Hyperparameter Tuning and Pipeline Optimization via Grid Search Method and Tree-Based AutoML in Breast Cancer Prediction.

Automated machine learning (AutoML) has been recognized as a powerful tool to build a system that automates the design and optimizes the model selection machine learning (ML) pipelines. In this study, we present a tree-based pipeline optimization tool (TPOT) as a method for determining ML models with significant performance and less complex breast cancer diagnostic pipelines. Some features of pre-processors and ML models are defined as expression trees and optimal gene programming (GP) pipelines, a stochastic search system. Features of radiomics have been presented as a guide for the ML pipeline selection from the breast cancer data set based on TPOT. Breast cancer data were used in a comparative analysis of the TPOT-generated ML pipelines with the selected ML classifiers, optimized by a grid search approach. The principal component analysis (PCA) random forest (RF) classification was proven to be the most reliable pipeline with the lowest complexity. The TPOT model selection technique exceeded the performance of grid search (GS) optimization. The RF classifier showed an outstanding outcome amongst the models in combination with only two pre-processors, with a precision of 0.83. The grid search optimized for support vector machine (SVM) classifiers generated a difference of 12% in comparison, while the other two classifiers, naïve Bayes (NB) and artificial neural network-multilayer perceptron (ANN-MLP), generated a difference of almost 39%. The method's performance was based on sensitivity, specificity, accuracy, precision, and receiver operating curve (ROC) analysis.

Effectiveness of Post-Mortem Computed Tomography (PMCT) in Comparison with Conventional Autopsy: A Systematic Review.

BACKGROUND: With the advancement of technology, Computed Tomography (CT) scan imaging can be used to gain deeper insight into the cause of death. AIMS: The purpose of this study was to perform a systematic review of the efficacy of Post- Mortem Computed Tomography (PMCT) scan compared with the conventional autopsies gleaned from literature published in English between the year 2009 and 2016. METHODOLOGY: A literature search was conducted on three databases, namely PubMed, MEDLINE, and Scopus. A total of 387 articles were retrieved, but only 21 studies were accepted after meeting the review criteria. Data, such as the number of victims, the number of radiologists and forensic pathologists involved, causes of death, and additional and missed diagnoses in PMCT scans were tabulated and analysed by two independent reviewers. RESULTS: Compared with the conventional autopsy, the accuracy of PMCT scans in detecting injuries and causes of death was observed to range between 20% and 80%. The analysis also showed that PMCT had more advantages in detecting fractures, fluid in airways, gas in internal organs, major hemorrhages, fatty liver, stones, and bullet fragments. Despite its benefits, PMCT could also miss certain important lesions in a certain region such as cardiovascular injuries and minor vascular injuries. CONCLUSION: This systematic review suggests that PMCT can replace most of the conventional autopsies in specific cases and is also a good complementary tool in most cases.

Estimating the refractive index of oxygenated and deoxygenated hemoglobin using genetic algorithm - support vector regression model.

BACKGROUND AND OBJECTIVES: The refractive index of hemoglobin plays important role in hematology due to its strong correlation with the pathophysiology of different diseases. Measurement of the real part of the refractive index remains a challenge due to strong absorption of the hemoglobin especially at relevant high physiological concentrations. So far, only a few studies on direct measurement of refractive index have been reported and there are no firm agreements on the reported values of refractive index of hemoglobin due to measurement artifacts. In addition, it is time consuming, laborious and expensive to perform several experiments to obtain the refractive index of hemoglobin. In this work, we proposed a very rapid and accurate computational intelligent approach using Genetic Algorithm/Support Vector Regression models to estimate the real part of the refractive index for oxygenated and deoxygenated hemoglobin samples. METHODS: These models utilized experimental data of wavelengths and hemoglobin concentrations in building highly accurate Genetic Algorithm/Support Vector Regression model (GA-SVR). RESULTS: The developed methodology showed high accuracy as indicated by the low root mean square error values of 4.65 × 10-4 and 4.62 × 10-4 for oxygenated and deoxygenated hemoglobin, respectively. In addition, the models exhibited 99.85 and 99.84% correlation coefficients (r) for the oxygenated and deoxygenated hemoglobin, thus, validating the strong agreement between the predicted and the experimental results CONCLUSIONS: Due to the accuracy and relative simplicity of the proposed models, we envisage that these models would serve as important references for future studies on optical properties of blood.