The actual economic situation, caused the change in the way that many people move on the cities, ... more The actual economic situation, caused the change in the way that many people move on the cities, due to this fact a big majority of the travelers, began to use public means of transport, both because these means are comfortable enough to make a day by day journey and because they are much more cheaper. However, there are routes where you need to use different means of public transport, on those cases improve the time management is crucial for the user, or he might arrive late on his destination. Countries in Europe like Belgium and Holland, have adopted intermodal route planners that gather different types of information, both form public and private means of transport, and also give tools to the user, for him to plan the journey with the less effort possible. The aim of this work is to analyze if this kind of platform could be introduced in Portugal. So a survey was made, in order to find what functionalities the user would hope to find in an application that could help him to plan a journey using public means of transport. At technical level other analysis was made in order to integrate the different tools that an intermodal route planner, could or should have, to help the user on the planification of a journey using public means of transport.
2016 International Conference on ReConFigurable Computing and FPGAs (ReConFig), 2016
Our work proposes a hardware architecture for a Long Short-Term Memory (LSTM) Neural Network, aim... more Our work proposes a hardware architecture for a Long Short-Term Memory (LSTM) Neural Network, aiming to outperform software implementations, by exploiting its inherent parallelism. The main design decisions are presented, along with the proposed network architecture. A description of the main building blocks of the network is also presented. The network is synthesized for various sizes and platforms, and the performance results are presented and analyzed. Our synthesized network achieves a 251 times speed-up over a custom-built software network, running on an i7-3770k Desktop computer, proving the benefits of parallel computation for this kind of network.
2019 International Young Engineers Forum (YEF-ECE), 2019
The AIA (Atmospheric Imaging Assembly) instrument, on-board the SDO (Solar Dynamics Observatory) ... more The AIA (Atmospheric Imaging Assembly) instrument, on-board the SDO (Solar Dynamics Observatory) satellite, provides high-resolution and high-cadence solar images since 2010. To extract scientific knowledge from those high-resolution images there is a need for efficient automatic web tools to detect and/or track the coronal bright points (CBPs). Identifying and tracking CBPs is essential for successfully calculate the solar corona rotation rate at different latitudes. Over the last years this topic has been an area of research in solar physics and some effective methods have been developed. The purpose of this work is to design an automatic and near real-time web tool that detects and tracks CBPs on solar images and publishes the results online, thus, allowing search and visualization functionalities to support astrophysicists perform improved solar analysis. The detection process uses a gradient based segmentation algorithm that has proved to provide accurate data about CBPs' dynamics. The paper propose to extend that work by using SunPy and OpenCV in Python, together with the Gradient Path Labeling (GPL) segmentation algorithm. The results obtained display good approximations, when compared with the ones from other authors, and the tool also seems reliable over long testing periods.
Promoters in Escherichia coli include an ‘OFF’ state, during which transcription is halted. Here,... more Promoters in Escherichia coli include an ‘OFF’ state, during which transcription is halted. Here, we propose a novel empirical method for assessing the time-length spent by promoters in this state. It relies on direct measurements of RNA production kinetics at the single molecule level at different induction levels, followed by an estimation of the RNA production rate under infinite induction, which is then compared to this rate under real, maximum induction. We apply it to the LacO3O1 promoter and infer that, under full induction, on average, 15% of the time between successful transcription events is spent in the OFF state. We verify this result by comparing the kinetics of a mutant strain lacking repressor molecules with that of the inferred rate under infinite induction. We expect this strategy of dissecting the kinetics of transcription repression to be applicable to a wide number of promoters in E. coli.
2019 IEEE Congress on Evolutionary Computation (CEC), 2019
Nowadays, the number of aerial unmanned vehicles (UAVs) is growing at a tremendous speed, as well... more Nowadays, the number of aerial unmanned vehicles (UAVs) is growing at a tremendous speed, as well as its technology. Therefore, it is essential to follow this growth with increasingly robust algorithms to be possible to exist cooperation between robots autonomously. One of the major events currently being developed in autonomous cooperation is relatively terrain classification where, this classification, is mainly important for emergency landings, mapping and decision making. This paper presents a robust computer vision system to sort terrain types using two main algorithms: Particle Swarm Optimization (PSO) and Gray-Level Co-Occurrence Matrix (GLCM). In addition to these two algorithms, a neural network was designed with the aim of increasing the probability of success of the proposed system. In order to evaluate this article, the system is validated using videos acquired onboard of a UAV with a RGB camera.
