Enhancing Multimodal Silent Speech Interfaces With Feature Selection

In research on Silent Speech Interfaces (SSI), different sources of information (modalities) have been combined, aiming at obtaining better performance than the individual modalities. However, when combining these modalities, the dimensionality of the feature space rapidly increases, yielding the well-known "curse of dimensionality". As a consequence, in order to extract useful information from this data, one has to resort to feature selection (FS) techniques to lower the dimensionality of the learning space. In this paper, we assess the impact of FS techniques for silent speech data, in a dataset with 4 non-invasive and promising modalities, namely: video, depth, ultrasonic Doppler sensing, and surface electromyography. We consider two supervised (mutual information and Fisher's ratio) and two unsupervised (meanmedian and arithmetic mean geometric mean) FS filters. The evaluation was made by assessing the classification accuracy (word recognition error) of three well-known classifiers (knearest neighbors, support vector machines, and dynamic time warping). The key results of this study show that both unsupervised and supervised FS techniques improve on the classification accuracy on both individual and combined modalities. For instance, on the video component, we attain relative performance gains of 36.2% in error rates. FS is also useful as pre-processing for feature fusion. Copyright © 2014 ISCA.

Fatty liver automatic diagnosis from ultrasound images

In this paper an automatic classification algorithm is proposed for the diagnosis of the liver steatosis, also known as, fatty liver, from ultrasound images. The features, automatically extracted from the ultrasound images used by the classifier, are basically the ones used by the physicians in the diagnosis of the disease based on visual inspection of the ultrasound images. The main novelty of the method is the utilization of the speckle noise that corrupts the ultrasound images to compute textural features of the liver parenchyma relevant for the diagnosis. The algorithm uses the Bayesian framework to compute a noiseless image, containing anatomic and echogenic information of the liver and a second image containing only the speckle noise used to compute the textural features. The classification results, with the Bayes classifier using manually classified data as ground truth show that the automatic classifier reaches an accuracy of 95% and a 100% of sensitivity.

Unsupervised automatic music genre classification

Trabalho apresentado no âmbito do Mestrado em Engenharia Informática, como requisito parcial para obtenção do grau de Mestre em Engenharia Informática

An automatic voice pleasantness classification system based on prosodic and acoustic patterns of voice preference

In the last few years the number of systems and devices that use voice based interaction has grown significantly. For a continued use of these systems the interface must be reliable and pleasant in order to provide an optimal user experience. However there are currently very few studies that try to evaluate how good is a voice when the application is a speech based interface. In this paper we present a new automatic voice pleasantness classification system based on prosodic and acoustic patterns of voice preference. Our study is based on a multi-language database composed by female voices. In the objective performance evaluation the system achieved a 7.3% error rate.

Stratigraphic interpretation of Well-Log data of the Athabasca Oil Sands of Alberta Canada through Pattern recognition and Artificial Intelligence

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies.

Automatic visual detection of human behavior: a review from 2000 to 2014

Due to advances in information technology (e.g., digital video cameras, ubiquitous sensors), the automatic detection of human behaviors from video is a very recent research topic. In this paper, we perform a systematic and recent literature review on this topic, from 2000 to 2014, covering a selection of 193 papers that were searched from six major scientific publishers. The selected papers were classified into three main subjects: detection techniques, datasets and applications. The detection techniques were divided into four categories (initialization, tracking, pose estimation and recognition). The list of datasets includes eight examples (e.g., Hollywood action). Finally, several application areas were identified, including human detection, abnormal activity detection, action recognition, player modeling and pedestrian detection. Our analysis provides a road map to guide future research for designing automatic visual human behavior detection systems.

An integrated system for human action recognition from video using hidden Markov model

In this paper, we present an integrated system for real-time automatic detection of human actions from video. The proposed approach uses the boundary of humans as the main feature for recognizing actions. Background subtraction is performed using Gaussian mixture model. Then, features are extracted from silhouettes and Vector Quantization is used to map features into symbols (bag of words approach). Finally, actions are detected using the Hidden Markov Model. The proposed system was validated using a newly collected real- world dataset. The obtained results show that the system is capable of achieving robust human detection, in both indoor and outdoor environments. Moreover, promising classification results were achieved when detecting two basic human actions: walking and sitting.

