ABSTRACT
PRCIS: Machine learning (ML) based on the optical coherence tomography angiography vessel density features with different thresholds using a support vector machine (SVM) model provides excellent performance for glaucoma detection. BACKGROUND: To assess the classification performance of ML based on the 4 vessel density features of peripapillary optical coherence tomography angiography for glaucoma detection. METHODS: Images from 119 eyes of 119 glaucoma patients and 76 eyes of 76 healthy individuals were included. Four vessel density features of optical coherence tomography angiography images were developed using a threshold-based segmentation method and were integrated into 3 models of machine learning classifiers. Images were divided into 70% training set and 30% test set. Classification performances of SVM, random forest, and Gaussian Naive Bayes models were evaluated with the area under the receiver operating characteristic curve (AUC) and accuracy. RESULTS: Glaucoma eyes had lower vessel densities at different thresholds. For differentiating glaucoma eyes, the best results were achieved with 70% and 100% thresholds, in which SVM classifier discriminated glaucoma from healthy eyes with an AUC of 1 and accuracy of 1. After SVM, the random forest classifier with 100% thresholds showed an AUC of 0.993 and an accuracy of 0.994. Furthermore, the AUC of our ML performance (SVM) was 0.96 in a subgroup analysis of mild and moderate glaucoma eyes. CONCLUSIONS: ML based on the combined peripapillary vessel density features of total vessels and capillaries in the whole image and ring image could provide excellent performance for glaucoma detection.
Subject(s)
Glaucoma, Open-Angle , Glaucoma , Humans , Glaucoma, Open-Angle/diagnosis , Fluorescein Angiography/methods , Retinal Vessels , Tomography, Optical Coherence/methods , Bayes Theorem , Intraocular Pressure , Retinal Ganglion Cells , Visual Fields , Glaucoma/diagnosis , Machine LearningABSTRACT
BACKGROUND: Radical cystectomy (RC) with lymph node dissection is the mainstay of treatment for patients with muscle-invasive bladder cancer (MIBC) and high risk non-MIBC. The American Joint Committee on Cancer's (AJCC) node staging and lymph node ratio (LNR) systems are used in estimating prognosis; however, they do not directly factor in negative dissected nodes. In this study, we evaluated the log odds of positive lymph nodes (LODDS), a novel measure of nodal involvement, as a predictor of survival. PATIENTS AND METHODS: Eighty-three patients who underwent RC were retrospectively included and their demographic and clinical data were collected. Kaplan-Meier curve and Cox regression were used for survival analyses. RESULTS: Median number of dissected lymph nodes was 13 (range 3-45). ROC curve analysis indicated -0.92 as the optimal LODDS cutoff. LODDS > -0.92 was associated with higher T stage, lymphovascular invasion, and significantly worse overall survival (OS) (mean OS 18.6 vs. 45.1 months, P-value < .001). Furthermore, we evaluated AJCC node staging, LNR, and LODDS in three separate multivariable Cox regression models. Among 3 different measures of nodal disease burden, only LODDS was an independent predictor of OS (HR 2.71, 95% CI 1.28-5.73, P = .009). CONCLUSIONS: Our results show that LODDS is an independent predictor of OS and outperforms AJCC node staging and LNR in forecasting prognosis among patients with urothelial bladder cancer who undergo RC.