2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, 2021
Wildfires are recurrent natural disasters in some regions of the globe, being Portugal one of the... more Wildfires are recurrent natural disasters in some regions of the globe, being Portugal one of these areas. The Portuguese Institute for Nature Conservation and Forests implemented a national fuel break (FB) network, with the goal of decreasing fire hazard. FBs are areas of reduced fuel load that slow down fire spread, creating firefighting opportunities. Its effectiveness relies on periodic treatments to maintain the fuel load in levels that can reduce the effects of wildfires. This paper proposes a methodology to assess the FB state, according to its fuel load, based on the analysis of the inter-annual variability of Sentinel-2 NDVI time series. Inter-annual comparison allows it to adapt to different regions. To assess the reliability of NDVI data to evaluate the FB state, a linear regression with the vegetation height acquired by the Global Ecosystem Dynamics Investigation (GEDI) mission was tested, achieving determination coefficients between 0.57 and 0.98.
Cell division in Escherichia coli is morphologically symmetric due to, among other, these cells&#... more Cell division in Escherichia coli is morphologically symmetric due to, among other, these cells' ability to place the Z-ring at midcell. Studies have reported that, at sub-optimal temperatures, this symmetry decreases at the single-cell level, but the causes remain unclear. Using fluorescence microscopy, we observe FtsZ-GFP and DAPI-stained nucleoids to assess the robustness of the symmetry of Z-ring formation and positioning in individual cells under sub-optimal and critical temperatures. We find the Z-ring formation and positioning to be robust at sub-optimal temperatures, as the Z-ring's mean width, density and displacement from midcell maintain similar levels of correlation to one another as in optimal temperatures. However, at critical temperatures, the Z-ring displacement from midcell is much increased. We show evidence that this is due to enhanced distance between the replicated nucleoids and, thus, reduced Z-ring density, which explains the weaker precision in settin...
Proceedings of the 1st International Living Usability Lab Workshop on AAL Latest Solutions, Trends and Applications, 2011
Living Usability Lab for Next Generation Networks (www.livinglab.pt) is a Portuguese industry-aca... more Living Usability Lab for Next Generation Networks (www.livinglab.pt) is a Portuguese industry-academia collaborative R&D project, active in the field of live usability testing, focusing on the development of technologies and services to support healthy, productive and active citizens. The project adopts the principles of universal design and natural user interfaces (speech, gesture) making use of the benefits of next generation networks and distributed computing. Therefore, it will have impact on the general population, including the elderly and citizens with permanent or situational special needs. This paper presents project motivations, conceptual model, architecture and work in progress.
Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems
Page 1. leigh, NC July 7-10, 1992 Proceedings of the 1992 EEE,RSJ International Conference on Int... more Page 1. leigh, NC July 7-10, 1992 Proceedings of the 1992 EEE,RSJ International Conference on Intelligent Robots and Systems Ship noise evaluation based on segmented decision trees Jose M. Fonseca [email protected] Fernando Moura-Pires fmp@ fct.unl.pt ...
The automatic recognition of noise situations in an underwater environment is an interesting issu... more The automatic recognition of noise situations in an underwater environment is an interesting issue, especially in coastal waters. In fact, recognition of ships by their acoustic signature captured by passive sonars can be very useful for automatic statistic tasks. This is a complex problem involving a large amount of signals with unstable characteristics. To achieve our goal, an artificial intelligence based approach was adopted in order to extract results from such large amounts of data. Traditional machine learning and neural network techniques were used and compared in their performance to select a technology able to produce a good classifier. Based on the neural network technology, a real-time classifier, using low cost hardware was developed. The adopted architecture is explained and the friendly user interface is presented.
We propose and evaluate an automatic segmentationmethod for extracting striatal brain structures ... more We propose and evaluate an automatic segmentationmethod for extracting striatal brain structures (caudate, putamen, and ventral striatum) from parametric 11C-raclopride positron emission tomography (PET) brain images. We focus on the images acquired using a novel brain dedicated high-resolution (HRRT) PET scanner. The segmentation method first extracts the striatum using a deformable surface model and then divides the striatum into its substructures based on a graph partitioning algorithm. The weighted kernel k-means algorithm is used to partition the graph describing the voxel affinities within the striatum into the desired number of clusters. Themethod was experimentally validated with synthetic and real image data. The experiments showed that our method was able to automatically extract caudate, ventral striatum, and putamen from the images. Moreover, the putamen could be subdivided into anterior and posterior parts. An automatic method for the extraction of striatal structures f...
Journal of voice : official journal of the Voice Foundation, Jan 12, 2016
Speech signal processing techniques have provided several contributions to pathologic voice ident... more Speech signal processing techniques have provided several contributions to pathologic voice identification, in which healthy and unhealthy voice samples are evaluated. A less common approach is to identify laryngeal pathologies, for which the use of a noninvasive method for pathologic voice identification is an important step forward for preliminary diagnosis. In this study, a hierarchical classifier and a combination of systems are used to improve the accuracy of a three-class identification system (healthy, physiological larynx pathologies, and neuromuscular larynx pathologies). Three main subject classes were considered: subjects with physiological larynx pathologies (vocal fold nodules and edemas: 59 samples), subjects with neuromuscular larynx pathologies (unilateral vocal fold paralysis: 59 samples), and healthy subjects (36 samples). The variables used in this study were a speech task (sustained vowel /a/ or continuous reading speech), features with or without perceptual info...
