Search results for: document classifier

Authors: Andrey V. Timofeev

Abstract:

The paper presents new results concerning selection of optimal information fusion formula for ensembles of C-OTDR channels. The goal of information fusion is to create an integral classificator designed for effective classification of seismoacoustic target events. The LPBoost (LP-β and LP-B variants), the Multiple Kernel Learning, and Weighing of Inversely as Lipschitz Constants (WILC) approaches were compared. The WILC is a brand new approach to optimal fusion of Lipschitz Classifiers Ensembles. Results of practical usage are presented.

Keywords: Lipschitz Classifier, classifiers ensembles, LPBoost, C-OTDR systems

Procedia PDF Downloads 435

Authors: Aayushi Pandey, Manoj Kumar Pandey, Ashutosh Tiwari, Kireet Kumar

Abstract:

Glaciers play an important role in climate change and are sensitive phenomena of global climate change scenario. Glaciers in Himalayas are unique as they are predominantly valley type and are located in tropical, high altitude regions. These glaciers are often covered with debris which greatly affects ablation rate of glaciers and work as a sensitive indicator of glacier health. The aim of this study is to map high altitude Glacier surface with a focus on glacial lake and debris estimation using different techniques in Nagling glacier of dhauliganga basin in Himalayan region. Different Image Classification techniques i.e. thresholding on different band ratios and supervised classification using maximum likelihood classifier (MLC) have been used on high resolution sentinel 2A level 1c satellite imagery of 14 October 2017.Here Near Infrared (NIR)/Shortwave Infrared (SWIR) ratio image was used to extract the glaciated classes (Snow, Ice, Ice Mixed Debris) from other non-glaciated terrain classes. SWIR/BLUE Ratio Image was used to map valley rock and Debris while Green/NIR ratio image was found most suitable for mapping Glacial Lake. Accuracy assessment was performed using high resolution (3 meters) Planetscope Imagery using 60 stratified random points. The overall accuracy of MLC was 85 % while the accuracy of Band Ratios was 96.66 %. According to Band Ratio technique total areal extent of glaciated classes (Snow, Ice ,IMD) in Nagling glacier was 10.70 km2 nearly 38.07% of study area comprising of 30.87 % Snow covered area, 3.93% Ice and 3.27 % IMD covered area. Non-glaciated classes (vegetation, glacial lake, debris and valley rock) covered 61.93 % of the total area out of which valley rock is dominant with 33.83% coverage followed by debris covering 27.7 % of the area in nagling glacier. Glacial lake and Debris were accurately mapped using Band ratio technique Hence, Band Ratio approach appears to be useful for the mapping of debris covered glacier in Himalayan Region.

Keywords: band ratio, Dhauliganga basin, glacier mapping, Himalayan region, maximum likelihood classifier (MLC), Sentinel-2 satellite image

Procedia PDF Downloads 203

Authors: Carlos Arias-Alcaide, Cristina Soguero-Ruiz, Paloma Santos-Álvarez, Adrián García-Romero, Inmaculada Mora-Jiménez

Abstract:

Analysing clinical data with computers in such a way that have an impact on the practitioners’ workflow is a challenge nowadays. This paper provides a first approach for monitoring the health status of COVID-19 patients through the use of some biomarkers (blood tests) and the simplest Naïve Bayes classifier. Data of two Spanish hospitals were considered, showing the potential of our approach to estimate reasonable posterior probabilities even some days before the event.

Keywords: Bayesian model, blood biomarkers, classification, health tracing, machine learning, posterior probability

Procedia PDF Downloads 187

Authors: Syahrul Bariah Abdul Hamid, Colin W. Binns, Jun Hui Chih

Abstract:

The total of obese mothers is growing throughout Asia. Breastfeeding provides the perfect nutrition for infants, by promoting a higher IQ and protecting against childhood and adult diseases. A prospective cohort study was carried out of mothers attending eight antenatal clinics run by the Ministry of Health in Selangor, Malaysia to document the prevalence of obesity and its relationship with breastfeeding outcomes. Mothers were enrolled during the antenatal period and followed up until 6 months postpartum to document breastfeeding outcomes. A total of 652 Malay mothers were recruited for the study a response rate of 93.1 %. The pre-pregnancy body mass index (BMI) of the mothers showed that 36.5% of the mothers were overweight or obese. There were a total of 78 obese mothers in the sample and 41 (52.6%) of these mothers were able to initiate breastfeeding within one hour of birth compared to 238/337 (70.6 %, χ² 9.35, p<0.001) of those with a normal BMI. At 6 months, 23.1 % of obese mothers were exclusively breastfeeding their infants, compared to 56.0 % of the normal BMI mothers. On the other hand, the rate of infant formula feeding was higher in the obese mothers by 53.8 % compared to 19.0 % among normal weight mothers, χ² 37.6, p<0.001). Further analysis suggested these factors were found to be positively associated with discontinued exclusive breastfeeding at 6 months among obese mothers; mothers whom delayed breastfeeding initiation, had health problems during pregnancy, caesarean delivery, reported had insufficient colostrum/milk and babies had sucking problems at or before 4 weeks. Besides that, mothers who perceived their biological mothers had preference towards formula feeding or were ambivalent about the feeding method and had biological mothers without experience in breastfeeding for more than 1 month also were more likely to discontinue exclusive breastfeeding at 6 months. These findings suggested that the greater the pre-pregnant BMI, the earlier the cessation of exclusive breastfeeding and they were also less likely to initiate breastfeeding and have less adequate milk supply. Future investigations of the effects of maternal obesity on breastfeeding outcomes should be conducted along with effective interventions to advance the care of obese women at reproductive age and their children.

