Comparison of the predictive outcomes for anti-tuberculosis drug-induced hepatotoxicity by different machine learning techniques.

BACKGROUND: The study compared the predictive outcomes of artificial neural network, support vector machine and random forest on the occurrence of anti-tuberculosis drug-induced hepatotoxicity. METHODS: The clinical and genomic data of patients treated with anti-tuberculosis drugs at Taipei Medical University-Wanfang Hospital were used as training sets, and those at Taipei Medical University-Shuang Ho Hospital served as test sets. Features were selected through a univariate risk factor analysis and literature evaluation. The accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve were calculated to compare the traditional, genomic, and combined models of the three techniques. RESULTS: Nine models were created with 7 clinical factors and 4 genotypes. Artificial neural network with clinical and genomic factors exhibited the best performance, with an accuracy of 88.67%, a sensitivity of 80%, and a specificity of 90.4% for the test set. The area under the receiver operating characteristic curve of this best model reached 0.894 for training set and 0.898 for test set, which was significantly better than 0.801 for training set and 0.728 for test set by support vector machine and 0.724 for training set and 0.718 for test set by random forest. CONCLUSIONS: Artificial neural network with clinical and genomic data can become a clinical useful tool in predicting anti-tuberculosis drug-induced hepatotoxicity. The machine learning technique can be an innovation to predict and prevent adverse drug reaction.

Microfluidic device for real-time formulation of reagents and their subsequent encapsulation into double emulsions.

Emulsion drops are often employed as picoliter-sized containers to perform screening assays. These assays usually entail the formation of drops encompassing discrete objects such as cells or microparticles and reagents to study interactions between the different encapsulants. Drops are also used to screen influences of reagent concentrations on the final product. However, these latter assays are less frequently performed because it is difficult to change the reagent concentration over a wide range and with high precision within a single experiment. In this paper, we present a microfluidic double emulsion drop maker containing pneumatic valves that enable real-time formulation of different reagents using pulse width modulation and consequent encapsulation of the mixed solutions. This device can produce drops from reagent volumes as low as 10 µL with minimal sample loss, thereby enabling experiments that would be prohibitively expensive using drop generators that do not contain valves. We employ this device to monitor the kinetics of the cell-free synthesis of green fluorescent proteins inside double emulsions. To demonstrate the potential of this device for real-time formulation, we perform DNA titration experiments to test the influence of DNA concentration on the amount of green fluorescence protein produced in double emulsions by a coupled cell-free transcription / translation system.

Recombinant human thyrotropin before (131)I therapy in patients with nodular goitre: a meta-analysis of randomized controlled trials.

BACKGROUND: Recombinant human thyrotropin (rhTSH) can be used to enhance radioiodine therapy for shrinking multinodular goitre. The aim of this meta-analysis was to compare the effectiveness of rhTSH pretreatment and radioiodine therapy with that of radioiodine alone for treating benign nodular goitre. METHODS: The PubMed, EMBASE, Cochrane Library, Scopus and ClinicalTrials.gov databases were searched to identify studies published before September 2014. A meta-analysis was performed to calculate the pooled effect size using random-effects models. The primary outcome was the reduction in thyroid volume. Secondary outcomes included thyroid function, extent of tracheal compression, radioactive iodine uptake, incidence of hypothyroidism and other complications. RESULTS: Nine RCTs including 416 patients were selected. The reductions in thyroid volume were significantly greater in the rhTSH pretreatment groups than those in the radioiodine alone groups at 12 months (weighted mean difference: 14·42%; 95% CI: 4·51-24·34% in high-dose rhTSH vs radioiodine alone; weighted mean difference: 19·66%; 95% CI: 3·67-35·65% in low-dose rhTSH vs radioiodine alone). The incidence of hypothyroidism in the high-dose rhTSH groups was significantly higher than that in the radioiodine alone groups. No significant difference in the incidence of hypothyroidism occurred between the low-dose rhTSH groups and the radioiodine alone groups. CONCLUSIONS: The overall results indicated that using rhTSH before radioiodine therapy resulted in a greater thyroid volume reduction than radioiodine therapy alone. An increased incidence of hypothyroidism was observed in patients receiving high-dose rhTSH. Low-dose rhTSH before radioiodine therapy is more efficacious than radioiodine therapy alone for treating nontoxic benign thyroid nodules.