Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Apr 19;16(8):1570.
doi: 10.3390/cancers16081570.

Machine Learning Applied to Pre-Operative Computed-Tomography-Based Radiomic Features Can Accurately Differentiate Uterine Leiomyoma from Leiomyosarcoma: A Pilot Study

Miriam Santoro  1 , Vladislav Zybin  2 , Camelia Alexandra Coada  3 , Giulia Mantovani  4 , Giulia Paolani  1 , Marco Di Stanislao  4   5 , Cecilia Modolon  2 , Stella Di Costanzo  4 , Andrei Lebovici  6   7 , Gloria Ravegnini  8 , Antonio De Leo  5   9 , Marco Tesei  4 , Pietro Pasquini  4   5 , Luigi Lovato  2 , Alessio Giuseppe Morganti  5   10 , Maria Abbondanza Pantaleo  5   11 , Pierandrea De Iaco  4   5 , Lidia Strigari  1 , Anna Myriam Perrone  4   5
Affiliations

Machine Learning Applied to Pre-Operative Computed-Tomography-Based Radiomic Features Can Accurately Differentiate Uterine Leiomyoma from Leiomyosarcoma: A Pilot Study

Miriam Santoro et al. Cancers (Basel). .

Abstract

Background: The accurate discrimination of uterine leiomyosarcomas and leiomyomas in a pre-operative setting remains a current challenge. To date, the diagnosis is made by a pathologist on the excised tumor. The aim of this study was to develop a machine learning algorithm using radiomic data extracted from contrast-enhanced computed tomography (CECT) images that could accurately distinguish leiomyosarcomas from leiomyomas.

Methods: Pre-operative CECT images from patients submitted to surgery with a histological diagnosis of leiomyoma or leiomyosarcoma were used for the region of interest identification and radiomic feature extraction. Feature extraction was conducted using the PyRadiomics library, and three feature selection methods combined with the general linear model (GLM), random forest (RF), and support vector machine (SVM) classifiers were built, trained, and tested for the binary classification task (malignant vs. benign). In parallel, radiologists assessed the diagnosis with or without clinical data.

Results: A total of 30 patients with leiomyosarcoma (mean age 59 years) and 35 patients with leiomyoma (mean age 48 years) were included in the study, comprising 30 and 51 lesions, respectively. Out of nine machine learning models, the three feature selection methods combined with the GLM and RF classifiers showed good performances, with predicted area under the curve (AUC), sensitivity, and specificity ranging from 0.78 to 0.97, from 0.78 to 1.00, and from 0.67 to 0.93, respectively, when compared to the results obtained from experienced radiologists when blinded to the clinical profile (AUC = 0.73 95%CI = 0.62-0.84), as well as when the clinical data were consulted (AUC = 0.75 95%CI = 0.65-0.85).

Conclusions: CECT images integrated with radiomics have great potential in differentiating uterine leiomyomas from leiomyosarcomas. Such a tool can be used to mitigate the risks of eventual surgical spread in the case of leiomyosarcoma and allow for safer fertility-sparing treatment in patients with benign uterine lesions.

Keywords: artificial intelligence; computed tomography; diagnosis; leiomyoma; machine learning; radiomics; sarcoma.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflicts of interest.

Figures

Figure 1
Figure 1
Workflow analysis composed of two steps: creation of the radiomic model (A) and comparison between the machine-learning-based optimal classifiers (B) as detailed in the manuscript. GLCM: Gray Level Co-occurrence Matrix; GLRM: Gray Level Run Length Matrix; GLSZM: Gray Level Size Zone Matrix; NGTDM: Neighboring Gray Tone Difference Matrix; GLM: generalized linear model; RF: random forest; SVM: support vector machine; ROC: receiver operating characteristic; LASSO: Least Absolute Shrinkage and Selection Operator; RFE: recursive feature elimination.
Figure 2
Figure 2
ROC curves for the diagnostic accuracy of radiologists’ expert opinions in the presence of CT scans only (A) and with clinical data at hand (B). ROC curves for the diagnostic accuracy of a CT scoring system containing nine CT items (C) and with clinical suspicion (D). (E–G) ROC curves on the test dataset for the diagnostic accuracy of the best GLM-based (E), SVM-based (F), and RF-based (G) classifier models for each radiomic feature selection method. AUC: area under the curve; ROC: receiver operating characteristic; CT: computed tomography; ML: machine learning; GLM: general linear model; SVM: support vector machine; RF: random forest; LASSO: Least Absolute Shrinkage and Selection Operator.
Figure 3
Figure 3
Autocorrelation plots of the features selected by the optimal models reported in Table 3. Pearson’s rho coefficients are reported in the upper right panels, while p-values (i.e., p < 0, 0.001, 0.01, 0.05, 0.1, and 1) are expressed as symbols (i.e., “***”, “**”, “*”, “.”, and “ “, respectively). The blue, orange, and green dots represent the features selected by the optimal Boruta, Least Absolute Shrinkage and Selection Operator (LASSO), and recursive feature elimination (RFE) machine-learning-based approaches, respectively.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. D’Angelo E., Prat J. Uterine Sarcomas: A Review. Gynecol. Oncol. 2010;116:131–139. doi: 10.1016/j.ygyno.2009.09.023. - DOI - PubMed
    1. Astolfi A., Nannini M., Indio V., Schipani A., Rizzo A., Perrone A.M., De Iaco P., Pirini M.G., De Leo A., Urbini M., et al. Genomic Database Analysis of Uterine Leiomyosarcoma Mutational Profile. Cancers. 2020;12:2126. doi: 10.3390/cancers12082126. - DOI - PMC - PubMed
    1. Roberts M.E., Aynardi J.T., Chu C.S. Uterine Leiomyosarcoma: A Review of the Literature and Update on Management Options. Gynecol. Oncol. 2018;151:562–572. doi: 10.1016/j.ygyno.2018.09.010. - DOI - PubMed
    1. Lacuna K., Bose S., Ingham M., Schwartz G. Therapeutic Advances in Leiomyosarcoma. Front. Oncol. 2023;13:1149106. doi: 10.3389/fonc.2023.1149106. - DOI - PMC - PubMed
    1. Ricci S., Stone R.L., Fader A.N. Uterine Leiomyosarcoma: Epidemiology, Contemporary Treatment Strategies and the Impact of Uterine Morcellation. Gynecol. Oncol. 2017;145:208–216. doi: 10.1016/j.ygyno.2017.02.019. - DOI - PubMed

LinkOut - more resources

-