Abbas Ali - Academia.edu (original) (raw)

Papers by Abbas Ali

Since the outbreak of Coronavirus Disease 2019 (COVID-19), most of the impacted patients have bee... more Since the outbreak of Coronavirus Disease 2019 (COVID-19), most of the impacted patients have been diagnosed with high fever, dry cough, and soar throat leading to severe pneumonia. Hence, to date, the diagnosis of COVID-19 from lung imaging is proved to be a major evidence for early diagnosis of the disease. Although nucleic acid detection using real-time reverse-transcriptase polymerase chain reaction (rRT-PCR) remains a gold standard for the detection of COVID-19, the proposed approach focuses on the automated diagnosis and prognosis of the disease from a non-contrast chest computed tomography (CT) scan for timely diagnosis and triage of the patient. The prognosis covers the quantification and assessment of the disease to help hospitals with the management and planning of crucial resources, such as medical staff, ventilators and intensive care units (ICUs) capacity. The approach utilises deep learning techniques for automated quantification of the severity of COVID-19 disease via measuring the area of multiple rounded ground-glass opacities (GGO) and consolidations in the periphery (CP) of the lungs and accumulating them to form a severity score. The severity of the disease can be correlated with the medicines prescribed during the triage to assess the effectiveness of the treatment. The proposed approach shows promising results where the classification model achieved 93% accuracy on hold-out data.

PRICAI 2019: Trends in Artificial Intelligence, 2019

In the last few years, we have witnessed a resurgence of interest in neural networks. The state-o... more In the last few years, we have witnessed a resurgence of interest in neural networks. The state-of-the-art deep neural network architectures are however challenging to design from scratch and requiring computationally costly empirical evaluations. Hence, there has been a lot of research effort dedicated to effective utilisation and adaptation of previously proposed architectures either by using transfer learning or by modifying the original architecture. The ultimate goal of designing a network architecture is to achieve the best possible accuracy for a given task or group of related tasks. Although there have been some efforts to automate network architecture design process, most of the existing solutions are still very computationally intensive. This work presents a framework to automatically find a good set of hyper-parameters resulting in reasonably good accuracy, which at the same time is less computationally expensive than the existing approaches. The idea presented here is to frame the hyper-parameter selection and tuning within the reinforcement learning regime. Thus, the parameters of a meta-learner, RNN, and hyper-parameters of the target network are tuned simultaneously. Our meta-learner is being updated using policy network and simultaneously generates a tuple of hyper-parameters which are utilized by another network. The network is trained on a given task for a number of steps and produces validation accuracy whose delta is used as reward. The reward along with the state of the network, comprising statistics of network's final layer outcome and training loss, are fed back to the meta-learner which in turn generates a tuned tuple of hyper-parameters for the next time-step. Therefore, the effectiveness of a recommended tuple can be tested very quickly rather than waiting for the network to converge. This approach produces accuracy close to the state-of-the-art approach and is found to be comparatively less computationally intensive.

2021 International Joint Conference on Neural Networks (IJCNN), 2021

Proceedings of the Conference of Language and …, 2009

Urdu language processing applications encounter non-Urdu text specifically English text frequentl... more Urdu language processing applications encounter non-Urdu text specifically English text frequently. The accuracy of these systems eg machine translation, text-to-speech etc. is highly undermined as they are unable to handle English text. One possibility could be addition ...

Evolutionary Intelligence

The first COVID-19 confirmed case was reported in Wuhan, China, and spread across the globe with ... more The first COVID-19 confirmed case was reported in Wuhan, China, and spread across the globe with an unprecedented impact on humanity. Since this pandemic requires pervasive diagnosis, developing smart, fast, and efficient detection techniques is significant. To this end, we have developed an Artificial Intelligence engine to classify the lung inflammation level (mild, progressive, severe stage) of the COVID-19 confirmed patient. In particular, the developed model consists of two phases; in the first phase, we calculate the volume and density of lesions and opacities of the CT scan images of the confirmed COVID-19 patient using Morphological approaches. The second phase classifies the pneumonia level of the confirmed COVID-19 patient. We use a modified Convolution Neural Network (CNN) and k-Nearest Neighbor; we also compared the results of both models to the other classification algorithms to precisely classify lung inflammation. The experiments show that the CNN model can provide testing accuracy up to 95.65% compared with exiting classification techniques. The proposed system in this work can be applied efficiently to CT scan and X-ray image datasets. Also, in this work, the Transfer Learning technique has been used to train the pre-trained modified CNN model on a smaller dataset than the original dataset; the modified CNN achieved 92.80% of testing accuracy for detecting pneumonia on chest X-ray images for the relatively extensive dataset.

