Differential diagnosis of Interstitial Lung Diseases using Deep Learning networks

ABSTRACT

An architecture for automatic lung tissue classification method based on the Deep Learning techniques is designed in this paper. Recent works on Deep Learning techniques achieved impressive results in the field of medical image classification. So, we designed a Convolution Neural Network (CNN) for the classification of five categories of Interstitial Lung Diseases (ILD) patterns in High-Resolution Computed Tomography (HRCT) images. The CNN consists of 3 Convolution layers, Leaky ReLU activation followed by Maximum pooling layer and dense layer. The last Fully Connected (FC) layer has 5 outputs equivalent to the classes considered such as Normal, Ground Glass (GG), Emphysema, Micro Nodules, and Fibrosis. The proposed CNN is trained and evaluated on the publicly available ILD database provided by the University Hospitals of Geneva (HUG). Experimental results are compared with the state-of-art, which shows an outstanding performance of the proposed CNN model giving 94.67% accuracy and 94.65% F_avg .     

A transfer of technology from engineering: use of ROC curves from signal detection theory to investigate information processing in the brain during sensory difference testing

This article reviews a beneficial effect of technology transfer from Electrical Engineering to Food Sensory Science. Specifically, it reviews the recent adoption in Food Sensory Science of the receiver operating characteristic (ROC) curve, a tool that is incorporated in the theory of signal detection. Its use allows the information processing that takes place in the brain during sensory difference testing to be studied and understood. The review deals with how Signal Detection Theory, also called Thurstonian modeling, led to the adoption of a more sophisticated way of analyzing the data from sensory difference tests, by introducing the signal-to-noise ratio, d', as a fundamental measure of perceived small sensory differences. Generally, the method of computation of d' is a simple matter for some of the better known difference tests like the triangle, duo-trio and 2-AFC. However, there are occasions when these tests are not appropriate and other tests like the same-different and the A Not-A test are more suitable. Yet, for these, it is necessary to understand how the brain processes information during the test before d' can be computed. It is for this task that the ROC curve has a particular use.     

Metalorganic chemical vapor deposition of Cu films from bis(t-butyl-3-oxo-butanoato)copper(II): thermodynamic investigation and experimental verification

For the metalorganic chemical vapor deposition (MOCVD) of copper films using the β-diketonate complex copper t-butylacetoacetate or bis(t-butyl-3-oxo-butanoato)copper(II) as the precursor material, the equilibrium concentrations of various condensed and gaseous phases have been thermodynamically calculated, using the criterion of minimization of total Gibbs free energy of the system. The study predicts the deposition of films of metallic copper without the presence of any oxides, even in an inert atmosphere of argon. The Cu films so formed would contain a small amount of carbon, which reduces to zero at higher temperatures. In the reactive ambient of hydrogen, carbon-free copper film is predicted over a wide range of substrate temperatures and total reactor pressures. The thermodynamic yield shows reasonable agreement with experimental observations, though the removal of carbon from the MOCVD-grown copper films in hydrogen is not as complete as the thermodynamic calculations predict.     

Fuzzy Logic Based Condition Rating of Existing Reinforced Concrete Bridges

Fuzzy logic is a means for modeling the uncertainty involved in describing an event/result using natural language. The fuzzy logic approach would be particularly useful for remedying the uncertainties and imprecision in bridge inspectors’ observations. This study explores the possibilities of using fuzzy mathematics for condition assessment and rating of bridges, developing a systematic procedure and formulations for rating existing bridges using fuzzy mathematics. Computer programs developed from formulations presented in this paper are used for evaluating the rating of existing bridges, and the details are presented in the paper. In this approach, the entire bridge has been divided into three major components—deck, superstructure, and substructure—each of which is further subdivided into a number of elements. Using fuzzy mathematics in combination with an eigenvector-based priority setting approach, the resultant rating set for the bridge has been evaluated based on the specified ratings and importance factors for all the elements of the bridge. Then the defuzzified value of the resultant rating fuzzy set becomes the rating value for the bridge as a whole. It is argued that the methodology presented in this paper would help the decision makers/bridge inspectors immensely.     

