RESUMO
PURPOSE: Predicting recurrence following stereotactic body radiotherapy (SBRT) for non-small cell lung cancer provides important information for the feasibility of the individualized radiotherapy and allows to select the appropriate treatment strategy based on the risk of recurrence. In this study, we evaluated the performance of both machine learning models using positron emission tomography (PET) and computed tomography (CT) radiomic features for predicting recurrence after SBRT. METHODS: Planning CT and PET images of 82 non-small cell lung cancer patients who performed SBRT at our hospital were used. First, tumors were delineated on each CT and PET of each patient, and 111 unique radiomic features were extracted, respectively. Next, the 10 features were selected using three different feature selection algorithms, respectively. Recurrence prediction models based on the selected features and four different machine learning algorithms were developed, respectively. Finally, we compared the predictive performance of each model for each recurrence pattern using the mean area under the curve (AUC) calculated following the 0.632+ bootstrap method. RESULTS: The highest performance for local recurrence, regional lymph node metastasis, and distant metastasis were observed in models using Support vector machine with PET features (mean AUC = 0.646), Naive Bayes with PET features (mean AUC = 0.611), and Support vector machine with CT features (mean AUC = 0.645), respectively. CONCLUSIONS: We comprehensively evaluated the performance of prediction model developed for recurrence following SBRT. The model in this study would provide information to predict the recurrence pattern and assist in making treatment strategies.
Assuntos
Carcinoma Pulmonar de Células não Pequenas , Neoplasias Pulmonares , Aprendizado de Máquina , Radiocirurgia , Planejamento da Radioterapia Assistida por Computador , Idoso , Idoso de 80 Anos ou mais , Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Algoritmos , Carcinoma Pulmonar de Células não Pequenas/diagnóstico por imagem , Carcinoma Pulmonar de Células não Pequenas/radioterapia , Carcinoma Pulmonar de Células não Pequenas/cirurgia , Processamento de Imagem Assistida por Computador/métodos , Neoplasias Pulmonares/diagnóstico por imagem , Neoplasias Pulmonares/radioterapia , Neoplasias Pulmonares/cirurgia , Recidiva Local de Neoplasia/diagnóstico por imagem , Tomografia por Emissão de Pósitrons combinada à Tomografia Computadorizada/métodos , Prognóstico , Radiômica , Radiocirurgia/métodos , Dosagem Radioterapêutica , Planejamento da Radioterapia Assistida por Computador/métodos , Radioterapia de Intensidade Modulada/métodos , Estudos RetrospectivosRESUMO
BACKGROUND: Surface-guided radiotherapy (SGRT) is adopted by several institutions; however, reports on the phantoms used to assess the precision of the SGRT setup are limited. PURPOSE: The purpose of this study was to develop a phantom to verify the accuracy of the irradiation position during skin mark-less SGRT. METHODS: An acrylonitrile butadiene styrene (ABS) plastic cube phantom with a diameter of 150 mm on each side containing a dummy target of 15 mm and two types of body surface-shaped phantoms (breast/face shape) that could be attached to the cube phantom were fabricated. Films can be inserted on four sides of the cubic phantom (left, right, anterior and posterior), and the center of radiation can be calculated by irradiating the dummy target with orthogonal MV beams. Three types of SGRT using a VOXELAN-HEV600M (Electronics Research&Development Corporation, Okayama, Japan) were evaluated using this phantom: (i) SGRTCT-a SGRT set-up based solely on a computed tomography (CT)-reference image. (ii) SGRTCT + CBCT-a method where cone beam computed tomography (CBCT) matching was performed after SGRTCT. (iii) SGRTScan-a resetup technique using a scan reference image obtained after completing the (ii) step. RESULTS: Both the breast and face phantoms were recognized in the SGRT system without problems. SGRTScan ensure precision within 1 mm/1° for breast and face verification, respectively. All SGRT methods showed comparable rotational accuracies with no significant disparities. CONCLUSIONS: The developed phantom was useful for verifying the accuracy of skin mark-less SGRT position matching. The SGRTScan demonstrated the feasibility of achieving skin-mark less SGRT with high accuracy, with deviations of less than 1 mm. Additional research is necessary to evaluate the suitability of the developed phantoms for use in various facilities and systems. This phantom could be used for postal surveys in the future.