ABSTRACT
Objective:Based on radiomics characteristics, different machine learning classification models are constructed to predict the gamma pass rate of dose verification in intensity-modulated radiotherapy for pelvic tumors, and to explore the best prediction model.Methods:The results of three-dimensional dose verification based on phantom measurement were retrospectively analyzed in 196 patients with pelvic tumor intensity-modulated radiotherapy plans. The gamma pass rate standard was 3%/2 mm and 10% dose threshold. Prediction models were constructed by extracting radiomic features based on dose documentation. Four machine learning algorithms, random forest, support vector machine, adaptive boosting, and gradient boosting decision tree were used to calculate the AUC value, sensitivity, and specificity respectively. The classification performance of the four prediction models was evaluated.Results:The sensitivity and specificity of the random forest, support vector machine, adaptive boosting, and gradient boosting decision tree models were 0.93, 0.85, 0.93, 0.96, 0.38, 0.69, 0.46, and 0.46, respectively. The AUC values were 0.81 and 0.82 for the random forest and adaptive boosting models, respectively, and 0.87 for the support vector machine and gradient boosting decision tree models.Conclusions:Machine learning method based on radiomics can be used to construct a prediction model of gamma pass rate for specific dosimetric verification of pelvic intensity-modulated radiotherapy. The classification performance of the support vector machine model and gradient boosting decision tree model is better than that of the random forest model and adaptive boosting model.
ABSTRACT
Objective: To compare the effectiveness of decompression and short fusion or long fusion for degenerative scoliosis (DS) with a Cobb angle of 20-40° combined with spinal stenosis. Methods: The clinical data of 50 patients with DS who were treated with decompression combined with short fusion or long fusion between January 2015 and May 2017 were retrospectively analysed. Patients were divided into long fusion group (fixed segments>3, 23 cases) and short fusion group (fixed segments≤3, 27 cases). There was no significant difference in gender, age, disease duration, and preoperative visual analogue scale (VAS) score of leg pain, Oswestry disability index (ODI), thoracic kyphosis (TK), thoracolumbar kyphosis (TLK), pelvic incidence (PI), pelvic title (PT), and sacral slope (SS) between the two groups ( P>0.05); however, the VAS score of low back pain, Cobb angle, and sagittal vertical axis (SVA) in long fusion group were significantly higher than those in short fusion group ( P0.05). The Cobb angle, SVA, TLK, and PT significantly decreased, while SS and LL significantly increased in the long fusion group ( P0.05). The improvements of Cobb angle, SVA, LL, PT, and SS in the long fusion group were significantly higher than those in the short fusion group at last follow-up ( P<0.05). There was no perioperative death in both groups. The incidence of complications in the long fusion group was 34.8% (8/23), which was significantly higher than that in the short fusion group [11.1% (3/27)] ( χ2=4.056, P=0.034). Conclusion: The DS patients with the Cobb angle of 20-40°can achieve satisfactory clinical outcomes and improve the spino-pelvic parameters by choosing appropriate fixation levels. Short fusion has less surgical trauma and fewer complications, whereas long fusion has more advantages in enhancing spino-pelvic parameters and relieving low back pain.