Search: WFRF:(Franchi Dorella) >
Multiclass risk mod...
Multiclass risk models for ovarian malignancy : an illustration of prediction uncertainty due to the choice of algorithm
-
- Ledger, Ashleigh (author)
- Catholic University of Leuven
-
- Ceusters, Jolien (author)
- Leuven Cancer Institute,Catholic University of Leuven
-
- Valentin, Lil (author)
- Lund University,Lunds universitet,Obstetrisk, gynekologisk och prenatal ultraljudsdiagnostik,Forskargrupper vid Lunds universitet,Obstetric, Gynaecological and Prenatal Ultrasound Research,Lund University Research Groups,Skåne University Hospital
-
show more...
-
- Testa, Antonia (author)
- Policlinico Universitario Agostino Gemelli,Catholic University of the Sacred Heart, Rome
-
- Van Holsbeke, Caroline (author)
- Hospital Oost-Limburg
-
- Franchi, Dorella (author)
- European Institute of Oncology
-
- Bourne, Tom (author)
- University Hospitals Leuven,Catholic University of Leuven,Queen Charlotte's and Chelsea Hospital
-
- Froyman, Wouter (author)
- Catholic University of Leuven,University Hospitals Leuven
-
- Timmerman, Dirk (author)
- Catholic University of Leuven,University Hospitals Leuven
-
- Van Calster, Ben (author)
- Leiden University Medical Centre,Catholic University of Leuven
-
show less...
-
(creator_code:org_t)
- 2023
- 2023
- English.
-
In: BMC Medical Research Methodology. - 1471-2288. ; 23:1
- Related links:
-
http://dx.doi.org/10... (free)
-
show more...
-
https://lup.lub.lu.s...
-
https://doi.org/10.1...
-
show less...
Abstract
Subject headings
Close
- Background: Assessing malignancy risk is important to choose appropriate management of ovarian tumors. We compared six algorithms to estimate the probabilities that an ovarian tumor is benign, borderline malignant, stage I primary invasive, stage II-IV primary invasive, or secondary metastatic. Methods: This retrospective cohort study used 5909 patients recruited from 1999 to 2012 for model development, and 3199 patients recruited from 2012 to 2015 for model validation. Patients were recruited at oncology referral or general centers and underwent an ultrasound examination and surgery ≤ 120 days later. We developed models using standard multinomial logistic regression (MLR), Ridge MLR, random forest (RF), XGBoost, neural networks (NN), and support vector machines (SVM). We used nine clinical and ultrasound predictors but developed models with or without CA125. Results: Most tumors were benign (3980 in development and 1688 in validation data), secondary metastatic tumors were least common (246 and 172). The c-statistic (AUROC) to discriminate benign from any type of malignant tumor ranged from 0.89 to 0.92 for models with CA125, from 0.89 to 0.91 for models without. The multiclass c-statistic ranged from 0.41 (SVM) to 0.55 (XGBoost) for models with CA125, and from 0.42 (SVM) to 0.51 (standard MLR) for models without. Multiclass calibration was best for RF and XGBoost. Estimated probabilities for a benign tumor in the same patient often differed by more than 0.2 (20% points) depending on the model. Net Benefit for diagnosing malignancy was similar for algorithms at the commonly used 10% risk threshold, but was slightly higher for RF at higher thresholds. Comparing models, between 3% (XGBoost vs. NN, with CA125) and 30% (NN vs. SVM, without CA125) of patients fell on opposite sides of the 10% threshold. Conclusion: Although several models had similarly good performance, individual probability estimates varied substantially.
Subject headings
- MEDICIN OCH HÄLSOVETENSKAP -- Klinisk medicin -- Cancer och onkologi (hsv//swe)
- MEDICAL AND HEALTH SCIENCES -- Clinical Medicine -- Cancer and Oncology (hsv//eng)
Keyword
- Calibration
- Machine learning
- Multiclass models
- Ovarian Neoplasms
- Prediction models
Publication and Content Type
- art (subject category)
- ref (subject category)
Find in a library
To the university's database
- By the author/editor
-
Ledger, Ashleigh
-
Ceusters, Jolien
-
Valentin, Lil
-
Testa, Antonia
-
Van Holsbeke, Ca ...
-
Franchi, Dorella
-
show more...
-
Bourne, Tom
-
Froyman, Wouter
-
Timmerman, Dirk
-
Van Calster, Ben
-
show less...
- About the subject
-
- MEDICAL AND HEALTH SCIENCES
-
MEDICAL AND HEAL ...
-
and Clinical Medicin ...
-
and Cancer and Oncol ...
- Articles in the publication
-
BMC Medical Rese ...
- By the university
-
Lund University