Conventionally, workflows examining transcription regulation networks from gene expression data involve distinct analytical steps. There is a need for pipelines that unify data mining and inference deduction into a singular framework to enhance interpretation and hypotheses generation. We propose a workflow that merges network construction with gene expression data mining focusing on regulation processes in the context of transcription factor driven gene regulation. The pipeline implements pathway-based modularization of expression profiles into functional units to improve biological interpretation. The integrated workflow was implemented as a web application software (TransReguloNet) with functions that enable pathway visualization and comparison of transcription factor activity between sample conditions defined in the experimental design. The pipeline merges differential expression, network construction, pathway-based abstraction, clustering and visualization. The framework was applied in analysis of actual expression datasets related to lung, breast and prostrate cancer.

This study proposes a method to estimate elastic deformation using silhouettes obtained from multiple endoscopic images. Our method can estimate the intraoperative deformation of organs using a volumetric mesh model reconstructed from preoperative CT data. We use this elastic body silhouette information of elastic bodies not to model the shape but to estimate the local displacements. The model shape is updated to satisfy the silhouette constraint while preserving the shape as much as possible. The result of the experiments showed that the proposed methods could estimate the deformation with root mean square (RMS) errors of 5.0–10 mm.

A 1-Mbps, 315MHz OOK transceiver in 40-nm CMOS for body area networks is developed. Both a 38-pJ/bit carrier-frequency-free intermittent sampling receiver with -55dBm sensitivity and a 36-pJ/bit transmitter applied dual supply voltage scheme with -20dBm output power achieve the lowest energy in the published transceivers for wireless sensor networks.

The well-known Chvátal--Erdös theorem states that every graph $G$ of order at least three with $\alpha(G)\le\kappa(G)$ has a Hamiltonian cycle, where $\alpha(G)$ and $\kappa(G)$ are the independence number and the connectivity of $G$, respectively. Oberly and Sumner [J. Graph Theory, 3 (1979), pp. 351--356] have proved that every connected, locally connected claw-free graph of order at least three has a Hamiltonian cycle. We study the connection of these two theorems. For $x\in V(G)$, let $B(x)$ denote the subgraph of $G$ induced by the closed neighborhood of $x$. Then the theorem by Oberly and Sumner says that a connected graph $G$ of order at least three satisfying $\alpha(B(x))\le 2\le \kappa(B(x))$ for every vertex $x$ has a Hamiltonian cycle. The comparison of this theorem with the Chvátal--Erdös theorem leads us to suspect that the threshold 2 between $\alpha(B(x))$ and $\kappa(B(x))$ is not necessary. We say that $G$ satisfies the local Chvátal--Erdös condition if $\alpha(B(x))\le\kappa(B(x))$ holds for every vertex $x$ in $G$. The second author conjectured that if the order of a connected graph $G$ is at least three and satisfies the local Chvátal--Erdös condition, then $G$ has a Hamiltonian cycle. In this paper, we support this conjecture by proving that under this assumption, $G$ is $1$-tough and has a $2$-factor.

A design methodology for sub-0.5V crystal oscillators is shown to realize an all-sub-0.5V ultra low power RF transceiver for wireless sensor networks. To reduce the minimum operating voltage (VDDmin) of the crystal oscillator, both the optimization of the gate width of the CMOS inverter in the crystal oscillator and the reduction in gate length by CMOS technology scaling are required. In accordance with the developed design methodology, a 39MHz crystal oscillator is designed and fabricated in a 40nm CMOS. The measured power consumption is 4.1μW at 0.35V and 39MHz, and the power supply voltage is the lowest among the previously reported crystal oscillators.

This paper proposes a Voice Activity Detection (VAD) algorithm based on Conditional Random Fields (CRF) using multiple features.VAD is a technique used to distinguish between speech and non-speech in noisy environments and is an important component in many real-world speech applications.The posterior probability of output labels in the proposed method is directly modeled by the weighted sum of the feature functions.Effective features are automatically selected by estimating appropriate weight parameters to improve the accuracy of VAD.Experimental results on the CENSREC-1-C database revealed that the proposed approach can decrease error rates by using CRF.

The purpose of this study is to investigate a new technique for image-based cancer cell classification and provide a more quantitative and objective characterization method for a diagnosis, which currently relies on qualitative and empirical judgment of pathologists. For this, a new method for chromatin texture characterization employing a new feature, contour complexity, is proposed and evaluated using nuclear images obtained from paraffin-wax embedded sections of human breast cancer on slides. The proposed feature is calculated on the basis of a contour length of nucleus obtained by setting different threshold values of intensity for a grayscale image, and it is a quantitative measure of chromatin texture. An expectation-maximization (EM) algorithm-based segmentation and an effective initial parameter search method for EM are used for the automatic calculation of the feature. The results for breast cancer cell detection showed that the average contour complexity value for malignant cells (19.6plusmn4.1) is found to be significantly greater (p-6, Kolmogorov-Smirnov test) than that of benign cells (0.35plusmn0.17). By the comparison with the conventional fractal dimension approach, it is shown that the proposed feature is much more sensitive feature than the fractal dimension for the individual cancer cell detection.

In this paper, analysis of the dynamics of the ultrasonic motors is introduced. Though static characteristics have in the past been clarified, the dynamic characteristics were still unclear. For the purpose of realizing high speed control of ultrasonic motors, we analyzed the dynamic characteristics through measurement and numerical simulation. Among the different types of ultrasonic motors, we focused on bar-type ultrasonic motors and first measured the step response using a laser Doppler velocimeter (LDV). We measured changing the input parameters and showed the relationship between input parameters and delay, overshoot of the rotational speed of the rotor. Next, we established a mathematical model of the bar-type ultrasonic motor. In our model, we approximated the stator as a rigid disc and torsional spring. We set the discrete spring at the contact area between rotor and stator, and simulated considering stick-slip at the contact area. The results of numerical simulation matched that of the measurements qualitatively and we clarified the dynamics of the contact condition and friction force at the contact area.