Scott R. Silva
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Articles (5)
Year:
2026
Year:
2026
Year:
2026
Background Gynecological cancers are among the most prevalent cancers in women worldwide. Brachytherapy, often used as a boost to external beam radiotherapy, is integral to treatment. Advances in computation, algorithms, and data availability have popularized machine learning. Objective To develop and compare machine learning models for predicting grade 3 or higher toxicities in gynecological cancer patients treated with high dose rate (HDR) brachytherapy, aiming to contribute to personalized radiation treatments. Methods A retrospective analysis on gynecological cancer patients who underwent HDR brachytherapy with Syed-Neblett or Tandem and Ovoid applicators from 2009 to 2023. After exclusions, 233 patients were included. Dosimetric variables for the high-risk clinical target volume (HR-CTV) and organs at risk, along with tumor, patient, and toxicity data, were collected and compared between groups with and without grade 3 or higher toxicities using statistical tests. Six supervised classification machine learning models (Logistic Regression, Random Forest, K-Nearest Neighbors, Support Vector Machines, Gaussian Naive Bayes, and Multi-Layer Perceptron Neural Networks) were constructed and evaluated. The construction process involved sequential feature selection (SFS) when appropriate, followed by hyperparameter tuning. Final model performance was characterized using a 25% withheld test dataset. Results The top three ranking models were Support Vector Machines, Random Forest, and Logistic Regression, with F1 testing scores of 0.63, 0.57, and 0.52; normMCC testing scores of 0.75, 0.77, and 0.71; and accuracy testing scores of 0.80, 0.85, and 0.81, respectively. The SFS algorithm selected 10 features for the highest-ranking model. In traditional statistical analysis, HR-CTV volume, Charlson Comorbidity Index, Length of Follow-Up, and D2cc - Rectum differed significantly between groups with and without grade 3 or higher toxicities. Conclusions Machine learning models were developed to predict grade 3 or higher toxicities, achieving satisfactory performance. Machine learning presents a novel solution to creating multivariable models for personalized radiation therapy care.
Year:
2024
Background Verrucous vulvar carcinoma (VC) is an uncommon and distinct histologic subtype of squamous cell carcinoma (SCC). The available literature on VC is currently limited to case reports and small single institution studies. Aims The goals of this study were to analyze data from the National Cancer Database (NCDB) to quantitate the incidence of VC and to investigate the effects of patient demographics, tumor characteristics, and treatment regimens on overall survival (OS) in women with verrucous vulvar carcinoma. Methods and results Patients diagnosed with vulvar SCC or VC between the years of 2004 and 2016 were identified in the NCDB. OS was assessed with Kaplan–Meier curves and the log‐rank test. Construction of a Cox model compared survival after controlling for confounding variables. The reported incidence of SCC of the vulva has significantly increased since 2004 ( p < .0001). In contrast, the incidence of VC has remained stable ( p = .344) since 2004. Compared to SCC, VC was significantly more likely to be diagnosed in older women ( p < .0001) and treated with surgery alone ( p < .0001). However, on propensity score weighted analysis there was a trend toward improved 5‐year OS in women with VC compared to those with SCC (63.4% vs. 57.7%, p = .0794). Multivariable Cox survival analysis showed an improvement in OS in VC patients treated with both primary site and regional lymph node surgery compared to primary site surgery alone (adjusted hazard ratio [aHR] 0.67, 95% confidence interval [CI] 0.46–0.97, p = .0357). Conclusion Verrucous carcinoma is more likely to present in older women. Regional lymph node surgery in addition to primary site surgery significantly improves OS in VC patients.
Year:
2022
Collaborators (5)
Magdalena Stankiewicz
MD PhD
Maria Sklodowska-Curie National Research Institute of Oncology
Lorraine Portelance
Professor/ Vice-Chair/Clinic Div Chief
H Lee Moffitt Cancer Center and Research Institute
Jeremy Gaskins
University of Louisville
Allison Quick
Ohio State University
Anupam Rishi
Assistant Professor
H Lee Moffitt Cancer Center and Research Institute
Make your next move.
Prepare your application
Review drafts before you send them.