The role of automatic shape and position recognitionin streamlining manufacturing

The main features of most components consist of simple basic functional geometries: planes, cylinders, spheres and cones. Shape and position recognition of these geometries is essential for dimensional characterization of components, and represent an important contribution in the life cycle of the product, concerning in particular the manufacturing and inspection processes of the final product. This work aims to establish an algorithm to automatically recognize such geometries, without operator intervention. Using differential geometry large volumes of data can be treated and the basic functional geometries to be dealt recognized. The original data can be obtained by rapid acquisition methods, such as 3D survey or photography, and then converted into Cartesian coordinates. The satisfaction of intrinsic decision conditions allows different geometries to be fast identified, without operator intervention. Since inspection is generally a time consuming task, this method reduces operator intervention in the process. The algorithm was first tested using geometric data generated in MATLAB and then through a set of data points acquired by measuring with a coordinate measuring machine and a 3D scan on real physical surfaces. Comparison time spent in measuring is presented to show the advantage of the method. The results validated the suitability and potential of the algorithm hereby proposed

Automatic eye blink detection using consumer web cameras

This research aims to advance blinking detection in the context of work activity. Rather than patients having to attend a clinic, blinking videos can be acquired in a work environment, and further automatically analyzed. Therefore, this paper presents a methodology to perform the automatic detection of eye blink using consumer videos acquired with low-cost web cameras. This methodology includes the detection of the face and eyes of the recorded person, and then it analyzes the low-level features of the eye region to create a quantitative vector. Finally, this vector is classified into one of the two categories considered —open and closed eyes— by using machine learning algorithms. The effectiveness of the proposed methodology was demonstrated since it provides unbiased results with classification errors under 5%

Forecasting human entrances at a commercial store using facial recognition data

Dissertação de mestrado integrado em Engenharia e Gestão de Sistemas de Informação

Automatic Mallampati Classification Using Active Appearance Models

Difficult tracheal intubation assessment is an important research topic in anesthesia as failed intubations are important causes of mortality in anesthetic practice. The modified Mallampati score is widely used, alone or in conjunction with other criteria, to predict the difficulty of intubation. This work presents an automatic method to assess the modified Mallampati score from an image of a patient with the mouth wide open. For this purpose we propose an active appearance models (AAM) based method and use linear support vector machines (SVM) to select a subset of relevant features obtained using the AAM. This feature selection step proves to be essential as it improves drastically the performance of classification, which is obtained using SVM with RBF kernel and majority voting. We test our method on images of 100 patients undergoing elective surgery and achieve 97.9% accuracy in the leave-one-out crossvalidation test and provide a key element to an automatic difficult intubation assessment system.

Speech by Mr. Micheál Martin TD – Meeting The Challenges Of Cultural Diversity In The Irish Healthcare Sector