Developing specialized software tools is essential to support studies of solar activity evolution... more Developing specialized software tools is essential to support studies of solar activity evolution. With new space missions such as Solar Dynamics Observatory (SDO), solar images are being produced in unprecedented volumes. To capitalize on that huge data availability, the scientific community needs a new generation of software tools for automatic and efficient data processing. In this paper a prototype of a modular framework for solar feature detection, characterization, and tracking is presented. To develop an efficient system capable of automatic solar feature tracking and measuring, a hybrid approach combining specialized image processing, evolutionary optimization, and soft computing algorithms is being followed. The specialized hybrid algorithm for tracking solar features allows automatic feature tracking while gathering characterization details about the tracked features. The hybrid algorithm takes advantages of the snake model, a specialized image processing algorithm widely used in applications such as boundary delineation, image segmentation, and object tracking. Further, it exploits the flexibility and efficiency of Particle Swarm Optimization (PSO), a stochastic population based optimization algorithm. PSO has been used successfully in a wide range of applications including combinatorial optimization, control, clustering, robotics, scheduling, and image processing and video analysis applications. The proposed tool, denoted PSO-Snake model, was already successfully tested in other works for tracking sunspots and coronal bright points. In this work, we discuss the application of the PSO-Snake algorithm for calculating the sidereal rotational angular velocity of the solar corona. To validate the results we compare them with published manual results performed by an expert.
In developed countries, the prevalence of infertility ranges from 3.5% to 16.7%. There are severa... more In developed countries, the prevalence of infertility ranges from 3.5% to 16.7%. There are several factors that affect the success rate of in vitro treatments and so every couple has a singular probability of success which can be predicted. As these treatments are complex and expensive with a variable probability of success, the most common question asked by in vitro fertilization patients is ‘‘What are my chances of conceiving?”. Classical statistics and artificial intelligence models have been published in the literature. So far, artificial intelligent prediction models are not aimed at live birth but rather at pregnancy and use undergoing treatment features. The main aim of this study is to develop a classification tree model that estimates the chance of a live birth before In Vitro Fertilization (IVF) treatments. This decision tree might result in a new clinical support system that helps physicians to deal with the couple's expectations. Keywords-artificial intelligence; dec...
The aim of this study was to build an Artificial Neural Network (ANN) complemented by a decision ... more The aim of this study was to build an Artificial Neural Network (ANN) complemented by a decision tree to predict the chance of live birth after an In Vitro Fertilization (IVF)/Intracytoplasmic Sperm Injection (ICSI) treatment, before the first embryo transfer, using demographic and clinical data. Overall, 26 demographic and clinical data from 1193 cycles who underwent an IVF/ICSI treatment at Centro de Infertilidade e Reprodução Medicamente Assistida, between 2012 and 2019, were analyzed. An ANN was constructed by selecting experimentally the input variables which most correlated to the target through Pearson correlation. The final used variables were: woman’s age, total dose of gonadotropin, number of eggs, number of embryos and Antral Follicle Count (AFC). A decision tree was developed considering as an initial set the input variables integrated in the previous model. The ANN model was validated by the holdout method and the decision tree model by the 10-fold cross method. The ANN...
Temperature shifts trigger genome-wide changes in Escherichia coli's gene expression. We studied ... more Temperature shifts trigger genome-wide changes in Escherichia coli's gene expression. We studied if chromosome integration impacts on a gene's sensitivity to these shifts, by comparing the single-RNA production kinetics of a P LacO3O1 promoter, when chromosomally-integrated and when singlecopy plasmid-borne. At suboptimal temperatures their induction range, fold change, and response to decreasing temperatures are similar. At critically low temperatures, the chromosome-integrated promoter becomes weaker and noisier. Dissection of its initiation kinetics reveals longer lasting states preceding open complex formation, suggesting enhanced supercoiling buildup. Measurements with Gyrase and Topoisomerase I inhibitors suggest hindrance to escape supercoiling buildup at low temperatures. Consistently, similar phenomena occur in energy-depleted cells by DNP at 30 °C. Transient, critically-low temperatures have no long-term consequences, as raising temperature quickly restores transcription rates. We conclude that the chromosomally-integrated P LacO3O1 has higher sensitivity to low temperatures, due to longer-lasting super-coiled states. A lesser active, chromosomeintegrated native lac is shown to be insensitive to Gyrase overexpression, even at critically low temperatures, indicating that the rate of escaping positive supercoiling buildup is temperature and transcription rate dependent. A genome-wide analysis supports this, since cold-shock genes exhibit atypical supercoiling-sensitivities. This phenomenon might partially explain the temperature-sensitivity of some transcriptional programs of E. coli. Escherichia coli has evolved sophisticated regulatory programs to adapt to fluctuating environments that allow tuning gene expression so as to trigger appropriate responses 1,2. In general, gene expression regulation occurs during transcription