Keywords: exclusive breastfeeding, body mass index (BMI), breastfeeding discontinuation, maternal obesity

Procedia PDF Downloads 171

Authors: Sofia Matoug, Amr Abdel-Dayem

Abstract:

This paper presents a comprehensive survey of recent research studies to segment and classify brain MR (magnetic resonance) images in order to detect significant changes to brain ventricles. The paper also presents a general framework for detecting regions that atrophy, which can help neurologists in detecting and staging Alzheimer. Furthermore, a prototype was implemented to segment brain MR images in order to extract the region of interest (ROI) and then, a classifier was employed to differentiate between normal and abnormal brain tissues. Experimental results show that the proposed scheme can provide a reliable second opinion that neurologists can benefit from.

Keywords: Alzheimer, brain images, classification techniques, Magnetic Resonance Images MRI

Procedia PDF Downloads 278

Authors: Masahiro Ohmura, Koh Kakusho, Takeshi Okadome

Abstract:

This paper presents a regression model with autocorrelated errors in which the inputs are social moods obtained by analyzing the adjectives in Twitter posts using a document topic model. The regression model predicts Dow Jones Industrial Average (DJIA) more precisely than autoregressive moving-average models.

Keywords: stock market prediction, social moods, regression model, DJIA

Procedia PDF Downloads 522

Authors: Muhammed Shahriar Haque

Abstract:

As an offshoot of journalistic discourse, the documentary should be objective in nature without harbouring any preconceived notion to foster ulterior motifs. When it comes to a social issue like rape in South Asian countries, as media in recent times is inundated with this violent act in India, Pakistan, Myanmar, Bangladesh, how does one document it in terms of objectivity and subjectivity? The objective of this study is twofold: to document the history of documentaries, and to critically analyze South Asian rape/flesh trade-documentaries. The overall goal is to trace the (re/de)construction of objectivity-subjectivity in documentaries. This paper adopts a qualitative approach to documentarist discourse through the lens of critical discourse analysis (CDA). Data was gathered for 10 documentaries on the theme of rape and/or flesh trade from eight South Asian countries, predominantly the South Asian Association of Regional Cooperation (SAARC) region. The documentaries were primarily categorised by using three frameworks based on six modes, six subgenres, and four basic approaches of documentary. Subsequently, the findings were critiqued from CDA perspective. The outcome suggests that there a two schools of thoughts regarding documentaries. According to journalistic ethics, news and/or documentaries should be objective in orientation and focus on informing the audience and/common people. The empirical findings tend to challenge ethical parameters of objectivity. At times, it seems that journalistic discourse is discursively (re)constructed to give an augmented simulation of objectivity. Based on the findings it may be recommended that if documentaries steer away from empirical facts and indulge in poetic naivety, their credibility could be questioned. A research of this nature is significant as it raises questions with regard to ethical and moral conscience of documentary filmmakers. Furthermore, it looks at whether they uphold journalistic integrity or succumb to their bias, and thereby depict subjective views, which could be tainted with political and/or propagandist ulterior motifs.

Keywords: discursive (re/de)construction, documentaries, journalistic integrity, rape/flesh trade

Procedia PDF Downloads 127

Authors: Petronella Jonck, Freda van der Walt

Abstract:

Despite the benefits of innovation diffusion in the South African public service, implementation thereof seems to be problematic, particularly with regard to e-governance which would enhance the quality of service delivery, especially accessibility, choice, and mode of operation. This paper reports on differences between the public service and the private sector in terms of innovation diffusion. Innovation diffusion will be investigated to explore identified obstacles that are hindering successful implementation of e-governance. The research inquiry is underpinned by the diffusion of innovation theory, which is premised on the assumption that innovation has a distinct channel, time, and mode of adoption within the organisation. A comparative thematic document analysis was conducted to investigate organisational differences with regard to innovation diffusion. A similar approach has been followed in other countries, where the same conceptual framework has been used to guide document analysis in studies in both the private and the public sectors. As per the recommended conceptual framework, three organisational characteristics were emphasised, namely the external characteristics of the organisation, the organisational structure, and the inherent characteristics of the leadership. The results indicated that the main difference in the external characteristics lies in the focus and the clientele of the private sector. With regard to organisational structure, private organisations have veto power, which is not the case in the public service. Regarding leadership, similarities were observed in social and environmental responsibility and employees’ attitudes towards immediate supervision. Differences identified included risk taking, the adequacy of leadership development, organisational approaches to motivation and involvement in decision making, and leadership style. Due to the organisational differences observed, it is recommended that differentiated strategies be employed to ensure effective innovation diffusion, and ultimately e-governance. It is recommended that the results of this research be used to stimulate discussion on ways to improve collaboration between the mentioned sectors, to capitalise on the benefits of each sector.

Keywords: E-governance, ICT, innovation diffusion, comparative analysis

Procedia PDF Downloads 328

Authors: Ahmed Abdullah Ahmed

Abstract:

The task of recognizing the writer of a handwritten text has been an attractive research problem in the document analysis and recognition community with applications in handwriting forensics, paleography, document examination and handwriting recognition. This research presents an automatic method for writer recognition from digitized images of unconstrained writings. Although a great effort has been made by previous studies to come out with various methods, their performances, especially in terms of accuracy, are fallen short, and room for improvements is still wide open. The proposed technique employs optimal codebook based writer characterization where each writing sample is represented by a set of features computed from two codebooks, beginning and ending. Unlike most of the classical codebook based approaches which segment the writing into graphemes, this study is based on fragmenting a particular area of writing which are beginning and ending strokes. The proposed method starting with contour detection to extract significant information from the handwriting and the curve fragmentation is then employed to categorize the handwriting into Beginning and Ending zones into small fragments. The similar fragments of beginning strokes are grouped together to create Beginning cluster, and similarly, the ending strokes are grouped to create the ending cluster. These two clusters lead to the development of two codebooks (beginning and ending) by choosing the center of every similar fragments group. Writings under study are then represented by computing the probability of occurrence of codebook patterns. The probability distribution is used to characterize each writer. Two writings are then compared by computing distances between their respective probability distribution. The evaluations carried out on ICFHR standard dataset of 206 writers using Beginning and Ending codebooks separately. Finally, the Ending codebook achieved the highest identification rate of 98.23%, which is the best result so far on ICFHR dataset.

Keywords: off-line text-independent writer identification, feature extraction, codebook, fragments

Procedia PDF Downloads 487

Authors: D. Touil, L. Ouadah

Abstract:

An industrial model of cement grinding circuit is proposed on the basis of sampling surveys undertaken in the Meftah cement plant in Algiers, Algeria. The ball mill is described by a series of equal fully mixed stages that incorporates the effect of air sweeping. The kinetic parameters of this material in the energy normalized form obtained using the data of batch dry ball milling are taken into account in developing the present scale-up procedure. The dynamic separator is represented by the air classifier selectivity equation corrected by empirical factors. The model is incorporated in computer program that predict full size distributions and mass flow rates for all streams in a circuit under a particular set of operating conditions.

Keywords: grinding circuit, clinker, cement, modeling, population balance, energy

Procedia PDF Downloads 503

Authors: Srushti Patil, Preethi Baligar, Gopalkrishna Joshi, Gururaj N. Bhadri

Abstract:

In designing learning environments, the instructional strategies can be tailored to suit the learning style of an individual to ensure effective learning. In this study, the information shared on social media like Facebook is being used to predict learning style of a learner. Previous research studies have shown that Facebook data can be used to predict user personality. Users with a particular personality exhibit an inherent pattern in their digital footprint on Facebook. The proposed work aims to correlate the user's’ personality, predicted from Facebook data to the learning styles, predicted through questionnaires. For Millennial learners, Facebook has become a primary means for information sharing and interaction with peers. Thus, it can serve as a rich bed for research and direct the design of learning environments. The authors have conducted this study in an undergraduate freshman engineering course. Data from 320 freshmen Facebook users was collected. The same users also participated in the learning style and personality prediction survey. The Kolb’s Learning style questionnaires and Big 5 personality Inventory were adopted for the survey. The users have agreed to participate in this research and have signed individual consent forms. A specific page was created on Facebook to collect user data like personal details, status updates, comments, demographic characteristics and egocentric network parameters. This data was captured by an application created using Python program. The data captured from Facebook was subjected to text analysis process using the Linguistic Inquiry and Word Count dictionary. An analysis of the data collected from the questionnaires performed reveals individual student personality and learning style. The results obtained from analysis of Facebook, learning style and personality data were then fed into an automatic classifier that was trained by using the data mining techniques like Rule-based classifiers and Decision trees. This helps to predict the user personality and learning styles by analysing the common patterns. Rule-based classifiers applied for text analysis helps to categorize Facebook data into positive, negative and neutral. There were totally two models trained, one to predict the personality from Facebook data; another one to predict the learning styles from the personalities. The results show that the classifier model has high accuracy which makes the proposed method to be a reliable one for predicting the user personality and learning styles.