Evaluation of micro-mechanical properties of nonlinear cementitious composite using large-displacement transient nanoindentation analysis

Nanoindentation technique is employed for evaluation of mechanical properties of homogeneous materials at micro-level. Many engineering materials, especially cement and concrete composites, which are extensively used as building materials, exhibit phase heterogeneity and are highly porous. The presence of pores highly influences the response obtained from nanoindentation tests. In this study, mechanical properties of Calcium-Silicate-Hydrate (C-S-H), the primary binding agent in cementitious composites, are investigated using a simulated nanoindentation technique. The influence of presence of pores and its geometrical distribution on the non-linear response of C-S-H phases and the stress distribution are critically analysed.     

Seismic performance evaluation of GLD beam–column sub-assemblages of three different scenarios

Gravity load designed (GLD) structures exhibit poor seismic performance due to inadequate reinforcement to cater for the seismic forces, lack of confinement, poor joint strength, improper anchorage, presence of weak column and strong beams. These structures are also susceptible to environmental impact leading to corrosion of reinforcement which would further degrade their seismic performance. In RC structures, beam–column joints are crucial members and dissipate the seismic energy imparted to the structure. Hence in this study, two typical cases of GLD (i) uncorroded GLD and (ii) corrosion affected GLD (corroded) exterior beam–column sub-assemblages subjected to reverse cyclic loading are investigated and compared. Reinforcement corrosion of one of the GLD specimens is accelerated by the impressed current technique. Based on the investigation on uncorroded GLD and corrosion affected GLD beam–column specimens, it is noted that the corrosion has significant effect on the seismic performance in terms of loss of energy dissipation, strength- and stiffness- degradation. To improve the seismic performance of uncorroded GLD specimens, a low invasive single steel haunch upgradation scheme (SHUS) is proposed in this study. The maximum load carried by steel haunch upgraded GLD specimen is about 30% higher than that of GLD specimen. Hence, it is noted that by implementing the proposed upgradation scheme for GLD beam–column sub-assemblage, remarkable improvement in the seismic response can be achieved. The present study provides insight into the behaviour of steel haunch upgraded GLD specimen and would also pave the way for formulating seismic upgradation of even corroded GLD beam–column sub-assemblage.     

Flame sprayed tri-metallic Pt–Sn–X/Al2O3 catalysts (X = Ce,Zn, and K) for propane dehydration

Trimetallic nanocrystalline Pt–Sn–X/Al₂O₃ catalysts (X = Ce, Zn, and K) consisting of 0.3 wt.% Pt, 1 wt.% Sn, and 0.5 wt.% X have been prepared by one-step flame spray pyrolysis (FSP). As shown by the X-ray diffraction (XRD) and the transmission electron microscopy (TEM) results, the as-synthesized FSP-made catalysts were consisted of single-crystalline γ-alumina particles with average primary particle sizes 8 to 9 nm. The N₂ physisorption results revealed that all the catalysts contained only the macropore structure. The catalytic properties of the FSP-made catalysts were investigated in the dehydration of propane. Addition of Ce during FSP synthesis resulted in higher Pt dispersion as well as improved catalytic activity and stability than the non-promoted Pt–Sn/Al₂O₃. An opposite trend was found with the ones doped with Zn and K in which high surface coverage of Zn and K resulted in a significant loss of Pt active sites. The mechanism for the formation of the trimetallic nanoparticles during one-step FSP synthesis appeared to depend strongly on the differences in the vapor pressure of the metals and the alumina support in flame.     

Thermodynamic investigation of the MOCVD of copper films from<Emphasis Type="Italic">bis</Emphasis>(2,2,6,6-tetramethyl-3,5-heptadionato)copper(II)

Equilibrium concentrations of various condensed and gaseous phases have been thermodynamically calculated, using the free energy minimization criterion, for the metalorganic chemical vapour deposition (MOCVD) of copper films usingbis(2,2,6,6-tetramethyl-3,5-heptadionato)copper(II) as the precursor material. From among the many chemical species that may possibly result from the CVD process, only those expected on the basis of mass spectrometric analysis and chemical reasoning to be present at equilibrium, under different CVD conditions, are used in the thermodynamic calculations. The study predicts the deposition of pure, carbon-free copper in the inert atmosphere of argon as well as in the reactive hydrogen atmosphere, over a wide range of substrate temperatures and total reactor pressures. Thin films of copper, grown on SiO₂/Si(100) substrates from this metalorganic precursor by low pressure CVD have been characterized by XRD and AES. The experimentally determined composition of CVD-grown copper films is in reasonable agreement with that predicted by thermodynamic analysis.