2020 International Joint Conference on Neural Networks (IJCNN)

This paper presents the design and development of multi-dialect automatic speech recognition for ... more This paper presents the design and development of multi-dialect automatic speech recognition for Arabic. Deep neural networks are becoming an effective tool to solve sequential data problems, particularly, adopting an end-to-end training of the system. Arabic speech recognition is a complex task because of the existence of multiple dialects, non-availability of large corpora, and missing vocalization. Thus, the first contribution of this work is the development of a large multi-dialectal corpus with either full or at least partially vocalized transcription. Additionally, the open-source corpus has been gathered from multiple sources that bring non-standard Arabic alphabets in transcription which are normalized by defining a common character-set. The second contribution is the development of a framework to train an acoustic model achieving state-of-the-art performance. The network architecture comprises of a combination of convolutional and recurrent layers. The spectrogram features of the audio data are extracted in the frequency vs time domain and fed in the network. The output frames, produced by the recurrent model, are further trained to align the audio features with its corresponding transcription sequences. The sequence alignment is performed using a beam search decoder with a tetra-gram language model. The proposed system achieved a 14% error rate which outperforms previous systems.

arXiv (Cornell University), Jan 23, 2022

Large Language Models (LLMs) have ushered in a major paradigm shift in Natural Language Processin... more Large Language Models (LLMs) have ushered in a major paradigm shift in Natural Language Processing (NLP), where large pre-trained Language models (LMs) have become a fundamental component of most NLP tasks. These models are intelligent enough to find relevant and meaningful representations of a language without any supervision. They are used to finetune typical NLP tasks with substantially higher precision than conventional shallow learning techniques. However, training these models requires a massively large corpus that adequately represents a language. Due to the availability of enormous corpora, English LLMs typically perform better than their counterparts. This effort focuses on the design and development of a large Arabic corpus. The corpus comprises over 500 GB of Arabic cleaned text, intended to improve cross-domain knowledge and downstream generalization capability of LLMs. The corpus was employed in the training of a large Arabic LLM. In order to assess the efficacy of the LLM, a variety of typical NLP tasks were fine-tuned. The fine-tuned tasks exhibited a significant boost in accuracy ranging between 4.5 and 8.5%, when compared to those downstreamed from multilingual BERT (mBERT). To the best of our knowledge, this is currently the largest clean and diverse Arabic corpus ever assembled.

Indian Journal of Pure & Applied Physics, 2017

The natural and artificial radioactivity in soil samples from some oil sites of Kirkuk-Iraq have ... more The natural and artificial radioactivity in soil samples from some oil sites of Kirkuk-Iraq have been estimated using a gamma spectrometry based on a high purity germanium (HPGe) detector. For this reason, soil samples have been collected from four sites; Henjera, Jabel Boor, Jambor and Qutan. It was found that the specific activity ranged from 7.31 to 63.33 Bq kg −1 for 226 Ra, from 3.54 to 42.95 Bq kg −1 for 232 Th, from 103.21 to 798.52 Bq kg −1 for 40 K and from 0.7 to 9.53 Bq kg −1 for 137 Cs. The results have been compared with the worldwide average values. The radium equivalent activity ( Ra eq ), the absorbed gamma dose rate ( D ), the annual effective dose rate (AEDE), the external hazard ( H ex ), the internal hazard ( H in ) and Gamma radiation representative level Index ( I γ ) have also been calculated. The Ra eq was 92.173 Bq kg −1 , the D was 45.53 nGyh -1 , the AEDE outdoor and indoor were 0.0959 and 0.224, respectively, the H ex was 0.242, the H in was 0.329, and I ...

Asian journal of management sciences & education, 2013

The paper sets to examine India’s educational management on the presumption that management shoul... more The paper sets to examine India’s educational management on the presumption that management should give enough value for money or adequate returns to investment. Literacy is a good enough test of educational management, universal literacy as a commonly accepted goal of education. Having failed to get it by about 25 points, Indian educational management appears to give not enough returns to investment.

