Using Imbalanced Triangle Synthetic Data for Machine Learning Anomaly Detection

The extreme imbalanced data problem is the core issue in anomaly detection.The amount of abnormal data is so small that we cannot get adequate information to analyze it.The mainstream methods focus on taking fully advantages of the normal data,of which the discrimination method is that the data not belonging to normal data distribution is the anomaly.From the view of data science,we concentrate on the abnormal data and generate artificial abnormal samples by machine learning method.In this kind of technologies,Synthetic Minority Over-sampling Technique and its improved algorithms are representative milestones,which generate synthetic examples randomly in selected line segments.In our work,we break the limitation of line segment and propose an Imbalanced Triangle Synthetic Data method.In theory,our method covers a wider range.In experiment with real world data,our method performs better than the SMOTE and its meliorations.

Four-hour evaluation of a medical food in subjects with type 2 diabetes receiving oral hypoglycemic medication

Background: Postprandial hyperglycemia is an independent risk factor for diabetes-associated complications in individuals with type 2 diabetes. Dietary modification plays an important role in glycemic control. This study was to examine the efficacy of a diabetes specific formula (DSF) during a 4-hour postprandial meal tolerance test (MTT) in Russian subjects with type 2 diabetes receiving oral hypoglycemic medication. Methods: In a randomized, cross-over design, 168 eligible subjects from 11 study centers consumed, in a random order, the DSF (230 mL) or a common light hospital breakfast (i.e. standard meal) on two different occasions. The amounts of macronutrients were similar between the two meals providing ~200 kcal, 11 g protein, 26 g carbohydrate and 8 g fat. Capillary glucose levels were measured at baseline (before meal consumption), and post-meal consumption at 30, 60, 90, 120, 180 and 240 min. Results: The DSF was well tolerated in all subjects. There were 111 subjects completed the study per protocol (mean ± SEM: age: 58.6 ± 0.8 yr, BMI: 31.8 ± 0.42 kg/m2, waist circumference: 101 ± 1.3 cm, HbA1c: 8.0% ± 0.1%). Glucose levels reached peak values at 60 min (median) and the lowest levels at the end of the 240-min MTT test. The mean positive area under the curve (PAUC), the primary outcome, was significantly smaller after DSF consumption (mean ± SEM: 183.02 ± 18.74, median: 132.55) than the PAUC after consumption of the standard meal (mean ± SEM: 239.95 ± 23.11, median: 166.89;p = 0.027). The actual and adjusted peak glucose concentrations were similar between the two treatments. Conclusions: In patients with type 2 diabetes receiving oral hypoglycemic agents, compared to a hospital meal, the DSF improves postprandial glucose control. Combining results from earlier studies, long-term use of DSF may be beneficial to improve glucose management and decrease diabetes-associated complications.

Misconceptions and truths for feeding patients in the intensive care unit: Case studies with practical nursing solutions

The Guidelines for the Provision and Assessment of Nutrition Support Therapy in the Adult Critically Ill Patient published in May 2009 (joint guidelines between the Society of Critical Care Medicine (SCCM) and the American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) have advanced our clinical practice for the nutritional management of critically ill patients. In the current article, we will review how to implement these guidelines using a case study model. Two mechanically ventilated and tube fed patients are discussed, one with pneumonia and the second with severe acute pancreatitis. We address the questions of the feeding timing, method of administration, and management of its complications for these two patients.