Cardiovascular disease (CVD) is the leading cause of death worldwide, and coronary artery disease (CAD) is a major contributor. Early-stage CAD can progress if undiagnosed and left untreated, leading to myocardial infarction (MI) that may induce irreversible heart muscle damage, resulting in heart chamber remodeling and eventual congestive heart failure (CHF). Electrocardiography (ECG) signals can be useful to detect established MI, and may also be helpful for early diagnosis of CAD. For the latter especially, the ECG perturbations can be subtle and potentially misclassified during manual interpretation and/or when analyzed by traditional algorithms found in ECG instrumentation. For automated diagnostic systems (ADS), deep 2 learning techniques are favored over conventional machine learning techniques, due to the automatic feature extraction and selection processes involved. This paper highlights various deep learning algorithms exploited for the classification of ECG signals into CAD, MI, and CHF conditions. The Convolutional Neural Network (CNN), followed by combined CNN and Long Short-Term Memory (LSTM) models, appear to be the most useful architectures for classification. A 16-layer LSTM model was developed in our study and validated using 10-fold cross-validation. A classification accuracy of 98.5% was achieved. Our proposed model has the potential to be a useful diagnostic tool in hospitals for the classification of abnormal ECG signals.

No abstract

No abstract

No abstract

ARSTRAff-Experimental and theoretical studies on the spontaneous ignition proccu of isolated fuel droplets were carried o u t Tiedependent ttmperature fields around the igniting droplets wcrc observed by interferometry so that two step temperature rise can be detected. Some experiments a n pcdomed under microgravity to obtain reference data. Induction times arc examined as a function of ambient temperature As a result a zero temperature codficient region h found, whicb is equivalent to the NTC (negative tempraturc coefficient) region for the ignition of premixed A numerical model is developed applying a simplified chemical reaction model tbet includts 4 e low and the high tempraturc reactions The model is able to rcprodua the two step temperature rise and the roles of the two kinds of reactions on the ignition praccu up to the establishment of a difusion fiame around the droplet are examined.

No abstract

Intravascular ultrasound (IVUS) plays an important role for the detection of arteriosclerosis, which causes the ischemic heart disease. In mechanical scanning-type IVUS, it is necessary to rotate a transducer or a reflecting mirror. A method that involves rotating the transducer using a torque wire causes image distortion (NURD: non uniform rotation distortion). For a method that involves placing an electromagnetic motor on the tip of an IVUS probe is difficult to miniaturize the probe. Our objectives are to miniaturize the probe (1 mm in diameter, 5 mm in length) and to remove NURD. Therefore, we conducted a study to assess the feasibility of attaining these objectives by constructing a prototype IVUS system, in which an ultrasound motor using a stator in the form of a helical coil (abbreviated as CS-USM: coiled stator-ultrasonic motor) is incorporated, and to clarify problems that need to be solved in constructing the probe.

The growth process of electroless nickel-phosphorous (NiP) films on nonconducting substrate was investigated quantitatively using tapping mode atomic force microscopy (TMAFM) with focus upon the nucleation density on the substrate surface. Three kinds of catalyzing processes were used for polyimide substrates to obtain nucleation densities of 900, 750, and 550 nuclei/gum 1 . The TMAFM observation showed that the film growth proceeds mainly through successive nucleation of fine "particles" with several nanometers in diameter and their three-dimensional growth, and that the nucleation density influences the state of aggregation of the particles and hence the resulting surface topography. The lower nucleation density developed the rougher surface with larger "grains" with a diameter of ca. 80 nm. However, the effect of nucleation density decayed with film growth and eventually vanished at a thickness of ca. 2000 nm due to successive nucleation. For comparison, the growth process of NiP on NiP coated Al substrate, which is used for rigid magnetic disk applications and has a higher nucleation density of 3000 nuclei/gm 2 , was also examined. This process resulted in a smooth surface without the formation of "aggregates." Based upon these results, the root-mean-square roughness and scaling analyses were performed to evaluate quantitatively the differences in growth process.