Keywords: educational data mining, Facebook, learning styles, personality traits

Procedia PDF Downloads 200

Authors: Touhami Tahenni

Abstract:

Excessively wide cracks are harmful to the serviceability of reinforced concrete (RC) beams and may lead to durability problems in the longer term. They also reduce the rigidity of RC sections, rendering the tensile concrete ineffective structurally. To reduce the negative effects of cracks, steel fibers are added to concrete mixes in the same manner as aggregates. In the present work, steel fibers reinforced concrete (SFRC) beams, made of normal strength and high strength concretes, were tested in a four-point bending test using a digital image correlation technique. The beams had different volume fractions of fibres and different aspect ratios (fiber length/fiber diameter). The evaluation of flexural cracking widths was determined using Gom-Aramis software. The experimental crack widths were compared with theoretical values predicted by the technical document of Rilem TC 162-TDF. The model proposed in this document seems to be the only one that considers the efficiency of steel fibres in restraining the crack widths. However, the model of Rilem takes into account only the aspect ratio of steel fibres to predict the crack width of SFRC beams. It has been reported in several pieces of research that the contribution of steel fibres to the limitation of flexural cracking widths is based on three essential parameters namely, the volume fraction, the orientation and the aspect ratio of fibres. Referring to the literature on the flexural cracking behavior of SFRC beams and the experimental observations of the present work, a correction of the Rilem model by the introduction of these parameters in the formula is proposed. The crack widths predicted by the new empirical model were compared with the experimental results and assessed against other test data on SFRC beams taken from the literature. The modified Rilem model gives better results and is found more satisfactory in predicting the crack widths of fibres concrete.

Keywords: stee fibres, reinforced concrete, flexural cracking, tensile strength, crack width

Procedia PDF Downloads 62

Authors: Suzan Wedyan

Abstract:

Data mining is one of the main phases in the Knowledge Discovery Database (KDD) which is responsible of finding hidden and useful knowledge from databases. There are many different tasks for data mining including regression, pattern recognition, clustering, classification, and association rule. In recent years a promising data mining approach called associative classification (AC) has been proposed, AC integrates classification and association rule discovery to build classification models (classifiers). This paper surveys and critically compares several AC algorithms with reference of the different procedures are used in each algorithm, such as rule learning, rule sorting, rule pruning, classifier building, and class allocation for test cases.

Keywords: associative classification, classification, data mining, learning, rule ranking, rule pruning, prediction

Procedia PDF Downloads 508

Authors: Haouam Djemaa

Abstract:

This study aimed to identify the possibility for applying fair value accounting by Algerian enterprises coted in capital maket (Algiers stock exchange). To achieve the objectives of this study, we made an interview with preparers of accounting information. The results document that enterprises are aware of fair value accounting in financial reporting because of its ability to provide useful accounting, but it depends on the availability of favorable circumstances for its application and this is what is missing in the Algerian environment.

Keywords: fair value, financial reporting, accounting information, valuation method

Procedia PDF Downloads 373

Authors: Km Tanu

Abstract:

It has been widely accepted that the learning style of adults and children is different, the learning motivation among adults vary, and even their learning preferences cannot be predetermined. In India, where the population is widely diverse and socio-economic and cultural disparities are there, the learning strategies should also be according to their needs and preferences. The present study explores the concept of adult learners in India in order to understand their needs and styles better. The adult learning principles of Malcolm Knowles have been analyzed, and its presence in the different policies and programs has been traced. To what extent these principles and other such concepts would be beneficial for the Indian population and for effective learning strategies, and what contextual understanding is needed, has been argued in the study. Descriptive research methodology, along with content and thematic analyses, has been used for the paper. It has been argued that there are four areas that play crucial roles in making learning effective. These are the learner, the facilitator, the resources and the policy. The prior experiences of the learners, their motivation, the group to which they belong (i.e., the learning styles and the strategies can be varied for the group of farmers and migrant laborers), and their expected outcome play an important role in making any adult education program successful but along with this, the role of facilitator or the educator is also very important as it is not easy to deal with the adult learners, the understanding that the task is not to teach the adult learners but to make them learn and to use their prior knowledge is a task in itself, proper training is needed for that matter. Many times, it has been seen that adult education programs are poorly funded, or even if they are funded, the fund is not utilized well; the unavailability of the resources is one of the reasons for the failure of adult education programs, and if we see these four points as a triangle, at the bottom, there is a policy document. A well-stated and described doable policy document is also equally important.