It gives me great pleasure to accept the invitation to address this conference on “Meeting the Challenges of Cultural Diversity in the Irish Healthcare Sector” which is being organised by the Irish Health Services Management Institute in partnership with the National Consultative Committee on Racism and Interculturalism. The conference provides an important opportunity to develop our knowledge and understanding of the issues surrounding cultural diversity in the health sector from the twin perspectives of patients and staff. Cultural diversity has over recent years become an increasingly visible aspect of Irish society bringing with it both opportunities and challenges. It holds out great possibilities for the enrichment of all who live in Ireland but it also challenges us to adapt creatively to the changes required to realise this potential and to ensure that the experience is a positive one for all concerned but particularly for those in the minority ethnic groups. In the last number of years in particular, the focus has tended to be on people coming to this country either as refugees, asylum seekers or economic migrants. Government figures estimate that as many as 340,000 immigrants are expected in the next six years. However ethnic and cultural diversity are not new phenomena in Ireland. Travellers have a long history as an indigenous minority group in Ireland with a strong culture and identity of their own. The changing experience and dynamics of their relationship with the wider society and its institutions over time can, I think, provide some valuable lessons for us as we seek to address the more numerous and complex issues of cultural diversity which have arisen for us in the last decade. Turning more specifically to the health sector which is the focus of this conference, culture and identity have particular relevance to health service policy and provision in that The first requirement is that we in the health service acknowledge cultural diversity and the differences in behaviours and in the less obvious areas of values and beliefs that this often implies. Only by acknowledging these differences in a respectful way and informing ourselves of them can we address them. Our equality legislation – The Employment Equality Act, 1998 and the Equal Status Act, 2000 – prohibits discrimination on nine grounds including race and membership of the Traveller community. The Equal Status Act prohibits discrimination on an individual basis in relation to the nine grounds while for groups it provides for the promotion of equality of opportunity. The Act applies to the provision of services including health services. I will speak first about cultural diversity in relation to the patient. In this respect it is worth mentioning that the recognition of cultural diversity and appropriate responses to it were issues which were strongly emphasised in the public consultation process which we held earlier this year in the context of developing National Anti-Poverty targets for the health sector and also our new national health strategy. Awareness and sensitivity training for staff is a key requirement for adapting to a culturally diverse patient population. The focus of this training should be the development of the knowledge and skills to provide services sensitive to cultural diversity. Such training can often be most effectively delivered in partnership with members of the minority groups themselves. I am aware that the Traveller community, for example, is involved in in-service training for health care workers. I am also aware that the National Consultative Committee on Racism and Interculturalism has been involved in training with the Eastern Regional Health Authority. We need to have more such initiatives. A step beyond the sensitivity training for existing staff is the training of members of the minority communities themselves as workers in our health services. Again the Traveller community has set an example in this area with its Primary Health Care Project for Travellers. The Primary Health Care for Travellers Project was established in 1994 as a joint partnership initiative with the Eastern Health Board and Pavee Point, with ongoing technical assistance being provided from the Department of Community Health and General Practice, Trinity College, Dublin. This project was the first of its kind in the country and has facilitated The project included a training course which concentrated on skills development, capacity building and the empowerment of Travellers. This confidence and skill allowed the Community Health Workers to go out and conduct a baseline survey to identify and articulate Travellers’ health needs. This was the first time that Travellers were involved in this process; in the past their needs were assumed. The results of the survey were fed back to the community and they prioritised their needs and suggested changes to the health services which would facilitate their access and utilisation. Ongoing monitoring and data collection demonstrates a big improvement in levels of satisfaction and uptake and ulitisation of health services by Travellers in the pilot area. This Primary Health Care for Travellers initiative is being replicated in three other areas around the country and funding has been approved for a further 9 new projects. This pilot project was the recipient of a WHO 50th anniversary commemorative award in 1998. The project is developing as a model of good practice which could inspire further initiatives of this type for other minority groups. Access to information has been identified in numerous consultative processes as a key factor in enabling people to take a proactive approach to managing their own health and that of their families and in facilitating their access to health services. Honouring our commitment to equity in these areas requires that information is provided in culturally appropriate formats. The National Health Promotion Strategy 2000-2005, for example, recognises that there exists within our society many groups with different requirements which need to be identified and accommodated when planning and implementing health promotion interventions. These groups include Travellers, refugees and asylum seekers, people with intellectual, physical or sensory disability and the gay and lesbian community. The Strategy acknowledges the challenge involved in being sensitive to the potential differences in patterns of poor health among these different groups. The Strategic aim is to promote the physical, mental and social well-being of individuals from these groups. The objective of the Strategy on these issues are: While our long term aim may be to mainstream responses so that our health services is truly multicultural, we must recognise the need at this point in time for very specific focused responses particularly for groups with poor health status such as Travellers and also for refugees and asylum seekers. In the case of refugees and asylum seekers examples of targeted services are screening for communicable diseases – offered on a voluntary basis – and psychological support services for those who have suffered trauma before coming here. The two approaches of targeting and mainstreaming are not mutually exclusive. A combination of both is required at this point in time but the balance between them must be kept under constant review in the light of changing needs. A major requirement if we are to meet the challenge of cultural diversity is an appropriate data and research base. I think it is important that we build up our information and research data base in partnership with the minority groups themselves. We must establish what the health needs of diverse groups are; we must monitor uptake of services and how well we are responding to needs and we must monitor outcomes and health status. We must also examine the impact of the policies in other sectors on the health of minority groups. The National Health Information Strategy, currently being developed, and the recently published National Strategy for Health Research – Making Knowledge Work for Health provide important frameworks within which we can improve our data and research base. A culturally diverse health sector workforce – challenges and opportunities The Irish health service can benefit greatly from successful international recruitment. There has been a strong non-national representation amongst the medical profession for more than 30 years. More recently there have been significant increases in other categories of health service workers from overseas. The Department recognises the enormous value that overseas recruitment brings over a wide range of services and supports the development of effective and appropriate recruitment strategies in partnership with health service employers. These changes have made cultural diversity an important issue for all health service organisations. Diversity in the workplace is primarily about creating a culture that seeks, respects, values and harnesses difference. This includes all the differences that when added together make each person unique. So instead of the focus being on particular groups, diversity is about all of us. Change is not about helping “them” to join “us” but about critically looking at “us” and rooting out all aspects of our culture that inappropriately exclude people and prevent us from being inclusive in the way we relate to employees, potential employees and clients of the health service. International recruitment benefits consumers, Irish employees and the overseas personnel alike. Regardless of whether they are employed by the health service, members of minority groups will be clients of our service and consequently we need to be flexible in order to accommodate different cultural needs. For staff, we recognise that coming from other cultures can be a difficult transition. Consequently health service employers have made strong efforts to assist them during this period. Many organisations provide induction courses, religious facilities (such as prayer rooms) and help in finding suitable accommodation. The Health Service Employers Agency (HSEA) is developing an equal opportunities/diversity strategy and action plans as well as training programmes to support their implementation, to ensure that all health service employment policies and practices promote the equality/diversity agenda to continue the development of a culturally diverse health service. The management of this new environment is extremely important for the health service as it offers an opportunity to go beyond set legal requirements and to strive for an acceptance and nurturing of cultural differences. Workforce cultural diversity affords us the opportunity to learn from the working practices and perspectives of others by allowing personnel to present their ideas and experience through teamwork, partnership structures and other appropriate fora, leading to further improvement in the services we provide. It is important to ensure that both personnel units and line managers communicate directly with their staff and demonstrate by their actions that they intend to create an inclusive work place which doesn´t demand that minority staff fit. Contented, valued employees who feel that there is a place for them in the organisation will deliver a high quality health service. Your conference here today has two laudable aims – to heighten awareness and assist health care staff to work effectively with their colleagues from different cultural backgrounds and to gain a greater understanding of the diverse needs of patients from minority ethnic backgrounds. There is a synergy in these aims and in the tasks to which they give rise in the management of our health service. The creative adaptations required for one have the potential to feed into the other. I would like to commend both organisations which are hosting this conference for their initiative in making this event happen, particularly at this time – Racism in the Workplace Week. I look forward very much to hearing the outcome of your deliberations. Thank you.

