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.

Background: Vitamin E supplements containing tocotrienols are now being recommended for optimum health but its effects are scarcely known. The objective was to determine the effects of Tocotrienol Rich Fraction (TRF) supplementation on lipid profile and oxidative status in healthy older individuals at a dose of 160 mg/day for 6 months. Methods: Sixty-two subjects were recruited from two age groups: 35-49 years (n = 31) and above 50 years (n = 31), and randomly assigned to receive either TRF or placebo capsules for six months. Blood samples were obtained at 0, 3^rd and 6^th months. Results: HDL-cholesterol in the TRF-supplemented group was elevated after 6 months (p < 0.01). Protein carbonyl contents were markedly decreased (p < 0.001), whereas AGE levels were lowered in the > 50 year-old group (p < 0.05). Plasma levels of total vitamin E particularly tocopherols were significantly increased in the TRF-supplemented group after 3 months (p < 0.01). Plasma total tocotrienols were only increased in the > 50 year-old group after receiving 6 months of TRF supplementation. Changes in enzyme activities were only observed in the > 50 year-old group. SOD activity was decreased after 3 (p < 0.05) and 6 (p < 0.05) months of TRF supplementation whereas CAT activity was decreased after 3 (p < 0.01) and 6 (p < 0.05) months in the placebo group. GPx activity was increased at 6 months for both treatment and placebo groups (p < 0.05). Conclusion: The observed improvement of plasma cholesterol, AGE and antioxidant vitamin levels as well as the reduced protein damage may indicate a restoration of redox balance after TRF supplementation, particularly in individuals over 50 years of age.

High-dose synchrotron microbeam radiation therapy (MRT) has shown the potential to deliver improved outcomes over conventional broadbeam (BB) radiation therapy. To implement synchrotron MRT clinically for cancer treatment, it is necessary to undertake dose equivalence studies to identify MRT doses that give similar outcomes to BB treatments. To develop an in vitro approach to determine biological dose equivalence between MRT and BB using two different cell-based assays. The acute response of tumour and normal cell lines (EMT6.5, 4T1.2, NMuMG, EMT6.5ch, 4T1ch5, SaOS-2) to MRT (50–560 Gy) and BB (1.5–10 Gy) irradiation was investigated using clonogenic and real time cell impedance sensing (RT-CIS)/xCELLigence assays. MRT was performed using a lattice of 25 or 50 µm-wide planar, polychromatic kilovoltage X-ray microbeams with 200 µm peak separation. BB irradiations were performed using a Co⁶⁰ teletherapy unit or a synchrotron radiation source. BB doses that would generate biological responses similar to MRT were calculated by data interpolation and verified by clonogenic and RT-CIS assays. For a given cell line, MRT equivalent BB doses identified by RT-CIS/xCELLigence were similar to those identified by clonogenic assays. Dose equivalence between MRT and BB were verified in vitro in two cell lines; EMT6.5ch and SaOS-2 by clonogenic assays and RT-CIS/xCELLigence. We found for example, that BB doses of 3.4±0.1 Gy and 4.40±0.04 Gy were radiobiologically equivalent to a peak, microbeam dose of 112 Gy using clonogenic and RT-CIS assays respectively on EMT6.5ch cells. Our data provides the first determination of biological dose equivalence between BB and MRT modalities for different cell lines and identifies RT-CIS/xCELLigence assays as a suitable substitute for clonogenic assays. These results will be useful for the safe selection of MRT doses for future veterinary and clinical trials.

Hepatocellular carcinoma (HCC) is considered as a complex liver disease and ranked as the eighth-highest mortality rate with a prevalence of 2.4% in Malaysia. Magnetic resonance imaging (MRI) has been acknowledged for its advantages, a gold technique for diagnosing HCC, and yet the false-negative diagnosis from the examinations is inevitable. In this study, 30 MR images from patients diagnosed with HCC is used to evaluate the robustness of semi-automatic segmentation using the flood fill algorithm for quantitative features extraction. The relevant features were extracted from the segmented MR images of HCC. Four types of features extraction were used for this study, which are tumour intensity, shape feature, textural feature and wavelet feature. A total of 662 radiomic features were extracted from manual and semi-automatic segmentation and compared using intra-class relation coefficient (ICC). Radiomic features extracted using semi-automatic segmentation utilized flood filling algorithm from 3D-slicer had significantly higher reproducibility (average ICC = 0.952 ± 0.009, p < 0.05) compared with features extracted from manual segmentation (average ICC = 0.897 ± 0.011, p > 0.05). Moreover, features extracted from semi-automatic segmentation were more robust compared to manual segmentation. This study shows that semi-automatic segmentation from 3D-Slicer is a better alternative to the manual segmentation, as they can produce more robust and reproducible radiomic features.