Keywords: adult education, Indian adult learner, effective learning styles, Malcolm Knowles learning principles, adult education policies and program

Procedia PDF Downloads 37

Authors: Lucas A. Silva, Carla R. Aguiar

Abstract:

This work presents a proposal to perform contextual sentiment analysis using a supervised learning algorithm and disregarding the extensive training of annotators. To achieve this goal, a web platform was developed to perform the entire procedure outlined in this paper. The main contribution of the pipeline described in this article is to simplify and automate the annotation process through a system of analysis of congruence between the notes. This ensured satisfactory results even without using specialized annotators in the context of the research, avoiding the generation of biased training data for the classifiers. For this, a case study was conducted in a blog of entrepreneurship. The experimental results were consistent with the literature related annotation using formalized process with experts.

Keywords: sentiment analysis, untrained annotators, naive bayes, entrepreneurship, contextualized classifier

Procedia PDF Downloads 368

Authors: Fortune Mangara

Abstract:

The world is undergoing the largest wave of urban growth in history due to rapid urbanization. Africa’s fast rate of urbanization is being driven by several factors such as population growth and migration. Urbanization results in development pressure on existing infrastructure, and numerous existing buildings are being destroyed in the process. Many of these buildings are built by environmental heritage resources which are part of the city's heritage and are therefore valuable. Many built environment heritage resources are currently being destroyed due to development pressure, while others are facing the risk of destruction or abandonment. There are different approaches that inform urban development and urban conservation. The modernist and post-modernist dichotomy has played an influencing role on how development or conservation of built environment heritage resources are approached. The fragmented nature of historical urban conservation paradigms and theories are also reflected in the evolution of policy and legislation that guide urban development and conservation of built heritage resources. Urban development and conservation have a long history of being guided by separated policies and legislation. However, recent international and South African policy and legislation had started to acknowledge the importance of integrating urban development and urban conservation. Spatial planning guides urban development and can be used as an integrative tool. With the aforementioned in mind, the main research question that guides this study is: What role does spatial planning play in the coexistence of urban development and urban conservation in a city in transition? The main purpose of this research is to use spatial planning as a tool for integrating urban conservation and urban development with reference to built environmental heritage resources. A qualitative research methodology is going to be employed in which a singular case study will be used as the research design. A qualitative document analysis will be used to collect data. Potchefstroom is going to be used as a case study as it is the oldest town in the North West province therefore is rich in built environmental heritage resources.

Keywords: built environmental heritage resources, document analysis, spatial planning, urban conservation, urban development

Procedia PDF Downloads 107

Authors: Jheanel E. Estrada, Larry A. Vea

Abstract:

This study developed model to detect proper/improper sitting posture using the built in web camera which detects the upper body points’ location and distances (chin, manubrium and acromion process). It also established relationships of human body frames and proper sitting posture. The models were developed by training some well-known classifiers such as KNN, SVM, MLP, and Decision Tree using the data collected from 60 students of different body frames. Decision Tree classifier demonstrated the most promising model performance with an accuracy of 95.35% and a kappa of 0.907 for head and shoulder posture. Results also showed that there were relationships between body frame and posture through Body Mass Index.

Keywords: posture, spinal points, gyroscope, image processing, ergonomics

Procedia PDF Downloads 305

Authors: Riona Dighe

Abstract:

This research attempts to identify an alternative approach to the measurement of Gross National Happiness (GNH). It uses artificial intelligence (AI), incorporating natural language processing (NLP) and sentiment analysis to measure GNH. We use ‘off the shelf’ NLP models responsible for the sentiment analysis of a sentence as a building block for this research. We constructed an algorithm using NLP models to derive a sentiment analysis score against sentences. This was then tested against a sample of 20 respondents to derive a sentiment analysis score. The scores generated resembled human responses. By utilising the MLP classifier, decision tree, linear model, and K-nearest neighbors, we were able to obtain a test accuracy of 89.97%, 54.63%, 52.13%, and 47.9%, respectively. This gave us the confidence to use the NLP models against sentences in websites to measure the GNH of a country.

