Abstract
This paper presents a behavioral model for amorphous indium-gallium-zinc oxide thin-film transistor using artificial neural network (ANN) based equivalent circuit (EC) approach to predict static and dynamic behavior of the device. In addition, TFT parasitic capacitances (C-GS and C-GD) characterization through measurements is also reported. In the proposed model, an EC is derived from the device structure, in terms of electrical lumped elements. Each electrical element in the EC is modeled with an ANN. Then these ANNs are connected together as per the EC and implemented in Verilog-A. The proposed model performance is validated by comparing the circuit simulation results with the measured response of a simple common-source amplifier, which has shown 12.2 dB gain, 50 mu W power consumption and 85 kHz 3-dB frequency with a power supply of 6 V. The same circuit is tested as an inverter and its response is also presented up to 50 kHz, from both simulations and measurements. These results show that the model is capable of capturing both small and large signal behavior of the device to good accuracy, even including the harmonic distortion of the signal (that emphasizes the nonlinear behavior of the parasitic capacitance), making the model suitable for IC design.

Abstract
This paper presents the characterization of fundamental analog and digital circuits with a-IGZO TFTs from measurements performed at normal ambient. The fundamental blocks considered in this work include digital logic gates, a low power single stage high-gain amplifier with capcacitive bootstrapping and a level shifter/buffer. These circuits are important functional blocks in analog/Mixed signal IC design with oxide TFTs. Being fabricated at low temperature (<200 degrees C), they can find potential applications in low-cost large-area flexible systems.

Abstract
A novel linear analog adder is proposed only with n-type enhancement IGZO TFTs that computes summation of four voltage signals. However, this design can be easily extended to perform summation of higher number of signals, just by adding a single TFT for each additional signal in the input block. The circuit needs few number of transistors, only a single power supply irrespective of the number of voltage signals to be added, and offers good accuracy over a reasonable range of input values. The circuit was fabricated on glass substrate with the annealing temperature not exceeding 200 degrees C. The circuit performance is characterized from measurements under normal ambient at room temperature, with a power supply voltage of 12 V and a load of approximate to 4pF. The designed circuit has shown a linearity error of 2.3% (until input signal peak to peak value is 2 V), a power consumption of 78 mu W and a bandwidth of approximate to 115 kHz, under the worst case condition (when it is adding four signals with the same frequency). In this test setup, it has been noticed that the second harmonic is 32 dB below the fundamental frequency component. This circuit could offer an economic alternative to the conventional approaches, being an important contribution to increase the functionality of large area flexible electronics.

Abstract
In different areas of knowledge, phenomena are represented by directional-angular or periodic-data; from wind direction and geographical coordinates to time references like days of the week or months of the calendar. These values are usually represented in a linear scale, and restricted to a given range (e.g. [0,2 pi)), hiding the real nature of this information. Therefore, dealing with directional data requires special methods. So far, the design of classifiers for periodic variables adopts a generative approach based on the usage of the von Mises distribution or variants. Since for non-periodic variables state of the art approaches are based on non-generative methods, it is pertinent to investigate the suitability of other approaches for periodic variables. We propose a discriminative Directional Logistic Regression model able to deal with angular data, which does not make any assumption on the data distribution. Also, we study the expressiveness of this model for any number of features. Finally, we validate our model against the previously proposed directional naive Bayes approach and against a Support Vector Machine with a directional Radial Basis Function kernel with synthetic and real data obtaining competitive results.

Abstract
Nowadays, breast cancer has become the most common cancer amongst females. As long as breast is assumed to be a feminine symbol, any imposed deformation of surgical procedures can affect the patients' quality of life. However, using a planning tool which is based on parametric modeling, not only improves surgeons' skills in order to perform surgeries with better cosmetic outcomes, but also increases the interaction between surgeons and patients during the decision for necessary procedures. In the current research, a methodology of parametric modeling, called Free-Form Deformation (FFD) is studied. Finally, confirmed by a quantitative analysis, we proposed two simplified versions of FFD methodology to increase model similarity to input data and decrease required fitting time.

Abstract
In classification, when there is a disproportion in the number of observations in each class, the data is said to be class imbalance. Class imbalance is pervasive in real world applications of data classification and has been the focus of much research. The minority class contributes too little to the decision boundary because the learning process learns from each observation in isolation. In this paper, we discuss the application of learning pairwise rankers as a solution to class imbalance. We compare ranking models to alternatives from the literature.