Articles — Recent

The aim of this thesis is to investigate the ability of cardiovascular biomarkers calculated from peripheral pulse waveforms to estimate central properties of the cardiovascular system (e.g. aortic stiffness) using nonlinear one-dimensional (1-D) modelling of pulse wave propagation in the arterial network. To test these biomarkers, I have produced novel 1-D models of pulse wave propagation under normal and pathological conditions. In the first part of my thesis, I extended the modelling capabilities of the existing 1-D/0-D code to represent arterial blood flow under diabetes, hypertension, and combined diabetes and hypertension. Cardiac and vascular parameters of the 1-D model were tailored to best match data available in the literature to produce generalised hypertensive, diabetic, and combined diabetic and hypertensive population models. Using these models, I have shown that the pulse waveform at the finger is strongly affected by the aortic flow wave and the muscular artery stiffness and diameter. Furthermore the peak to peak time measured from the pulse waveform at the finger can identify hypertensive from diabetic patients. In the second part, I developed a new methodology for optimising the number of arterial segments in 1-D modelling required to simulate precisely the blood pressure and flow waveforms at an arbitrary arterial location. This is achieved by systematically lumping peripheral 1-D model branches into 0-D models that preserve the net resistance and total compliance of the original model. The methodology is important to simplify the computational domain while maintaining the precision of the numerical predictions — an important step to translate 1-D modelling to the clinic. This thesis provides novel computational tools of blood flow modelling and waveform analysis for the design, development and testing of pulse wave biomarkers. These tools may help bridge the gap between clinical and computational approaches.

Trabalho apresentado ao currículo da Disciplina de Programação de Computadores 2

Haroon Ahmed's CV

A quick explanation of why the Alexander polynomial is an invariant for knots.

This article proposes to obtain a statistical model of the daily peak electricity load of a household located in Austin-TX,USA. The Box-Jenkins methodology was followed to obtain the best fit for the time-series. Four models provided a good fit: ARIMA(0,1,2), ARIMA(1,1,2), SARIMA(0,1,2)(0,1,1) and SARIMA(1,1,2)(0,1,1). The model with the highest Akaike Information Criteria was the ARIMA(1,2,2). However, the model with the highest forecast accuracy was the SARIMA(1,1,2)(0,1,1), which obtained an RMSE of 0.296 and a MAPE Of 15.00.

EPR paradox as a result of non-force interaction nonlocal quantum objects This is a LaTeX template (version from 2016 Feb. 17) for preparing documents for All-Russian Scientific Conference of the Mathematical Modeling and Boundary Value Problems [Matem. Mod. Kraev. Zadachi, Samara, Russian Federation]. It was submitted by an author writing for the 10th All-Russian Scientific Conference with international participation (MMiKZ’16).

Imaduddin Khan's CV

Fluid Report

The goal of the Angry Birds AI competition is to build an intelligent agent which can complete the levels of the game better than human players. This task is very challenging, because humans have a good prediction about the physic world, while for computers it is hard to reason about an unknown environment. In this paper we describe our DualHEX AI agent, which is based on the AngryHEX agent of participants of the Angry Birds competition 2013. Our agent models the knowledge of the game by means of Answer Set Programming. In this project we improved the AngryHEX approach by extending the knowledge base of the domain. Our DualHEX agent plans a shot taking into consideration the current and the next bird. It compares the damage probability of both birds to discard targets that suits more the next bird features.