: 6125-6135https://doi.org/10.1007/s00330-020-07678-9
- Diagnosis of ovarian neoplasms using nomogram in combination with ultrasound image-based radiomics signature and clinical factors.Front Genet. 2021; 12753948https://doi.org/10.3389/fgene.2021.753948
Article info
Publication history
Published online: July 07, 2022
Accepted:
June 8,
2022
Received in revised form:
May 31,
2022
Received:
May 4,
2022
Identification
Copyright
© 2022 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.
