ObjectiveTo observe the effectiveness of radiotherapy for refractory choroidal hemangioma. MethodsEight patients (8 eyes) with choroidal hemangioma were enrolled in this retrospective study. All the patients had received laser or photodynamic therapy before without effectiveness. The patients included 7 males and 1 females. The age was ranged from 11 to 54 years old, with an average of (27.50±15.18) years. All the patients were affected unilaterally, including 3 right eyes and 5 left eyes. There were 5 eyes with circumscribed choroidal hemangioma, 3 eyes with diffused choroidal hemangioma. All eyes had extensively exudative retinal detachment. The vision was from light sensation to 0.01. The volume of the tumors was ranged from 1.96 to 5.35 cm3, with a mean of (3.37±1.06) cm3. All the patients were treated with X rays by conventional fractional radiotherapy. Four of 8 patients were applied 24Gy totally in 8 fractions, while the other 4 patients were applied 46Gy in 23 fractions. Follow-up period ranged from 7 to 95 months, with medium of 42 months. ResultsRetinas reattached in all the eyes while exudation being absorbed. No retinal detachment happened again. To the last follow-up, the vision was from light sensation to 0.6. Visual activity improved in 6 eyes while 2 eyes improved obviously. Visual acuity was stable in remaining 2 eyes. The volume of the tumors decreased to 1.24-2.16 cm3, with a mean of (1.68±0.30) cm3. The percentage of the tumor decreased by 14.6-72.7, with an average of (44.89±21.30)%. No radiotherapy-associated complication occurred. ConclusionRadiotherapy is an efficient and safe treatment for refractory choroidal hemangioma.
This study aims to predict expression of Ki67 molecular marker in pancreatic cystic neoplasm using radiomics. We firstly manually segmented tumor area in multi-detector computed tomography (MDCT) images. Then 409 high-throughput features were automatically extracted and the least absolute shrinkage selection operator (LASSO) regression model was used for feature selection. After 200 bootstrapping repetitions of LASSO, 20 most frequently selected features made up the optimal feature set. Then 200 bootstrapping repetitions of support vector machine (SVM) classifier with 10-fold cross-validation were used to avoid overfitting and accurately predict the Ki67 expression. The highest prediction accuracy could achieve 85.29% and the highest area under the receiver operating characteristic curve (AUC) was 91.54% with a sensitivity (SENS) of 81.88% and a specificity (SPEC) of 86.75%. According to the results of experiment, the feasibility of predicting expression of Ki67 in pancreatic cystic neoplasm based on radiomics was verified.