An empirical evaluation of the Random Forests classifier models for variable selection in a large-scale lung cancer case-control study

Random Forests™ is reported to be one of the most accurate classification algorithms in complex data analysis. It shows excellent performance even when most predictors are noisy and the number of variables is much larger than the number of observations. In this thesis Random Forests was applied to a large-scale lung cancer case-control study. A novel way of automatically selecting prognostic factors was proposed. Also, synthetic positive control was used to validate Random Forests method. Throughout this study we showed that Random Forests can deal with large number of weak input variables without overfitting. It can account for non-additive interactions between these input variables. Random Forests can also be used for variable selection without being adversely affected by collinearities. ^ Random Forests can deal with the large-scale data sets without rigorous data preprocessing. It has robust variable importance ranking measure. Proposed is a novel variable selection method in context of Random Forests that uses the data noise level as the cut-off value to determine the subset of the important predictors. This new approach enhanced the ability of the Random Forests algorithm to automatically identify important predictors for complex data. The cut-off value can also be adjusted based on the results of the synthetic positive control experiments. ^ When the data set had high variables to observations ratio, Random Forests complemented the established logistic regression. This study suggested that Random Forests is recommended for such high dimensionality data. One can use Random Forests to select the important variables and then use logistic regression or Random Forests itself to estimate the effect size of the predictors and to classify new observations. ^ We also found that the mean decrease of accuracy is a more reliable variable ranking measurement than mean decrease of Gini. ^

An interactive intervention to impact adherence to antiretroviral treatment among treatment-naive HIV-infected adults in Thailand: A pilot randomized controlled study

Background. Consistent adherence to antiretroviral treatment is necessary for a treatment success. Improving and maintaining adherence rate >95% are challenging for health care professionals. This pilot randomized controlled study aimed to evaluate the impact of the interactive intervention on adherence to GPO-VIR, to describe the feasibility of the interactive intervention in Thailand, and to illustrate the adherence self-efficacy concept among HIV treatment-naïve patients in Thailand who were starting antiretroviral treatment. ^ Methods. The study took place at three HIV clinics located in Phayao, Thailand. Twenty-three patients were randomly assigned into the experimental (n=11) and the control groups (n=12). Each participant in the experimental group and a significant person to the patient received 5 educational sessions with a nurse at the clinics and at their homes. They also received 3 follow-up evaluations during the 6-month period of the study. The participants in the control group received the standard of care provided by HIV clinical personnel plus three follow-up evaluations at the clinic. ^ Results. Seventeen patients (7 in the experimental and 10 in the control group) completed the study. The 4-day recall on the Thai ACTG Adherence Scale demonstrated adherence rate >95% for most participants from both groups. After the first measurement, no experimental group patients reporting missing ART, while one control group participant continuously skipped ART. Participants from both groups had significantly increased CD4 cell counts after the study (F(1, 15) = 29.30, p = .000), but no differences were found between two groups (F(1, 15) = .001, p = .98). Examination of the intervention showed limitations and possibilities to implement it in Thailand. Qualitative data demonstrated self-efficacy expectations, resignation and acceptance as related concepts to improve adherence outcomes. ^ Conclusions. This interactive intervention, after appropriate modifications, is feasible to apply for Thai HIV-treatment naïve patients. Because of limitations the study could not demonstrate whether the interactive intervention improved adherence to ART among HIV-treatment naïve in Thailand. A longitudinal study in a larger sample would be required to test the impact of the intervention. ^ Keyword: antiretroviral treatment, adherence, treatment-naïve, Thailand, randomized controlled study ^