Prediction of Tuberculosis Using an Automated Machine Learning Platform for Models Trained on Synthetic Data.

High-quality medical data is critical to the development and implementation of machine learning (ML) algorithms in healthcare; however, security, and privacy concerns continue to limit access. We sought to determine the utility of "synthetic data" in training ML algorithms for the detection of tuberculosis (TB) from inflammatory biomarker profiles. A retrospective dataset (A) comprised of 278 patients was used to generate synthetic datasets (B, C, and D) for training models prior to secondary validation on a generalization dataset. ML models trained and validated on the Dataset A (real) demonstrated an accuracy of 90%, a sensitivity of 89% (95% CI, 83-94%), and a specificity of 100% (95% CI, 81-100%). Models trained using the optimal synthetic dataset B showed an accuracy of 91%, a sensitivity of 93% (95% CI, 87-96%), and a specificity of 77% (95% CI, 50-93%). Synthetic datasets C and D displayed diminished performance measures (respective accuracies of 71% and 54%). This pilot study highlights the promise of synthetic data as an expedited means for ML algorithm development.

Lyophilized oral sustained release polymeric nanoparticles of nateglinide.

The objective of this study is to formulate lyophilized oral sustained release polymeric nanoparticles of nateglinide in order to decrease dosing frequency, minimize side effects, and increase bioavailability. Nateglinide-loaded poly Ɛ-caprolactone nanoparticles were prepared by emulsion solvent evaporation with ultrasonication technique and subjected to various studies for characterization including scanning electron microscopy (SEM), Fourier transform infrared spectroscopy, photon correlation spectroscopy and evaluated for in vitro drug release and pharmacodynamic studies. The influence of increase in polymer concentration, ultrasonication time, and solvent evaporation rate on nanoparticle properties was investigated. The formulations were optimized based on the above characterization, and the formulation using 5% polymer, 3-min sonication time, and rota-evaporated was found to have the best drug entrapment efficiency of 64.09±4.27% and size of 310.40±11.42 nm. Based on SEM, nanoparticles were found to be spherical with a smooth surface. In vitro drug release data showed that nanoparticles sustained the nateglinide release for over 12 h compared to conventional tablets (Glinate 60 mg), and drug release was found to follow Fickian mechanism. In vivo studies showed that nanoparticles prolonged the antidiabetic activity of nateglinide in rats significantly (p≤0.05) compared to the conventional tablets (Glinate 60 mg) over a period of 12 h. Accelerated stability data indicated that there was minimal to no change in drug entrapment efficiency.