Iraqi Geological Journal, 2021

This study is based on agricultural soil samples which are collected along three traverses (A, B,... more This study is based on agricultural soil samples which are collected along three traverses (A, B, C), near the gypsum quarries located near the village of Bajwan north of Kirkuk city, to conduct a geochemical analysis and determining some heavy elements (Co, Cr, Ni, Cu, Sr, Pb, Cd) levels in the surface and subsurface soil horizons, and to indicate the potential sources of contamination with these elements. Accordingly, 30 samples were collected (six samples from travers A, five samples from travers B, and four samples from travers C) from the soil for each of the surface and subsurface levels. The results showed that the average concentrations of most studied elements increased in the subsurface soil compared to the surface soil, as a result of the influence of different geological and environmental conditions on the distribution of these elements in different soil horizons. The concentrations of the studied elements (Co, Ni, Cd) are more than their natural concentrations when comp...

Applied Sciences, 2018

Decision making is a cognitive process for evaluating data with certain attributes to come up wit... more Decision making is a cognitive process for evaluating data with certain attributes to come up with the best option, in terms of the preferences of decision makers. Conflicts and disagreements occur in most real world problems and involve the applications of mathematical tools dealing with uncertainty, such as rough set theory in decision making and conflict analysis processes. Afterwards, the Pawlak conflict analysis model based on rough set theory was established. Subsequently, Deja put forward some questions that are not answered by the Pawlak conflict analysis model and Sun’s model. In the present paper, using the notions of soft preference relation, soft dominance relation, and their roughness, we analyzed the Middle East conflict and answered the questions posed by Deja in a good manner.

Proceedings of International Structural Engineering and Construction, 2014

The objective of the work is to investigate the structural behavior of reinforced-concrete flange... more The objective of the work is to investigate the structural behavior of reinforced-concrete flanged deep beams with web openings, particularly in regards to the web reinforcement effect. This paper reports on experimental work on five test specimens, with square openings 25% of web depth at mid-height, through a critical shear path, subjected to repeated loading on simply-supported T-beams with two-point loads. Results showed that increasing the ratio of reinforcement of web increased the cracking and ultimate load, while increasing the vertical and horizontal reinforcement of web gives better results. Thus an inclined reinforcing of web gives higher ultimate load.

Journal of Geoscience and Environment Protection, 2018

Heavy metals (i.e. Cr, Co, Ni, Cu, Zn, Rb, Sr, Ba, Pb, V and Ga) distribution and their correlati... more Heavy metals (i.e. Cr, Co, Ni, Cu, Zn, Rb, Sr, Ba, Pb, V and Ga) distribution and their correlation with clay fraction were investigated. Fifteen samples of stream sediments were collected from the Lesser Zab River (LZR), which represent one of three major tributaries of the Tigris River at northeastern Iraq. Grain size distributions and textural composition indicate that these sediments are mainly characterized as clayey silt and silty sand. This indicates that the fluctuation in the relative variation of the grain size distribution in the studied sediments is due local contrast in the hydrological conditions, such as stream speed, energy of transportation and geological, geomorphological and climatic characterizations that influenced sediments properties. On the other hand, clay mineral assemblages consist of palygorskite, kaolinite, illite, chlorite and smectite, which in turn reveals that these sediments were derived from rocks of similar mineralogical and chemical composition as it is coincided with other published works. The clay mineral assemblages demonstrate that major phase transformations were not observed except for the palygorskite formation from smectite, since the minerals pair exhibit good negative correlation (−0.598) within the Lesser Zab River (LZR) sediments. To determine interrelation between the heavy metals and the clay fractions in the studied samples, correlation coefficients and factor analysis were performed. Heavy metals provide significant positive correlation with themselves and with Al 2 O 3 , Fe 2 O 3 and MnO. In addition, the results of factor analysis extracted two major factors; the first factor loading with the highest percent of variation (60%) from the major (Fe 2 O 3 , Al 2 O 3 and MnO in weight %), heavy metals and clay fraction. While the second factor with the (14%) of variance includes Cr and silt fraction, which indicate the affinity of the heavy metals How to cite this paper:

Religions: A Scholarly Journal, 2015

Perceptions on of work and its relation to human life have occupied man for centuries and discuss... more Perceptions on of work and its relation to human life have occupied man for centuries and discussions about the concept of work among different traditions and cultures appear to be under influence of a certain ideology or particular religion. In this article, Professor Ali has analyzed the understanding of work ethics based on the Islamic perspective. However, as an introduction to this theme, he firstly elaborated the concept of work ethics in a historical context, drawing from Greek civilization up to our time. The idea of work ethics and its understanding among the key makers of Greco-Roman and later Judeo-Christian civilizations up to the industrial revolution and technological age of our times went through the profound changes. In contrast to the Islamic perspective of work ethics, the author claims that in the Western civilization the essential purpose of work despite all drastic comprehension of this notion remains of profitable nature. In Islam on the other hand, the questio...