initiation 3 and it can be performed, e.g., by transcription factors 4,5 , which act locally, affecting specific genes, and by σ factors 6-9 , which have more genome-wide effects. Similarly, environmental changes can affect chromosomal DNA compaction, which is associated to supercoiling 10,11 and is regulated by nucleoid associated proteins (NAPs) 12,13. Interestingly, changes in DNA compaction has genome-wide effects 13-15 , causing the expression of some genes to increase while in others it decreases 4,16-18. DNA compaction and supercoiling have distinct effects on plasmid-borne and chromosome integrated genes (see e.g. 19). One reason for this is that the chromosome has topologically constrained segments that allow supercoiling buildup 12,20-22 , as transcription occurs, since this process generates positive supercoiling ahead of the RNA polymerase (RNAP) and negative supercoiling behind it 23,24. Meanwhile, plasmids lack discrete constraints. Thus, when positive and negative supercoiling emerge, they freely diffuse in opposite directions and annihilate each other 19. Thus, in general, the transcriptional activity in plasmids is only affected by transient constraints due to, e.g., transient protein binding 19,25. Exceptions are, e.g., plasmids encoding membrane-associated proteins that, by anchoring to the membrane 26-29 , can form longer lasting constraints. Other exceptions are plasmids
Proceedings of the 2018 Federated Conference on Computer Science and Information Systems, 2018
The information generated by a computer vision system capable of labelling a land surface as wate... more The information generated by a computer vision system capable of labelling a land surface as water, vegetation, soil or other type, can be used for mapping and decision making. For example, an unmanned aerial vehicle (UAV) can use it to find a suitable landing position or to cooperate with other robots to navigate across an unknown region. Previous works on terrain classification from RGB images taken onboard of UAVs shown that only static pixel-based features were tested with a considerable classification error. This paper proposes a robust and efficient computer vision algorithm capable of classifying the terrain from RGB images with improved accuracy. The algorithm complement the static image features with dynamic texture patterns produced by UAVs rotors downwash effect (visible at lower altitudes) and machine learning methods to classify the underlying terrain. The system is validated using videos acquired onboard of a UAV.
Unmanned Aerial Vehicles (UAVs), although hardly a new technology, have recently gained a promine... more Unmanned Aerial Vehicles (UAVs), although hardly a new technology, have recently gained a prominent role in many industries being widely used not only among enthusiastic consumers, but also in high demanding professional situations, and will have a massive societal impact over the coming years. However, the operation of UAVs is fraught with serious safety risks, such as collisions with dynamic obstacles (birds, other UAVs, or randomly thrown objects). These collision scenarios are complex to analyze in real-time, sometimes being computationally impossible to solve with existing State of the Art (SoA) algorithms, making the use of UAVs an operational hazard and therefore significantly reducing their commercial applicability in urban environments. In this work, a conceptual framework for both stand-alone and swarm (networked) UAVs is introduced, with a focus on the architectural requirements of the collision avoidance subsystem to achieve acceptable levels of safety and reliability. T...
Journal of Automation, Mobile Robotics and Intelligent Systems, 2019
�nowing how to iden�fy terrain types is especially important in the autonomous naviga�on, mapping... more �nowing how to iden�fy terrain types is especially important in the autonomous naviga�on, mapping, decision making and emergency landings areas. For example, an unmanned aerial vehicle (UAV) can use it to find a suitable landing posi�on or to cooperate with other robots to navigate across an unknown region. Previous works on terrain classifica�on from RGB images taken onboard of UAVs shown that only sta�c pixel-based features were tested with a considerable classifica�on error. This paper presents a computer vision algorithm capable of iden�fying the terrain from RGB images with improved accuracy. The algorithm complement the sta�c image features and dynamic texture pa�erns produced by UAVs rotors downwash e�ect (visible at lower al�tudes) and machine learning methods to classify the underlying terrain. The system is validated using videos acquired onboard of a UAV with a RGB camera.
The actual economic situation, caused the change in the way that many people move on the cities, ... more The actual economic situation, caused the change in the way that many people move on the cities, due to this fact a big majority of the travelers, began to use public means of transport, both because these means are comfortable enough to make a day by day journey and because they are much more cheaper. However, there are routes where you need to use different means of public transport, on those cases improve the time management is crucial for the user, or he might arrive late on his destination. Countries in Europe like Belgium and Holland, have adopted intermodal route planners that gather different types of information, both form public and private means of transport, and also give tools to the user, for him to plan the journey with the less effort possible. The aim of this work is to analyze if this kind of platform could be introduced in Portugal. So a survey was made, in order to find what functionalities the user would hope to find in an application that could help him to plan a journey using public means of transport. At technical level other analysis was made in order to integrate the different tools that an intermodal route planner, could or should have, to help the user on the planification of a journey using public means of transport.
