Predicting the onset of type 2 diabetes using wide and deep learning with electronic health records.

OBJECTIVE: Diabetes is responsible for considerable morbidity, healthcare utilisation and mortality in both developed and developing countries. Currently, methods of treating diabetes are inadequate and costly so prevention becomes an important step in reducing the burden of diabetes and its complications. Electronic health records (EHRs) for each individual or a population have become important tools in understanding developing trends of diseases. Using EHRs to predict the onset of diabetes could improve the quality and efficiency of medical care. In this paper, we apply a wide and deep learning model that combines the strength of a generalised linear model with various features and a deep feed-forward neural network to improve the prediction of the onset of type 2 diabetes mellitus (T2DM). MATERIALS AND METHODS: The proposed method was implemented by training various models into a logistic loss function using a stochastic gradient descent. We applied this model using public hospital record data provided by the Practice Fusion EHRs for the United States population. The dataset consists of de-identified electronic health records for 9948 patients, of which 1904 have been diagnosed with T2DM. Prediction of diabetes in 2012 was based on data obtained from previous years (2009-2011). The imbalance class of the model was handled by Synthetic Minority Oversampling Technique (SMOTE) for each cross-validation training fold to analyse the performance when synthetic examples for the minority class are created. We used SMOTE of 150 and 300 percent, in which 300 percent means that three new synthetic instances are created for each minority class instance. This results in the approximated diabetes:non-diabetes distributions in the training set of 1:2 and 1:1, respectively. RESULTS: Our final ensemble model not using SMOTE obtained an accuracy of 84.28%, area under the receiver operating characteristic curve (AUC) of 84.13%, sensitivity of 31.17% and specificity of 96.85%. Using SMOTE of 150 and 300 percent did not improve AUC (83.33% and 82.12%, respectively) but increased sensitivity (49.40% and 71.57%, respectively) with a moderate decrease in specificity (90.16% and 76.59%, respectively). DISCUSSION AND CONCLUSIONS: Our algorithm has further optimised the prediction of diabetes onset using a novel state-of-the-art machine learning algorithm: the wide and deep learning neural network architecture.

Antimalarial and cytotoxic activities of ethnopharmacologically selected medicinal plants from South Vietnam.

Malaria is a major global public health problem and the alarming spread of drug resistance and limited number of effective drugs now available underline how important it is to discover new antimalarial compounds. An ethnopharmacological investigation was undertaken of medicinal plants traditionally used to treat malaria in the South Vietnam. Forty-nine plants were identified, 228 extracts were prepared and tested for their in vitro activity against Plasmodium falciparum, and assessed for any cytotoxicity against the human cancer cell line HeLa and the embryonic lung MRC5 cell line. In a first screening at a concentration of 10 microg/ml, 92 extracts from 46 plants showed antiplasmodial activity (parasite growth inhibition >30%). The IC(50) values of the most active extracts were determined as well as their selectivity towards Plasmodium falciparum in comparison to their cytotoxic effects against the human cell lines. Six plants showed interesting antiplasmodial activity (IC(50) ranging from 0.4 to 8.6 microg/ml) with a good selectivity: two Menispermaceae, Arcangelisia flava (L.) Merr. and Fibraurea tinctoria Lour., and also Harrisonia perforata (Blanco) Merr. (Simaroubaceae), Irvingia malayana Oliv. ex Benn. (Irvingiaceae), Elaeocarpus kontumensis Gagn. (Elaeocarpaceae) and Anneslea fragrans Wall. (Theaceae).

A new species of Eustigma (Hamamelidaceae) from Hon Ba Nature Reserve, Vietnam.

A new species of Hamamelidaceae, Eustigma honbaense H.Toyama, Tagane & V.S.Dang, sp. nov., is described from Hon Ba Nature Reserve, Vietnam. This species is similar to Eustigma oblongifolium Gardner & Champ., but differs from it in having entire leaves, longer infructescences, capsules with a longer apical part and seeds with a larger hilum. A description, preliminary conservation assessment, illustration and photographs of the new species are provided, as well as an updated key to the genus Eustigma.

Goniothalamusflagellistylus Tagane & V. S. Dang (Annonaceae), a new species from Mt. Hon Ba, Vietnam.

A new species, Goniothalamusflagellistylus Tagane & V. S. Dang, sp. nov. from Hon Ba Nature Reserve in Khanh Hoa Province, South Vietnam is described and illustrated. This species is most similar to Goniothalamustortilipetalus M.R.Hend., but distinct in having 308-336 stamens (vs. ca. 170-260) and ca.120 carpels (vs. ca. 50-100) per flower, and Stigma and pseudostyles ca.8.5 mm (vs. 4-4.5 mm) long.

Aporosa tetragona Tagane & V. S. Dang (Phyllanthaceae), a new species from Mt. Hon Ba, Vietnam.

A new species, Aporosa tetragona Tagane & V. S. Dang, sp. nov., is described and illustrated from Mt. Hon Ba located in the Khanh Hoa Province, South Vietnam. This species is characterized by tetragonal pistillate flowers and fruits, which are clearly distinguishable from the other previously known species of the genus. The morphology and phylogeny based on rbcL and matK of this species indicated that the new species belongs to section Appendiculatae Pax & K. Hoffm.