RESUMEN
The examination of Alzheimer's disease (AD) using adaptive machine learning algorithms has unveiled promising findings. However, achieving substantial credibility in medical contexts necessitates a combination of notable accuracy, minimal processing time, and universality across diverse populations. Therefore, we have formulated a hybrid methodology in this study to classify AD by employing a brain MRI image dataset. We incorporated an averaging filter during preprocessing in the initial stage to reduce extraneous details. Subsequently, a combined strategy was utilized, involving principal component analysis (PCA) in conjunction with stepwise linear discriminant analysis (SWLDA), followed by an artificial neural network (ANN). SWLDA employs a combination of forward and backward recursion methods to choose a restricted set of features. The forward recursion identifies the most interconnected features based on partial Z-test values. Conversely, the backward recursion method eliminates the least correlated features from the same feature space. After the extraction and selection of features, an optimized artificial neural network (ANN) was utilized to differentiate the various classes of AD. To demonstrate the significance of this hybrid approach, we utilized publicly available brain MRI datasets using a 10-fold cross-validation strategy. The proposed method excelled over existing state-of-the-art systems, attaining weighted average recognition rates of 99.35% and 96.66%, respectively, across all the datasets.
RESUMEN
PURPOSE: The aim of this work was to evaluate the use of an angularly independent silicon detector (edgeless diodes) developed for dosimetry in megavoltage radiotherapy for Cyberknife in a phantom and for patient quality assurance (QA). METHOD: The characterization of the edgeless diodes has been performed on Cyberknife with fixed and IRIS collimators. The edgeless diode probes were tested in terms of basic QA parameters such as measurements of tissue-phantom ratio (TPR), output factor and off-axis ratio. The measurements were performed in both water and water-equivalent phantoms. In addition, three patient-specific plans have been delivered to a lung phantom with and without motion and dose measurements have been performed to verify the ability of the diodes to work as patient-specific QA devices. The data obtained by the edgeless diodes have been compared to PTW 60016, SN edge, PinPoint ionization chamber, Gafchromic EBT3 film, and treatment planning system (TPS). RESULTS: The TPR measurement performed by the edgeless diodes show agreement within 2.2% with data obtained with PTW 60016 diode for all the field sizes. Output factor agrees within 2.6% with that measured by SN EDGE diodes corrected for their field size dependence. The beam profiles' measurements of edgeless diodes match SN EDGE diodes with a measured full width half maximum (FWHM) within 2.3% and penumbra widths within 0.148 mm. Patient-specific QA measurements demonstrate an agreement within 4.72% in comparison with TPS. CONCLUSION: The edgeless diodes have been proved to be an excellent candidate for machine and patient QA for Cyberknife reproducing commercial dosimetry device measurements without need of angular dependence corrections. However, further investigation is required to evaluate the effect of their dose rate dependence on complex brain cancer dose verification.