2016 International Conference on ReConFigurable Computing and FPGAs (ReConFig), 2016
Our work proposes a hardware architecture for a Long Short-Term Memory (LSTM) Neural Network, aim... more Our work proposes a hardware architecture for a Long Short-Term Memory (LSTM) Neural Network, aiming to outperform software implementations, by exploiting its inherent parallelism. The main design decisions are presented, along with the proposed network architecture. A description of the main building blocks of the network is also presented. The network is synthesized for various sizes and platforms, and the performance results are presented and analyzed. Our synthesized network achieves a 251 times speed-up over a custom-built software network, running on an i7-3770k Desktop computer, proving the benefits of parallel computation for this kind of network.
2019 International Young Engineers Forum (YEF-ECE), 2019
The AIA (Atmospheric Imaging Assembly) instrument, on-board the SDO (Solar Dynamics Observatory) ... more The AIA (Atmospheric Imaging Assembly) instrument, on-board the SDO (Solar Dynamics Observatory) satellite, provides high-resolution and high-cadence solar images since 2010. To extract scientific knowledge from those high-resolution images there is a need for efficient automatic web tools to detect and/or track the coronal bright points (CBPs). Identifying and tracking CBPs is essential for successfully calculate the solar corona rotation rate at different latitudes. Over the last years this topic has been an area of research in solar physics and some effective methods have been developed. The purpose of this work is to design an automatic and near real-time web tool that detects and tracks CBPs on solar images and publishes the results online, thus, allowing search and visualization functionalities to support astrophysicists perform improved solar analysis. The detection process uses a gradient based segmentation algorithm that has proved to provide accurate data about CBPs' dynamics. The paper propose to extend that work by using SunPy and OpenCV in Python, together with the Gradient Path Labeling (GPL) segmentation algorithm. The results obtained display good approximations, when compared with the ones from other authors, and the tool also seems reliable over long testing periods.
Promoters in Escherichia coli include an ‘OFF’ state, during which transcription is halted. Here,... more Promoters in Escherichia coli include an ‘OFF’ state, during which transcription is halted. Here, we propose a novel empirical method for assessing the time-length spent by promoters in this state. It relies on direct measurements of RNA production kinetics at the single molecule level at different induction levels, followed by an estimation of the RNA production rate under infinite induction, which is then compared to this rate under real, maximum induction. We apply it to the LacO3O1 promoter and infer that, under full induction, on average, 15% of the time between successful transcription events is spent in the OFF state. We verify this result by comparing the kinetics of a mutant strain lacking repressor molecules with that of the inferred rate under infinite induction. We expect this strategy of dissecting the kinetics of transcription repression to be applicable to a wide number of promoters in E. coli.
2019 IEEE Congress on Evolutionary Computation (CEC), 2019
Nowadays, the number of aerial unmanned vehicles (UAVs) is growing at a tremendous speed, as well... more Nowadays, the number of aerial unmanned vehicles (UAVs) is growing at a tremendous speed, as well as its technology. Therefore, it is essential to follow this growth with increasingly robust algorithms to be possible to exist cooperation between robots autonomously. One of the major events currently being developed in autonomous cooperation is relatively terrain classification where, this classification, is mainly important for emergency landings, mapping and decision making. This paper presents a robust computer vision system to sort terrain types using two main algorithms: Particle Swarm Optimization (PSO) and Gray-Level Co-Occurrence Matrix (GLCM). In addition to these two algorithms, a neural network was designed with the aim of increasing the probability of success of the proposed system. In order to evaluate this article, the system is validated using videos acquired onboard of a UAV with a RGB camera.
2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, 2021
Wildfires are recurrent natural disasters in some regions of the globe, being Portugal one of the... more Wildfires are recurrent natural disasters in some regions of the globe, being Portugal one of these areas. The Portuguese Institute for Nature Conservation and Forests implemented a national fuel break (FB) network, with the goal of decreasing fire hazard. FBs are areas of reduced fuel load that slow down fire spread, creating firefighting opportunities. Its effectiveness relies on periodic treatments to maintain the fuel load in levels that can reduce the effects of wildfires. This paper proposes a methodology to assess the FB state, according to its fuel load, based on the analysis of the inter-annual variability of Sentinel-2 NDVI time series. Inter-annual comparison allows it to adapt to different regions. To assess the reliability of NDVI data to evaluate the FB state, a linear regression with the vegetation height acquired by the Global Ecosystem Dynamics Investigation (GEDI) mission was tested, achieving determination coefficients between 0.57 and 0.98.
