In modern-day medicine, medical imaging has undergone immense advancements and can capture several biomedical images from patients. In the wake of this, to assist medical specialists, these images can be used and trained in an intelligent system in order to aid the determination of the different diseases that can be identified from analyzing these images. Classification plays an important role in this regard; it enhances the grouping of these images into categories of diseases and optimizes the next step of a computer-aided diagnosis system. The concept of classification in machine learning deals with the problem of identifying to which set of categories a new population belongs. When category membership is known, the classification is done on the basis of a training set of data containing observations. The goal of this paper is to perform a survey of classification algorithms for biomedical images. The paper then describes how these algorithms can be applied to a big data architecture by using the Spark framework. This paper further proposes the classification workflow based on the observed optimal algorithms, Support Vector Machine and Deep Learning as drawn from the literature. The algorithm for the feature extraction step during the classification process is presented and can be customized in all other steps of the proposed classification workflow.

The capacitance-voltage (C-V) characteristics of thin films of ZrO2 deposited by thermal metal-organic chemical vapour deposition (MOCVD) have been analyzed. The films were grown at three different temperatures (500, 550 and 600 ºC) and 1 mbar pressure from a novel monomeric zirconium amide-guanidinate complex [Zr(NEtMe)2(guanidinate)2]. The true capacitance was determined from measurements made at different frequencies in order to account for the series and shunt parasitic resistances during C-V measurements. Films grown at 500 and 550 ºC showed no hysteresis while those grown at 600 ºC exhibited a very small hysteresis window ≈0.16 V for O2 flow of 100 sccm and ≈0.19 V for 50 sccm O2 flow. A very small voltage shift is also obtained for the device under 10 hr voltage stress. These preliminary in-depth electrical results suggest that quality ZrO2 can be grown from the novel [Zr(NEtMe)2(guanidinate)2] complex precursor paving the way for their use as future gate dielectrics.

Thin Film Transistors (TFTs) are the active elements for future large area electronic applications, in which low cost, low temperature processes and optical transparency are required. Zinc oxide (ZnO) thin film transistors (TFTs) on SiO 2 /n+-Si substrate are fabricated with the channel thicknesses ranging from 20 nm to 60 nm. It is found that both the performance and gate bias stress related instabilities of the ZnO TFTs fabricated were influenced by the thickness of ZnO active channel layer. The effective mobility was found to improve with increasing ZnO thickness by up to an order in magnitude within the thickness range investigated (20-60 nm). However, thinner films were found to exhibit greater stability in threshold voltage and turn-on voltage shifts with respect to both positive and negative gate bias stress. It was also observed that both the turn on voltage (V on ) and the threshold voltage (V T ) decrease with increasing channel thickness. Moreover, the variations in subthreshold slope (S) with ZnO thickness as well as variations in V T and V on suggest a possible dependence of trap states in the ZnO on the ZnO thickness. This is further correlated by the dependence of V T and V on instabilities with gate bias stress.

The recent use of hybrid renewable energy systems (HRESs) is considered one of the most reliable ways to improve energy access to decentralized communities because of their techno-economic and environmental benefits. Many distant locales, such as camps in war-torn nations, lack basic necessities like power. This study proposes a remedy for power outages in these areas; by designing an HRES and a control system for monitoring, distributing, and managing the electrical power from sustainable energy sources to supply the load. Hence, providing affordable, reliable, and clean energy for all (Sustainable Development Goal 7). In this study, the feasibility and techno-economic performance of an HRES for a refugee camp was evaluated under load following (LF), cycle charging (CC), and predictive control strategy (PS). The optimization results revealed that the PS was the most suitable, as it had the lowest cost and was more eco-friendly and energy-efficient. The predictive control strategy had a 48-h foresight of the load demand and resource potential and hence could effectively manage the HRES. The total net present cost (NPC) for the electrification of this refugee camp was $3,809,822.54, and the cost of electricity generated for every kWh is $0.2018. Additionally, 991,240.32 kg of emissions can be avoided annually through the hybridization of the diesel generator under the PS.

No abstract