Keywords: artificial intelligence, NLP, sentiment analysis, gross national happiness

Procedia PDF Downloads 75

Authors: Duncan Wallace, M-Tahar Kechadi

Abstract:

In recent years, Machine Learning (ML) approaches have been successfully applied to an analysis of patient symptom data in the context of disease diagnosis, at least where such data is well codified. However, much of the data present in Electronic Health Records (EHR) are unlikely to prove suitable for classic ML approaches. Furthermore, as scores of data are widely spread across both hospitals and individuals, a decentralized, computationally scalable methodology is a priority. The focus of this paper is to develop a method to predict outliers in an out-of-hours healthcare provision center (OOHC). In particular, our research is based upon the early identification of patients who have underlying conditions which will cause them to repeatedly require medical attention. OOHC act as an ad-hoc delivery of triage and treatment, where interactions occur without recourse to a full medical history of the patient in question. Medical histories, relating to patients contacting an OOHC, may reside in several distinct EHR systems in multiple hospitals or surgeries, which are unavailable to the OOHC in question. As such, although a local solution is optimal for this problem, it follows that the data under investigation is incomplete, heterogeneous, and comprised mostly of noisy textual notes compiled during routine OOHC activities. Through the use of Deep Learning methodologies, the aim of this paper is to provide the means to identify patient cases, upon initial contact, which are likely to relate to such outliers. To this end, we compare the performance of Long Short-Term Memory, Gated Recurrent Units, and combinations of both with Convolutional Neural Networks. A further aim of this paper is to elucidate the discovery of such outliers by examining the exact terms which provide a strong indication of positive and negative case entries. While free-text is the principal data extracted from EHRs for classification, EHRs also contain normalized features. Although the specific demographical features treated within our corpus are relatively limited in scope, we examine whether it is beneficial to include such features among the inputs to our neural network, or whether these features are more successfully exploited in conjunction with a different form of a classifier. In this section, we compare the performance of randomly generated regression trees and support vector machines and determine the extent to which our classification program can be improved upon by using either of these machine learning approaches in conjunction with the output of our Recurrent Neural Network application. The output of our neural network is also used to help determine the most significant lexemes present within the corpus for determining high-risk patients. By combining the confidence of our classification program in relation to lexemes within true positive and true negative cases, with an inverse document frequency of the lexemes related to these cases, we can determine what features act as the primary indicators of frequent-attender and non-frequent-attender cases, providing a human interpretable appreciation of how our program classifies cases.

Keywords: artificial neural networks, data-mining, machine learning, medical informatics

Procedia PDF Downloads 104

Authors: Zhuang Yiwen

Abstract:

The paper examines the challenges facing the Singapore healthcare system related to the tracking and traceability of medical devices. One of the major challenges identified is the lack of a standard coding system for medical devices, which makes it difficult to track them effectively. The paper suggests the use of the Unique Device Identifier (UDI) as a single standard for medical devices to improve tracking and reduce errors. The paper also explores the use of barcoding and image recognition to identify and document medical devices in nursing practices. In nursing practices, the use of barcodes for identifying medical devices is common. However, the information contained in these barcodes is often inconsistent, making it challenging to identify which segment contains the model identifier. Moreover, the use of barcodes may be improved with the use of UDI, but many subsidized accessories may still lack barcodes. The paper suggests that the readiness for UDI and barcode standardization requires standardized information, fields, and logic in electronic medical record (EMR), operating theatre (OT), and billing systems, as well as barcode scanners that can read various formats and selectively parse barcode segments. Nursing workflow and data flow also need to be taken into account. The paper also explores the use of image recognition, specifically the Tesseract OCR engine, to identify and document implants in public hospitals due to limitations in barcode scanning. The study found that the solution requires an implant information database and checking output against the database. The solution also requires customization of the algorithm, cropping out objects affecting text recognition, and applying adjustments. The solution requires additional resources and costs for a mobile/hardware device, which may pose space constraints and require maintenance of sterile criteria. The integration with EMR is also necessary, and the solution require changes in the user's workflow. The paper suggests that the long-term use of Systematized Nomenclature of Medicine Clinical Terms (SNOMED CT) as a supporting terminology to improve clinical documentation and data exchange in healthcare. SNOMED CT provides a standardized way of documenting and sharing clinical information with respect to procedure, patient and device documentation, which can facilitate interoperability and data exchange. In conclusion, the paper highlights the challenges facing the Singapore healthcare system related to the tracking and traceability of medical devices. The paper suggests the use of UDI and barcode standardization to improve tracking and reduce errors. It also explores the use of image recognition to identify and document medical devices in nursing practices. The paper emphasizes the importance of standardized information, fields, and logic in EMR, OT, and billing systems, as well as barcode scanners that can read various formats and selectively parse barcode segments. These recommendations could help the Singapore healthcare system to improve tracking and traceability of medical devices and ultimately enhance patient safety.

Keywords: medical device tracking, unique device identifier, barcoding and image recognition, systematized nomenclature of medicine clinical terms

Procedia PDF Downloads 50

Authors: Djamila Meghraoui, Bachir Boudraa, Thouraya Meksen, M.Boudraa

Abstract:

Parkinson's disease is a pathology that involves characteristic perturbations in patients’ voices. This paper describes a proposed method that aims to diagnose persons with Parkinson (PWP) by analyzing on line their voices signals. First, Thresholds signals alterations are determined by the Multi-Dimensional Voice Program (MDVP). Principal Analysis (PCA) is exploited to select the main voice principal componentsthat are significantly affected in a patient. The decision phase is realized by a Mul-tinomial Bayes (MNB) Classifier that categorizes an analyzed voice in one of the two resulting classes: healthy or PWP. The prediction accuracy achieved reaching 98.8% is very promising.

