Sensitivity and specificity of machine learning classifiers for glaucoma diagnosis using Spectral Domain OCT and standard automated perimetry

PURPOSE: To evaluate the sensitivity and specificity of machine learning classifiers (MLCs) for glaucoma diagnosis using Spectral Domain OCT (SD-OCT) and standard automated perimetry (SAP). METHODS: Observational cross-sectional study. Sixty two glaucoma patients and 48 healthy individuals were included. All patients underwent a complete ophthalmologic examination, achromatic standard automated perimetry (SAP) and retinal nerve fiber layer (RNFL) imaging with SD-OCT (Cirrus HD-OCT; Carl Zeiss Meditec Inc., Dublin, California). Receiver operating characteristic (ROC) curves were obtained for all SD-OCT parameters and global indices of SAP. Subsequently, the following MLCs were tested using parameters from the SD-OCT and SAP: Bagging (BAG), Naive-Bayes (NB), Multilayer Perceptron (MLP), Radial Basis Function (RBF), Random Forest (RAN), Ensemble Selection (ENS), Classification Tree (CTREE), Ada Boost M1(ADA),Support Vector Machine Linear (SVML) and Support Vector Machine Gaussian (SVMG). Areas under the receiver operating characteristic curves (aROC) obtained for isolated SAP and OCT parameters were compared with MLCs using OCT+SAP data. RESULTS: Combining OCT and SAP data, MLCs' aROCs varied from 0.777(CTREE) to 0.946 (RAN).The best OCT+SAP aROC obtained with RAN (0.946) was significantly larger the best single OCT parameter (p<0.05), but was not significantly different from the aROC obtained with the best single SAP parameter (p=0.19). CONCLUSION: Machine learning classifiers trained on OCT and SAP data can successfully discriminate between healthy and glaucomatous eyes. The combination of OCT and SAP measurements improved the diagnostic accuracy compared with OCT data alone.

Cálculos teóricos de afinidades por próton de n-alquilaminas usando o método ONIOM

The ONIOM method was used to calculate the proton affinities (PA) of n-alkylamines (CnH2n+1NH2, n = 3 to 6, 8, 10, 12, 14, 16 and 18). The calculations were carried out at several levels (HF, MP2, B3LYP, QCISD(T), ...) using Pople basis sets and at the QCISD(T) level using basis sets developed by the generator coordinate method (GCM) and adapted to effective core potentials. PAs were also obtained through the GCM and high level methods, like ONIOM[QCISD(T)/6-31+G(2df,p):MP2/6-31G+G(d,p))//ONIOM[MP2/6-31+G(d,p):HF/6-31G]. The average error using the GCM, with respect to experimental data, was 3.4 kJ mol-1.

Métodos probabilístico e determinístico para diagnóstico da necessidade de subsolagem de solos agrícolas

This paper presents two techniques to evaluate soil mechanical resistance to penetration as an auxiliary method to help in a decision-making in subsoiling operations. The decision is based on the volume of soil mobilized as a function of the considered critical soil resistance to penetration in each case. The first method, probabilistic, uses statistical techniques to define the volume of soil to be mobilized. The other method, deterministic, determines the percentage of soil to be mobilized and its spatial distribution. Both cases plot the percentage curves of experimental data related to the soil mechanical resistance to penetration equal or larger to the established critical level and the volume of soil to be mobilized as a function of critical level. The deterministic method plots showed the spatial distribution of the data with resistance to penetration equal or large than the critical level. The comparison between mobilized soil curves as a function of critical level using both methods showed that they can be considered equivalent. The deterministic method has the advantage of showing the spatial distribution of the critical points.

Metodo de reconstrução tridimensional para avaliação postural

Universidade Estadual de Campinas. Faculdade de Educação Física