Your browser doesn't support javascript.
loading
Mostrar: 20 | 50 | 100
Resultados 1 - 2 de 2
Filtrar
Mais filtros

Base de dados
Ano de publicação
Tipo de documento
País de afiliação
Intervalo de ano de publicação
1.
Eur J Nucl Med Mol Imaging ; 51(5): 1451-1461, 2024 Apr.
Artigo em Inglês | MEDLINE | ID: mdl-38133687

RESUMO

PURPOSE: To evaluate if a machine learning prediction model based on clinical and easily assessable imaging features derived from baseline breast [18F]FDG-PET/MRI staging can predict pathologic complete response (pCR) in patients with newly diagnosed breast cancer prior to neoadjuvant system therapy (NAST). METHODS: Altogether 143 women with newly diagnosed breast cancer (54 ± 12 years) were retrospectively enrolled. All women underwent a breast [18F]FDG-PET/MRI, a histopathological workup of their breast cancer lesions and evaluation of clinical data. Fifty-six features derived from positron emission tomography (PET), magnetic resonance imaging (MRI), sociodemographic / anthropometric, histopathologic as well as clinical data were generated and used as input for an extreme Gradient Boosting model (XGBoost) to predict pCR. The model was evaluated in a five-fold nested-cross-validation incorporating independent hyper-parameter tuning within the inner loops to reduce the risk of overoptimistic estimations. Diagnostic model-performance was assessed by determining the area under the curve of the receiver operating characteristics curve (ROC-AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy. Furthermore, feature importances of the XGBoost model were evaluated to assess which features contributed most to distinguish between pCR and non-pCR. RESULTS: Nested-cross-validation yielded a mean ROC-AUC of 80.4 ± 6.0% for prediction of pCR. Mean sensitivity, specificity, PPV, and NPV of 54.5 ± 21.3%, 83.6 ± 4.2%, 63.6 ± 8.5%, and 77.6 ± 8.1% could be achieved. Histopathological data were the most important features for classification of the XGBoost model followed by PET, MRI, and sociodemographic/anthropometric features. CONCLUSION: The evaluated multi-source XGBoost model shows promising results for reliably predicting pathological complete response in breast cancer patients prior to NAST. However, yielded performance is yet insufficient to be implemented in the clinical decision-making process.


Assuntos
Neoplasias da Mama , Humanos , Feminino , Neoplasias da Mama/diagnóstico por imagem , Neoplasias da Mama/terapia , Fluordesoxiglucose F18 , Estudos Retrospectivos , Imageamento por Ressonância Magnética/métodos , Tomografia por Emissão de Pósitrons , Aprendizado de Máquina
2.
PLoS One ; 19(7): e0307998, 2024.
Artigo em Inglês | MEDLINE | ID: mdl-39074093

RESUMO

PURPOSE: This study aimed to evaluate the prognostic potential of pre-therapeutic [18F]FDG-PET/CT variables regarding prediction of progression-free survival (PFS) and overall survival (OS) in NSCLC-patients. METHOD: NSCLC-patients who underwent pre-therapeutic [18F]FDG-PET/CT were retrospectively analyzed. The following imaging features were collected from the primary tumor: tumor size, tumor density, central necrosis, spicules and SUVmax. For standardization, an indexSUVmax was calculated (SUVmax primary tumor/SUVmax liver). Descriptive statistics and correlations of survival time analyses for PFS and OS were calculated using the Kaplan-Meier method and Cox regression including a hazard ratio (HR). A value of p < 0.05 was set as statistically significant. The 95%-confidence intervals (CI) were calculated. The median follow-up time was 63 (IQR 27-106) months. RESULTS: This study included a total of 82 patients (25 women, 57 men; mean age: 66 ± 9 years). IndexSUVmax (PFS: HR = 1.0, CI: 1.0-1.1, p = 0.49; OS: HR = 1.0, CI: 0.9-1.2, p = 0.41), tumor size (PFS: HR = 1.0, CI: 0.9-1.0, p = 0.08; OS: HR = 1.0, CI: 0.9-1.0, p = 0.07), tumor density (PFS: HR = 0.9, CI: 0.6-1.4, p = 0.73; OS: HR = 0.3; CI: 0.1-1.1; p = 0.07), central necrosis (PFS: HR = 1.0, CI: 0.6-1.8, p = 0.98; OS: HR = 0.6, CI: 0.2-1.9, p = 0.40) and spicules (PFS: HR = 1.0, CI: 0.6-1.9, p = 0.91; OS: HR = 1.3, CI: 0.4-3.7, p = 0.65) did not significantly affect PFS and OS in the study population. An optimal threshold value for the indexSUVmax was determined by ROC analysis and Youden's index. There was no significant difference in PFS with an indexSUVmax-threshold of 3.8 (13 vs. 27 months; p = 0.45) and in OS with an indexSUVmax-threshold of 4.0 (113 vs. 106 months; p = 0.40). CONCLUSIONS: SUVmax and morphologic parameters from pre-therapeutic [18F]FDG-PET/CT were not able to predict PFS and OS in NSCLC-patients.


Assuntos
Carcinoma Pulmonar de Células não Pequenas , Fluordesoxiglucose F18 , Neoplasias Pulmonares , Tomografia por Emissão de Pósitrons combinada à Tomografia Computadorizada , Intervalo Livre de Progressão , Humanos , Carcinoma Pulmonar de Células não Pequenas/diagnóstico por imagem , Carcinoma Pulmonar de Células não Pequenas/patologia , Carcinoma Pulmonar de Células não Pequenas/mortalidade , Masculino , Feminino , Idoso , Tomografia por Emissão de Pósitrons combinada à Tomografia Computadorizada/métodos , Neoplasias Pulmonares/diagnóstico por imagem , Neoplasias Pulmonares/mortalidade , Neoplasias Pulmonares/patologia , Pessoa de Meia-Idade , Prognóstico , Estudos Retrospectivos , Compostos Radiofarmacêuticos , Estimativa de Kaplan-Meier
SELEÇÃO DE REFERÊNCIAS
DETALHE DA PESQUISA