Journal of Remote Sensing & GIS, 2015

There are more than 119 million mines were buried in 71 countries in the world. The number of min... more There are more than 119 million mines were buried in 71 countries in the world. The number of mine victims is greater than the number of the victims of nuclear and chemical weapons together. Egypt is one of the countries that suffer from the presence of landmines in its soil. Hence, around 21 million landmines are found in several locations, especially at El-Alameen and Sinai Peninsula.

Journal of Agriculture and Rural Development in the Tropics and Subtropics

This study was conducted in 2010 in Eastern Nuba Mountains, Sudan to investigate ethnobotanical f... more This study was conducted in 2010 in Eastern Nuba Mountains, Sudan to investigate ethnobotanical food and nonfood uses of 16 wild edible fruit producing trees. Quantitative and qualitative information was collected from 105 individuals distributed in 7 villages using a semi-structured questionnaire. Also gathering of data was done using a number of rapid rural appraisal techniques, including key informant interviews, group discussion, secondary data sources and direct observations. Data was analysed using fidelity level and informant consensus factor methods to reveal the cultural importance of species and use category. Utilizations for timber products were found of most community importance than food usages, especially during cultivated food abundance. Balanites aegyptiaca, Ziziphus spina-christi and Tamarindus indica fruits were asserted as most preferable over the others and of high marketability in most of the study sites. Harvesting for timber-based utilizations in addition to agricultural expansion and overgrazing were the principal threats to wild edible food producing trees in the area. The on and off prevailing armed conflict in the area make it crucial to conserve wild food trees which usually play a more significant role in securing food supply during emergency times, especially in times of famine and wars. Increasing the awareness of population on importance of wild food trees and securing alternative income sources, other than wood products, is necessary in any rural development programme aiming at securing food and sustaining its resources in the area.

This study was undertaken at the Research Farm of KP Agricultural University, Peshawar during 201... more This study was undertaken at the Research Farm of KP Agricultural University, Peshawar during 2012 to evaluate half-sib maize families for grain yield and agronomic traits. Forty nine half-sib families derived from maize variety Sarhad White were used in the study. The experiment was laid out in 7×7 partial balance lattice design with two replications. Results indicated that the half-sib families were significantly different from one another for all the traits studied. Among the 49 half-sib families minimum days to 50% tasseling (50.0) were recorded for HS-32 while HS-48 showed maximum days to 50% tasseling (60.0). Minimum days to 50% silking (51.0) were observed for HS-08 while HS-48 showed maximum days to mid tasseling. Minimum plant height of 110.50 cm was recorded for HS-41 while HS-08 showed maximum plant height (160.60 cm), minimum ear height (40.70 cm) was recorded for HS-41 and maximum ear height (79.60 cm) for HS-48. Minimum grain moisture (14.0%) was recorded for HS-03 whi...

Journal of Islamic Banking and Finance, 2014

The aim of this paper is to study is to review the performance of Asset based (Mudharabah) and co... more The aim of this paper is to study is to review the performance of Asset based (Mudharabah) and commodity based (Murabaha) Islamic securities on a yearly basis in the Islamic Inter-Bank Money Market (IIMM) in Malaysia. The money market is a key appendage of the banking system and banking system is considered as very important for the development of any economy. Banks depend on the money market to manage their liquidity. While liquidity management is not the only use of money markets for banks, it is by far the most important. Recent statistics shows that the Malaysian Islamic banking sector continues to outperform the conventional banking sector with average annual asset growth rate of 18.6% during 2008-2012, in comparison to the conventional banking growth of 9.3% during the same period. The paper will also highlight the kinds of risk for Asset and commodity based Islamic securities in IIMM which are likely to be face by participant of the market.

Journal of Medical Sciences, 2003