Keywords: Parkinson’s disease recognition, PCA, MDVP, multinomial Naive Bayes

Procedia PDF Downloads 251

Authors: Punnoose A. K.

Abstract:

This paper presents an approach for identifying noisy speech recording using a multi-layer perception (MLP) trained to predict phonemes from acoustic features. Characteristics of the MLP posteriors are explored for clean speech and noisy speech at the frame level. Appropriate density functions are used to fit the softmax probability of the clean and noisy speech. A function that takes into account the ratio of the softmax probability density of noisy speech to clean speech is formulated. These phoneme independent scoring is weighted using a phoneme-specific weightage to make the scoring more robust. Simple thresholding is used to identify the noisy speech recording from the clean speech recordings. The approach is benchmarked on standard databases, with a focus on precision.

Keywords: noisy speech identification, speech pre-processing, noise robustness, feature engineering

Procedia PDF Downloads 101

Authors: Elia Favarelli, Enrico Testi, Andrea Giorgetti

Abstract:

Autonomous structural health monitoring (SHM) of many structures and bridges became a topic of paramount importance for maintenance purposes and safety reasons. This paper proposes a set of machine learning (ML) tools to perform automatic feature selection and detection of anomalies in a bridge from vibrational data and compare different feature extraction schemes to increase the accuracy and reduce the amount of data collected. As a case study, the Z-24 bridge is considered because of the extensive database of accelerometric data in both standard and damaged conditions. The proposed framework starts from the first four fundamental frequencies extracted through operational modal analysis (OMA) and clustering, followed by density-based time-domain filtering (tracking). The fundamental frequencies extracted are then fed to a dimensionality reduction block implemented through two different approaches: feature selection (intelligent multiplexer) that tries to estimate the most reliable frequencies based on the evaluation of some statistical features (i.e., mean value, variance, kurtosis), and feature extraction (auto-associative neural network (ANN)) that combine the fundamental frequencies to extract new damage sensitive features in a low dimensional feature space. Finally, one class classifier (OCC) algorithms perform anomaly detection, trained with standard condition points, and tested with normal and anomaly ones. In particular, a new anomaly detector strategy is proposed, namely one class classifier neural network two (OCCNN2), which exploit the classification capability of standard classifiers in an anomaly detection problem, finding the standard class (the boundary of the features space in normal operating conditions) through a two-step approach: coarse and fine boundary estimation. The coarse estimation uses classics OCC techniques, while the fine estimation is performed through a feedforward neural network (NN) trained that exploits the boundaries estimated in the coarse step. The detection algorithms vare then compared with known methods based on principal component analysis (PCA), kernel principal component analysis (KPCA), and auto-associative neural network (ANN). In many cases, the proposed solution increases the performance with respect to the standard OCC algorithms in terms of F1 score and accuracy. In particular, by evaluating the correct features, the anomaly can be detected with accuracy and an F1 score greater than 96% with the proposed method.

Keywords: anomaly detection, frequencies selection, modal analysis, neural network, sensor network, structural health monitoring, vibration measurement

Procedia PDF Downloads 97

Authors: Pilar Rey del Castillo

Abstract:

This document describes the building of an average indicator of the general sentiments about the future exposed in the newspapers in Spain. The raw data are collected through the scraping of the Digital Periodical and Newspaper Library website. Basic tools of natural language processing are later applied to the collected information to evaluate the sentiment strength of each word in the texts using a polarized dictionary. The last step consists of summarizing these sentiments to produce daily indices. The results are a first insight into the applicability of these techniques to produce periodic sentiment indicators.

Keywords: natural language processing, periodic indicator, sentiment analysis, web scraping

Procedia PDF Downloads 107

Authors: A. Gonzalez Aguna, J. M. Santamaria Garcia, J. L. Gomez Gonzalez, R. Barchino Plata, M. Fernandez Batalla, S. Herrero Jaen

Abstract:

