Intelligent Systems Technology (United States)

Diabetes causes a lot of damage to the internal organs and body parts. To find out the primary symptoms or features to detect and prevent the disease from its severity, an intelligent healthcare system is needed to analyse the disease data to save the patients with proper medication beforehand. The modern healthcare system requires proper IT solutions to process the medical data in disease management for diagnosis and prevention. Data analytics in the healthcare field demands both medical expertise and IT expertise. The feature selection process selects the important features from the feature pool and is generally applied as a pre-processing step prior to extracting the interesting classification rules from the data. In the paper, a rough set theory-based genetic algorithm (GA) method is proposed to select optimal feature set (called reduct)/symptoms from the diabetes dataset to predict the disease efficiently. Benchmark disease datasets are collected from the UCI repository. The experimental result shows that the selected features are important to predict the diabetes by providing good classification accuracy on existing benchmark classifiers, which proves the efficiency of the method.

Visual Lifelogging is the process of keeping track of one's life through wearable cameras. The focus of this research is to automatically classify images, captured from a wearable camera, into indoor and outdoor scenes. The results of this classification may be used in several applications. For instance, one can quantify the time a person spends outdoors and indoors which may give insights about the psychology of the concerned person. We use transfer learning from two VGG convolutional neural networks (CNN), one that is pre-trained on the ImageNet data set and the other on the Places data set. We investigate two methods of combining features from the two pre-trained CNNs. We evaluate the performance on the new UBRug data set and the benchmark SUN397 data set and achieve accuracy rates of 98.24% and 97.06%, respectively. Features obtained from the ImageNet pretrained CNN turned out to be more effective than those obtained from the Places pre-trained CNN. Fusing the feature vectors obtained from these two CNNs is an effective way to improve the classification. In particular, the performance that we achieve on the SUN397 data set outperforms the state-of-the-art.

Background Urinary steroid metabolite profiling is an accurate reflection of adrenal and gonadal steroid output and metabolism in peripheral target cells of steroid action. Measurement of steroid metabolite excretion by gas chromatography-massspectrometry (GC–MS) is considered reference standard for biochemical diagnosis of steroidogenic disorders. However, performance of GC–MS analysis and interpretation of the resulting data requires significant expertise and age- and sex-specific reference ranges. Here we developed novel computational approaches for rapid interpretation of GC–MS data for diagnosis of inborn steroidogenic disorders Methods We analysed the urinary steroid metabolome by GC–MS in 829 healthy controls(302 neonates and infants, 167 children and 360 adults) and 118 untreated patients with genetically confirmed inborn disorders (21-hydroxylase deficiency,17-hydroxylase deficiency, POR deficiency, 11b-hydroxylase deficiency, 3b-HSD2 deficiency, 17b-HSD3 deficiency, 5a-reductase type 2 deficiency, cytochrome b5 deficiency). We calculated age-related normative values for established metabolite ratios representing distinct enzymatic functions. We developed a novel interpretable machine learning technique, Angle Learning Vector Quantisation (ALVQ), which looks at all possible metabolite ratios, computationally reduces these to the most relevant for discrimination, and differentiates disease states by comparison to a representative prototype. The method runs independent of sex and age information, units of measurement and method of urine collection. Results Conventional biochemical ratios had 100% sensitivity but only very poor specificity. By contrast, ALVQ predicted 'affected urine' vs 'healthy urine' with 100% sensitivity and 97% specificity. For our three most prevalent conditions(PORD, SRD5A2 and CYP21A2), the specific condition was identified correctly in 96% of cases. Conclusion We developed a novel Steroid Metabolomics approach to automatically diagnose inborn steroidogenic disorders with very high sensitivity and specificity, superior to current methods, and with high potential for implementation in routine clinical care.

Place recognition is one of the most challenging problems in computer vision, and has become a key part in mobile robotics and autonomous driving applications for performing loop closure in visual SLAM systems. Moreover, the difficulty of recognizing a revisited location increases with appearance changes caused, for instance, by weather or illumination variations, which hinders the long-term application of such algorithms in real environments. In this paper we present a convolutional neural network (CNN), trained for the first time with the purpose of recognizing revisited locations under severe appearance changes, which maps images to a low dimensional space where Euclidean distances represent place dissimilarity. In order for the network to learn the desired invariances, we train it with triplets of images selected from datasets which present a challenging variability in visual appearance. The triplets are selected in such way that two samples are from the same location and the third one is taken from a different place. We validate our system through extensive experimentation, where we demonstrate better performance than state-of-art algorithms in a number of popular datasets.

This report documents the program and the outcomes of Dagstuhl Seminar 11341 Learning in the context of very high dimensional data. The aim of the seminar was to bring together researchers who develop, investigate, or apply machine learning methods for very high dimensional data to advance this important field of research. The focus was be on broadly applicable methods and processing pipelines, which offer efficient solutions for high-dimensional data analysis appropriate for a wide range of application scenarios.

The infusion of engineering (CogE) methods and tools into traditional systems engineering processes and practices is becoming an important priority as the systems engineering community begins to tackle system of systems (SoS) problems. As one illustration, the military is interested in creating cognitively-inspired systems (as small as a PDA and as large as a weapon platform) that maximally exploit human potential while also accelerating both human supported and automated decision making. To date, engineering has been successfully applied to the planning and requirements definition phases of systems engineering (also known as front-end analysis) but has yet to be infused into the remaining phases of systems engineering. To span the full systems engineering lifecycle, it is important, first and foremost, to communicate the return-on-investment to the various stakeholders to get their buy in. The second question that needs to be answered is whether or not cognitive engineering is ready for primetime. This paper presents promising technical and management strategies to overcome the challenges in introducing engineering into system-of-systems engineering (SOSE). Specifically, it presents a representative set of engineering methods and tools to span the full SoS life-cycle as well as engineering awareness initiatives to penetrate both the DoD acquisition community and commercial industry.