Cell division in Escherichia coli is morphologically symmetric due to, among other, these cells&#... more Cell division in Escherichia coli is morphologically symmetric due to, among other, these cells' ability to place the Z-ring at midcell. Studies have reported that, at sub-optimal temperatures, this symmetry decreases at the single-cell level, but the causes remain unclear. Using fluorescence microscopy, we observe FtsZ-GFP and DAPI-stained nucleoids to assess the robustness of the symmetry of Z-ring formation and positioning in individual cells under sub-optimal and critical temperatures. We find the Z-ring formation and positioning to be robust at sub-optimal temperatures, as the Z-ring's mean width, density and displacement from midcell maintain similar levels of correlation to one another as in optimal temperatures. However, at critical temperatures, the Z-ring displacement from midcell is much increased. We show evidence that this is due to enhanced distance between the replicated nucleoids and, thus, reduced Z-ring density, which explains the weaker precision in settin...
Proceedings of the 1st International Living Usability Lab Workshop on AAL Latest Solutions, Trends and Applications, 2011
Living Usability Lab for Next Generation Networks (www.livinglab.pt) is a Portuguese industry-aca... more Living Usability Lab for Next Generation Networks (www.livinglab.pt) is a Portuguese industry-academia collaborative R&D project, active in the field of live usability testing, focusing on the development of technologies and services to support healthy, productive and active citizens. The project adopts the principles of universal design and natural user interfaces (speech, gesture) making use of the benefits of next generation networks and distributed computing. Therefore, it will have impact on the general population, including the elderly and citizens with permanent or situational special needs. This paper presents project motivations, conceptual model, architecture and work in progress.
Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems
Page 1. leigh, NC July 7-10, 1992 Proceedings of the 1992 EEE,RSJ International Conference on Int... more Page 1. leigh, NC July 7-10, 1992 Proceedings of the 1992 EEE,RSJ International Conference on Intelligent Robots and Systems Ship noise evaluation based on segmented decision trees Jose M. Fonseca [email protected] Fernando Moura-Pires fmp@ fct.unl.pt ...
The automatic recognition of noise situations in an underwater environment is an interesting issu... more The automatic recognition of noise situations in an underwater environment is an interesting issue, especially in coastal waters. In fact, recognition of ships by their acoustic signature captured by passive sonars can be very useful for automatic statistic tasks. This is a complex problem involving a large amount of signals with unstable characteristics. To achieve our goal, an artificial intelligence based approach was adopted in order to extract results from such large amounts of data. Traditional machine learning and neural network techniques were used and compared in their performance to select a technology able to produce a good classifier. Based on the neural network technology, a real-time classifier, using low cost hardware was developed. The adopted architecture is explained and the friendly user interface is presented.
We propose and evaluate an automatic segmentationmethod for extracting striatal brain structures ... more We propose and evaluate an automatic segmentationmethod for extracting striatal brain structures (caudate, putamen, and ventral striatum) from parametric 11C-raclopride positron emission tomography (PET) brain images. We focus on the images acquired using a novel brain dedicated high-resolution (HRRT) PET scanner. The segmentation method first extracts the striatum using a deformable surface model and then divides the striatum into its substructures based on a graph partitioning algorithm. The weighted kernel k-means algorithm is used to partition the graph describing the voxel affinities within the striatum into the desired number of clusters. Themethod was experimentally validated with synthetic and real image data. The experiments showed that our method was able to automatically extract caudate, ventral striatum, and putamen from the images. Moreover, the putamen could be subdivided into anterior and posterior parts. An automatic method for the extraction of striatal structures f...
Journal of voice : official journal of the Voice Foundation, Jan 12, 2016
Speech signal processing techniques have provided several contributions to pathologic voice ident... more Speech signal processing techniques have provided several contributions to pathologic voice identification, in which healthy and unhealthy voice samples are evaluated. A less common approach is to identify laryngeal pathologies, for which the use of a noninvasive method for pathologic voice identification is an important step forward for preliminary diagnosis. In this study, a hierarchical classifier and a combination of systems are used to improve the accuracy of a three-class identification system (healthy, physiological larynx pathologies, and neuromuscular larynx pathologies). Three main subject classes were considered: subjects with physiological larynx pathologies (vocal fold nodules and edemas: 59 samples), subjects with neuromuscular larynx pathologies (unilateral vocal fold paralysis: 59 samples), and healthy subjects (36 samples). The variables used in this study were a speech task (sustained vowel /a/ or continuous reading speech), features with or without perceptual info...