Background: Computer Science is a field that transcends other disciplines of knowledge because it allows to support all kinds of physical and mental tasks. Health centres have a greater number and complexity of technological devices and the population consume and demand services derived from technology. Also, nursing education plans have included competencies related to and, even, courses about new technologies are offered to health professionals. However, nurses still limit their performance to the use and evaluation of products previously built. Objective: Develop a teaching-learning methodology for acquiring skills on designing applications for care. Methodology: Blended learning teaching with a group of graduate nurses through official training within a Master's Degree. The study sample was selected by intentional sampling without exclusion criteria. The study covers from 2015 to 2017. The teaching sessions included a four-hour face-to-face class and between one and three tutorials. The assessment was carried out by written test consisting of the preparation of an IEEE 830 Standard Specification document where the subject chosen by the student had to be a problem in the area of care. Results: The sample is made up of 30 students: 10 men and 20 women. Nine students had a degree in nursing, 20 diploma in nursing and one had a degree in Computer Engineering. Two students had a degree in nursing specialty through residence and two in equivalent recognition by exceptional way. Except for the engineer, no subject had previously received training in this regard. All the sample enrolled in the course received the classroom teaching session, had access to the teaching material through a virtual area and maintained at least one tutoring. The maximum of tutorials were three with an hour in total. Among the material available for consultation was an example of a document drawn up based on the IEEE Standard with an issue not related to care. The test to measure competence was completed by the whole group and evaluated by a multidisciplinary teaching team of two computer engineers and two nurses. Engineers evaluated the correctness of the characteristics of the document and the degree of comprehension in the elaboration of the problem and solution elaborated nurses assessed the relevance of the chosen problem statement, the foundation, originality and correctness of the proposed solution and the validity of the application for clinical practice in care. The results were of an average grade of 8.1 over 10 points, a range between 6 and 10. The selected topic barely coincided among the students. Examples of care areas selected are care plans, family and community health, delivery care, administration and even robotics for care. Conclusion: The applied methodology of learning-teaching for the design of technologies demonstrates the success in the training of nursing professionals. The role of expert is essential to create applications that satisfy the needs of end users. Nursing has the possibility, the competence and the duty to participate in the process of construction of technological tools that are going to impact in care of people, family and community.

Keywords: care, learning, nursing, technology

Procedia PDF Downloads 110

Authors: Xavi Vilajosana Guillen, Emilio José Pérez Salgado

Abstract:

This document presents experimental results obtained in a realistic environment using an underwater LoRa link. It aims to analyze the behavior of electromagnetic waves underwater and determine this communication capability. With this it has been tried to empirically evaluate the results obtained in the mathematical model using a commercial device with low cost and low consumption that works at frequency 433Mhz. The mathematical results obtained for wireless communication at 433Mhz underwater indicate that a communication of up to 7.5 m is possible, however experimentally 8 m has been achieved.

Keywords: 433Mhz link, internet of things, LoRa link, underwater communication

Procedia PDF Downloads 44

Authors: Fatima Akhmedova, Simon Liao

Abstract:

One of the most critical decision points in the design of a face recognition system is the choice of an appropriate face representation. Effective feature descriptors are expected to convey sufficient, invariant and non-redundant facial information. In this work, we propose a set of Hahn moments as a new approach for feature description. Hahn moments have been widely used in image analysis due to their invariance, non-redundancy and the ability to extract features either globally and locally. To assess the applicability of Hahn moments to Face Recognition we conduct two experiments on the Olivetti Research Laboratory (ORL) database and University of Notre-Dame (UND) X1 biometric collection. Fusion of the global features along with the features from local facial regions are used as an input for the conventional k-NN classifier. The method reaches an accuracy of 93% of correctly recognized subjects for the ORL database and 94% for the UND database.

Keywords: face recognition, Hahn moments, recognition-by-parts, time-lapse

Procedia PDF Downloads 341

Authors: Somaiyeh Dehghan, Mohammad Reza Kheirkhahan Haghighi

Abstract:

The rapid growth of domains that have data with a large number of features, while the number of samples is limited has caused difficulty in constructing strong classifiers. To reduce the dimensionality of the feature space becomes an essential step in classification task. Random subspace method (or attribute bagging) is an ensemble classifier that consists of several classifiers that each base learner in ensemble has subset of features. In the present paper, we introduce Random Subspace Ensemble of CMAC neural network (RSE-CMAC), each of which has training with subset of features. Then we use this model for classification task. For evaluation performance of our model, we compare it with bagging algorithm on 36 UCI datasets. The results reveal that the new model has better performance.

Keywords: classification, random subspace, ensemble, CMAC neural network

Procedia PDF Downloads 305

Authors: Refilwe Moeletsi

Abstract:

Dimension stone quarrying around Brits and Belfast areas started in the early 1930s and has been growing rapidly since then. Environmental impacts associated with these quarries have not been documented, and hence this study aims at detecting any change in the environment that might have been caused by these activities. Landsat images that were used to assess land use/land cover changes in Brits quarries from 1998 - 2015. A supervised classification using maximum likelihood classifier was applied to classify each image into different land use/land cover types. Classification accuracy was assessed using Google Earth™ as a source of reference data. Post-classification change detection method was used to determine changes. The results revealed significant increase in granite quarries and corresponding decrease in vegetation cover within the study region.

Keywords: remote sensing, GIS, change detection, granite quarries

Procedia PDF Downloads 286