Background: Neoadjuvant chemoradiotherapy (NCRT) followed by surgery is the standard treatment for patients with locally advanced rectal cancer. This study developed a random forest (RF) model to predict pathological complete response (pCR) based on radiomics derived from baseline 18 F-fluorodeoxyglucose ([ 18 F] FDG)-positron emission tomography (PET)/computed tomography (CT).Methods: This study included 169 patients with newly diagnosed rectal cancer. All patients received 18 F[FDG]-PET/CT, NCRT, and surgery. In total, 68 radiomic features were extracted from the metabolic tumor volume. The numbers of splits in a decision tree and trees in an RF were determined based on their effects on predictive performance. Receiver operating characteristic curve analysis was performed to evaluate predictive performance and ascertain the optimal threshold for maximizing prediction accuracy.Results: After NCRT, 22 patients (13%) achieved pCR, and 42 features that could differentiate tumors with pCR were used to construct the RF model. Six decision trees and seven splits were suitable. Accordingly, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 81.8%, 97.3%, 81.8%, 97.3%, and 95.3%, respectively. Conclusions: By using an RF, we determined that radiomics derived from baseline 18 F[FDG]-PET/CT could accurately predict pCR in patients with rectal cancer. Highly accurate and predictive values can be achieved but should be externally validated.
Background: Nonalcoholic fatty liver disease (NAFLD) is one of the most common chronic liver diseases. Studies have shown that sleep apnea is associated with NAFLD. However, studies on the association between sleep disorders in general and NAFLD are limited. We conducted a nationwide population-based longitudinal study to evaluate this potential association. Methods: We identified patients diagnosed with sleep disorders in the years 2000 through 2005 in Taiwan using the National Health Insurance Research Database and selected an equal number of patients without sleep disorders from the same database as the comparison cohort. The patients were followed from the index date to the diagnosis of NAFLD or the end of 2013. We used Cox proportional hazards models to estimate the risk of NAFLD associated with sleep disorders. Results: A total of 33,045 patients with sleep disorders were identified. The incidence of NAFLD was 14.0 per 10,000 person-year in patients with sleep disorders and 6.2 per 10,000 person-year in the comparison cohort. The adjusted hazard ratio (AHR) of NAFLD associated with sleep disorders was 1.78 (95% confidence interval [95%CI]: 1.46-2.16), and other independent risk factors included male sex (AHR = 1.31, 95%CI: 1.12-1.54), age 40-59 years (AHR = 1.49, 95%CI: 1.21-1.82), and dyslipidemia (AHR = 2.51, 95%CI: 2.08-3.04). In the subgroup analyses, both patients with (AHR = 2.24, 95%CI: 1.05-4.77) and without (AHR = 1.77, 95%CI: 1.46-2.15) sleep apnea had an increased risk of NAFLD. Conclusions: Sleep disorders are associated with NAFLD, even in patients without sleep apnea. Further studies are warranted to explore the mechanisms of the association.
BackgroundTo improve local control rate in patients with breast cancer receiving adjuvant radiotherapy after breast conservative surgery, additional boost dose to the tumor bed could be delivered simultaneously via the simultaneous integrated boost (SIB) modulated technique. However, the position of tumor bed kept changing during the treatment course as the treatment position was aligned to bony anatomy. This study aimed to analyze the positional uncertainties between bony anatomy and tumor bed, and a topology-based approach was derived to stratify patients with high variation in tumor bed localization.MethodsSixty patients with early-stage breast cancer or ductal carcinoma in situ were enrolled. All received adjuvant whole breast radiotherapy with or without local boost via SIB technique. The delineation of tumor bed was defined by incorporating the anatomy of seroma, adjacent surgical clips, and any architectural distortion on computed tomography simulation. A total of 1740 on-board images were retrospectively analyzed. Positional uncertainty of tumor bed was assessed by four components: namely systematic error (SE), and random error (RE), through anterior-posterior (AP), cranial-caudal (CC), left-right (LR) directions and couch rotation (CR). Age, tumor location, and body-mass factors including volume of breast, volume of tumor bed, breast thickness, and body mass index (BMI) were analyzed for their predictive role. The appropriate margin to accommodate the positional uncertainty of the boost volume was assessed, and the new plans with this margin for the tumor bed was designed as the high risk planning target volume (PTV-H) were created retrospectively to evaluate the impact on organs at risk.ResultsIn univariate analysis, a larger breast thickness, larger breast volume, higher BMI, and different tumor locations correlated with a greater positional uncertainty of tumor bed. However, BMI was the only factor associated with displacements of surgical clips in the multivariate analysis and patients with higher BMI were stratified as high variation group. When image guidance was aligned to bony structures, the SE and RE of clip displacement were consistently larger in the high variation group. The corresponding PTV-H margins for the high- and low-variation groups were 7, 10, 10 mm and 4, 9, 6 mm in AP, CC, LR directions, respectively. The heart dose between the two plans was not significantly different, whereas the dosimetric parameters for the ipsilateral lung were generally higher in the new plans.ConclusionsIn patients with breast cancer receiving adjuvant radiotherapy, a higher BMI is associated with a greater positional uncertainty of the boost tumor volume. More generous margin should be considered and it can be safely applied through proper design of beam arrangement with advanced treatment techniques.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.