Developing specialized software tools is essential to support studies of solar activity evolution... more Developing specialized software tools is essential to support studies of solar activity evolution. With new space missions such as Solar Dynamics Observatory (SDO), solar images are being produced in unprecedented volumes. To capitalize on that huge data availability, the scientific community needs a new generation of software tools for automatic and efficient data processing. In this paper a prototype of a modular framework for solar feature detection, characterization, and tracking is presented. To develop an efficient system capable of automatic solar feature tracking and measuring, a hybrid approach combining specialized image processing, evolutionary optimization, and soft computing algorithms is being followed. The specialized hybrid algorithm for tracking solar features allows automatic feature tracking while gathering characterization details about the tracked features. The hybrid algorithm takes advantages of the snake model, a specialized image processing algorithm widely used in applications such as boundary delineation, image segmentation, and object tracking. Further, it exploits the flexibility and efficiency of Particle Swarm Optimization (PSO), a stochastic population based optimization algorithm. PSO has been used successfully in a wide range of applications including combinatorial optimization, control, clustering, robotics, scheduling, and image processing and video analysis applications. The proposed tool, denoted PSO-Snake model, was already successfully tested in other works for tracking sunspots and coronal bright points. In this work, we discuss the application of the PSO-Snake algorithm for calculating the sidereal rotational angular velocity of the solar corona. To validate the results we compare them with published manual results performed by an expert.
In developed countries, the prevalence of infertility ranges from 3.5% to 16.7%. There are severa... more In developed countries, the prevalence of infertility ranges from 3.5% to 16.7%. There are several factors that affect the success rate of in vitro treatments and so every couple has a singular probability of success which can be predicted. As these treatments are complex and expensive with a variable probability of success, the most common question asked by in vitro fertilization patients is ‘‘What are my chances of conceiving?”. Classical statistics and artificial intelligence models have been published in the literature. So far, artificial intelligent prediction models are not aimed at live birth but rather at pregnancy and use undergoing treatment features. The main aim of this study is to develop a classification tree model that estimates the chance of a live birth before In Vitro Fertilization (IVF) treatments. This decision tree might result in a new clinical support system that helps physicians to deal with the couple's expectations. Keywords-artificial intelligence; dec...
The aim of this study was to build an Artificial Neural Network (ANN) complemented by a decision ... more The aim of this study was to build an Artificial Neural Network (ANN) complemented by a decision tree to predict the chance of live birth after an In Vitro Fertilization (IVF)/Intracytoplasmic Sperm Injection (ICSI) treatment, before the first embryo transfer, using demographic and clinical data. Overall, 26 demographic and clinical data from 1193 cycles who underwent an IVF/ICSI treatment at Centro de Infertilidade e Reprodução Medicamente Assistida, between 2012 and 2019, were analyzed. An ANN was constructed by selecting experimentally the input variables which most correlated to the target through Pearson correlation. The final used variables were: woman’s age, total dose of gonadotropin, number of eggs, number of embryos and Antral Follicle Count (AFC). A decision tree was developed considering as an initial set the input variables integrated in the previous model. The ANN model was validated by the holdout method and the decision tree model by the 10-fold cross method. The ANN...
Temperature shifts trigger genome-wide changes in Escherichia coli's gene expression. We studied ... more Temperature shifts trigger genome-wide changes in Escherichia coli's gene expression. We studied if chromosome integration impacts on a gene's sensitivity to these shifts, by comparing the single-RNA production kinetics of a P LacO3O1 promoter, when chromosomally-integrated and when singlecopy plasmid-borne. At suboptimal temperatures their induction range, fold change, and response to decreasing temperatures are similar. At critically low temperatures, the chromosome-integrated promoter becomes weaker and noisier. Dissection of its initiation kinetics reveals longer lasting states preceding open complex formation, suggesting enhanced supercoiling buildup. Measurements with Gyrase and Topoisomerase I inhibitors suggest hindrance to escape supercoiling buildup at low temperatures. Consistently, similar phenomena occur in energy-depleted cells by DNP at 30 °C. Transient, critically-low temperatures have no long-term consequences, as raising temperature quickly restores transcription rates. We conclude that the chromosomally-integrated P LacO3O1 has higher sensitivity to low temperatures, due to longer-lasting super-coiled states. A lesser active, chromosomeintegrated native lac is shown to be insensitive to Gyrase overexpression, even at critically low temperatures, indicating that the rate of escaping positive supercoiling buildup is temperature and transcription rate dependent. A genome-wide analysis supports this, since cold-shock genes exhibit atypical supercoiling-sensitivities. This phenomenon might partially explain the temperature-sensitivity of some transcriptional programs of E. coli. Escherichia coli has evolved sophisticated regulatory programs to adapt to fluctuating environments that allow tuning gene expression so as to trigger appropriate responses 1,2. In general, gene expression regulation occurs during transcription initiation 3 and it can be performed, e.g., by transcription factors 4,5 , which act locally, affecting specific genes, and by σ factors 6-9 , which have more genome-wide effects. Similarly, environmental changes can affect chromosomal DNA compaction, which is associated to supercoiling 10,11 and is regulated by nucleoid associated proteins (NAPs) 12,13. Interestingly, changes in DNA compaction has genome-wide effects 13-15 , causing the expression of some genes to increase while in others it decreases 4,16-18. DNA compaction and supercoiling have distinct effects on plasmid-borne and chromosome integrated genes (see e.g. 19). One reason for this is that the chromosome has topologically constrained segments that allow supercoiling buildup 12,20-22 , as transcription occurs, since this process generates positive supercoiling ahead of the RNA polymerase (RNAP) and negative supercoiling behind it 23,24. Meanwhile, plasmids lack discrete constraints. Thus, when positive and negative supercoiling emerge, they freely diffuse in opposite directions and annihilate each other 19. Thus, in general, the transcriptional activity in plasmids is only affected by transient constraints due to, e.g., transient protein binding 19,25. Exceptions are, e.g., plasmids encoding membrane-associated proteins that, by anchoring to the membrane 26-29 , can form longer lasting constraints. Other exceptions are plasmids
Proceedings of the 2018 Federated Conference on Computer Science and Information Systems, 2018
The information generated by a computer vision system capable of labelling a land surface as wate... more The information generated by a computer vision system capable of labelling a land surface as water, vegetation, soil or other type, can be used for mapping and decision making. For example, an unmanned aerial vehicle (UAV) can use it to find a suitable landing position or to cooperate with other robots to navigate across an unknown region. Previous works on terrain classification from RGB images taken onboard of UAVs shown that only static pixel-based features were tested with a considerable classification error. This paper proposes a robust and efficient computer vision algorithm capable of classifying the terrain from RGB images with improved accuracy. The algorithm complement the static image features with dynamic texture patterns produced by UAVs rotors downwash effect (visible at lower altitudes) and machine learning methods to classify the underlying terrain. The system is validated using videos acquired onboard of a UAV.