A new approach to real time traffic sign recognition based on colour information

The automatic interpretation of conventional traffic signs is very complex and time consuming. The paper concerns an automatic warning system for driving assistance. It does not interpret the standard traffic signs on the roadside; the proposal is to incorporate into the existing signs another type of traffic sign whose information will be more easily interpreted by a processor. The type of information to be added is profuse and therefore the most important object is the robustness of the system. The basic proposal of this new philosophy is that the co-pilot system for automatic warning and driving assistance can interpret with greater ease the information contained in the new sign, whilst the human driver only has to interpret the "classic" sign. One of the codings that has been tested with good results and which seems to us easy to implement is that which has a rectangular shape and 4 vertical bars of different colours. The size of these signs is equivalent to the size of the conventional signs (approximately 0.4 m2). The colour information from the sign can be easily interpreted by the proposed processor and the interpretation is much easier and quicker than the information shown by the pictographs of the classic signs

New perspectives in the use of ink evidence in forensic science: Part II. Development and testing of mathematical algorithms for the automatic comparison of ink samples analysed by HPTLC

In the first part of this research, three stages were stated for a program to increase the information extracted from ink evidence and maximise its usefulness to the criminal and civil justice system. These stages are (a) develop a standard methodology for analysing ink samples by high-performance thin layer chromatography (HPTLC) in reproducible way, when ink samples are analysed at different time, locations and by different examiners; (b) compare automatically and objectively ink samples; and (c) define and evaluate theoretical framework for the use of ink evidence in forensic context. This report focuses on the second of the three stages. Using the calibration and acquisition process described in the previous report, mathematical algorithms are proposed to automatically and objectively compare ink samples. The performances of these algorithms are systematically studied for various chemical and forensic conditions using standard performance tests commonly used in biometrics studies. The results show that different algorithms are best suited for different tasks. Finally, this report demonstrates how modern analytical and computer technology can be used in the field of ink examination and how tools developed and successfully applied in other fields of forensic science can help maximising its impact within the field of questioned documents.

OSIRISv1.2: a named entity recognition system for sequence variants of genes in biomedical literature

Background: Single Nucleotide Polymorphisms, among other type of sequence variants, constitute key elements in genetic epidemiology and pharmacogenomics. While sequence data about genetic variation is found at databases such as dbSNP, clues about the functional and phenotypic consequences of the variations are generally found in biomedical literature. The identification of the relevant documents and the extraction of the information from them are hampered by the large size of literature databases and the lack of widely accepted standard notation for biomedical entities. Thus, automatic systems for the identification of citations of allelic variants of genes in biomedical texts are required. Results: Our group has previously reported the development of OSIRIS, a system aimed at the retrieval of literature about allelic variants of genes http://ibi.imim.es/osirisform.html. Here we describe the development of a new version of OSIRIS (OSIRISv1.2, http://ibi.imim.es/OSIRISv1.2.html webcite) which incorporates a new entity recognition module and is built on top of a local mirror of the MEDLINE collection and HgenetInfoDB: a database that collects data on human gene sequence variations. The new entity recognition module is based on a pattern-based search algorithm for the identification of variation terms in the texts and their mapping to dbSNP identifiers. The performance of OSIRISv1.2 was evaluated on a manually annotated corpus, resulting in 99% precision, 82% recall, and an F-score of 0.89. As an example, the application of the system for collecting literature citations for the allelic variants of genes related to the diseases intracranial aneurysm and breast cancer is presented. Conclusion: OSIRISv1.2 can be used to link literature references to dbSNP database entries with high accuracy, and therefore is suitable for collecting current knowledge on gene sequence variations and supporting the functional annotation of variation databases. The application of OSIRISv1.2 in combination with controlled vocabularies like MeSH provides a way to identify associations of biomedical interest, such as those that relate SNPs with diseases.