Unmanned Aerial Vehicles (UAVs), although hardly a new technology, have recently gained a promine... more Unmanned Aerial Vehicles (UAVs), although hardly a new technology, have recently gained a prominent role in many industries being widely used not only among enthusiastic consumers, but also in high demanding professional situations, and will have a massive societal impact over the coming years. However, the operation of UAVs is fraught with serious safety risks, such as collisions with dynamic obstacles (birds, other UAVs, or randomly thrown objects). These collision scenarios are complex to analyze in real-time, sometimes being computationally impossible to solve with existing State of the Art (SoA) algorithms, making the use of UAVs an operational hazard and therefore significantly reducing their commercial applicability in urban environments. In this work, a conceptual framework for both stand-alone and swarm (networked) UAVs is introduced, with a focus on the architectural requirements of the collision avoidance subsystem to achieve acceptable levels of safety and reliability. T...
Journal of Automation, Mobile Robotics and Intelligent Systems, 2019
�nowing how to iden�fy terrain types is especially important in the autonomous naviga�on, mapping... more �nowing how to iden�fy terrain types is especially important in the autonomous naviga�on, mapping, decision making and emergency landings areas. For example, an unmanned aerial vehicle (UAV) can use it to find a suitable landing posi�on or to cooperate with other robots to navigate across an unknown region. Previous works on terrain classifica�on from RGB images taken onboard of UAVs shown that only sta�c pixel-based features were tested with a considerable classifica�on error. This paper presents a computer vision algorithm capable of iden�fying the terrain from RGB images with improved accuracy. The algorithm complement the sta�c image features and dynamic texture pa�erns produced by UAVs rotors downwash e�ect (visible at lower al�tudes) and machine learning methods to classify the underlying terrain. The system is validated using videos acquired onboard of a UAV with a RGB camera.
This thesis assesses decision tree algorithms, covering very popular ones such as ID3, C4.5 and C... more This thesis assesses decision tree algorithms, covering very popular ones such as ID3, C4.5 and CART. An original decision tree algorithm is proposed. After a brief introduction, this thesis presents the justification for selecting decision trees for the processing of large quantities of data and automatic classifiers generation. Then, a selection of techniques for the generation of decision trees is presented. After this theoretical overview, HistClass, a non-parametric, induction algorithm is introduced. This algorithm, designed essentially for the processing of large problems involving large amounts of data and containing discrete and/or continuous attributes is described in detail. Further, the software specifically developed for evaluation of the algorithm performance is discussed. Finally, the algorithms ID3, CART, C4.5 and HistClass are compared from the point of view of performance. For the performance evaluation, a complete set of typical, public domain, machine learning examples is used. These examples allow the comparison of the results obtained here and others developed in this area.
Uploads
Papers by Jose Fonseca
After a brief introduction, this thesis presents the justification for selecting decision trees for the processing of large quantities of data and automatic classifiers generation. Then, a selection of techniques for the generation of decision trees is presented. After this theoretical overview, HistClass, a non-parametric, induction algorithm is introduced. This algorithm, designed essentially for the processing of large problems involving large amounts of data and containing discrete and/or continuous attributes is described in detail. Further, the software specifically developed for evaluation of the algorithm performance is discussed.
Finally, the algorithms ID3, CART, C4.5 and HistClass are compared from the point of view of performance. For the performance evaluation, a complete set of typical, public domain, machine learning examples is used. These examples allow the comparison of the results obtained here and others developed in this area.