It offers exciting and practical examples of the use of intelligent algorithms in ambient and biomedical computing. It contains topics such as bioscience computing, database design, machine consciousness, scheduling, video summarization, audio classification, semantic reasoning, machine learning, tracking and localization, secure computing, and communication. Read more Read less. Amazon Global Store US International products have separate terms, are sold from abroad and may differ from local products, including fit, age ratings, and language of product, labeling or instructions.
Manufacturer warranty may not apply Learn more about Amazon Global Store. From the Back Cover The rapid growth in electronic systems in the past decade has boosted research in the area of computational intelligence. No customer reviews. Share your thoughts with other customers. Krose Bayesian networks for dynamic systems analysis Localization of a mobile platform Tracking with distributed cameras Conclusions and remaining issues References Preliminaries Secure approximate matching w.
Hamming distance Conclusion References Contents Shin Station scheduler Local scheduler LS Numerical and simulation results Experimental setup and results Conclusions References ix Motivation Introduction to Reed Solomon Channel model Architecture design options Results Benchmarking Optimisations to design Qualcomm Standards Engineering Dept. As it has become increasingly easy to generate, collect, transport, process, and store huge amounts of data, the role of intelligent algorithms has become prominent in order to visualize, manipulate, retrieve, and interpret the data.
For instance, intelligent search techniques have been developed to search for relevant items in huge collections of web pages, and data mining and interpretation techniques play a very important role in making sense out of huge amounts of biomolecular measurements. As a result, the added value of many modern systems is no longer determined by hardware only, but increasingly by the intelligent software that supports and facilitates the user in realizing his or her objectives. Over the past years, considerable progress has been made in the area of computational intelligence, which can be positioned at the intersection of computer science, discrete mathematics, and cognitive science.
This has led to a growing community of practitioners within Philips Research that develop, analyze, and apply intelligent algorithms. The Symposium on Intelligent Algorithms SOIA intends to provide this community of practitioners with a platform to exchange information. Again a selection of papers was edited, resulting in the present book. It consists of 17 chapters, divided over three parts corresponding to the strategic topics mentioned above. Below we present more detailed information about the individual chapters. In Chapter 1, Chris Clack discusses the topic of modeling biological systems, thus allowing to perform in-silico experiments by means of computer simulation, to formulate hypotheses.
In Chapter 2, Nevenka Dimitrova gives an overview of the reverse approach, where one does not use computers to simulate biological processes, but where one uses biology to perform computations, in DNA computing and synthetic biology. In Chapter 3, Martin Kersten and Arno Siebes discuss data management inspired by biology, resulting in an organic database system. In Chapter 4, Kees van Zon discusses how to achieve machine consciousness, and how it can be applied. Par It I consists of eight chapters, addressing problems from the area of content management and retrieval.
In Chapter 5, Wim Verhaegh discusses the problem of making a schedule of preferred TV programs, while at the same time selecting TV programs for recording, under the assumption of a limited number of tuners. In Chapter 6, Mauro Barbieri, Nevenka Dimitrova, and Lalitha Agnihotri present a technique to automatically summarize video into a condensed preview, allowing one to quickly browse and access large amounts of stored programs.
Chapters 7—9 concerns audio applications. In Chapter 8, Steffen Pauws presents a technique to automatically extract the key from a piece of music, providing an emotional connotation to it, and making it possible to build well-sounding music mixes. In Chapter 9, Zharko Aleksovski, Warner ten Kate, and Frank van Harmelen address the problem of combining multiple databases of music data in a semantic way, by approximating matches of music classes.
Part III consists of six chapters, focusing on the technology underlying intelligent algorithms and intelligent systems. Finally, Chapters 16 and 17 address resource issues in intelligent systems. In Chapter 16, Sai Shankar N. Finally, in Chapter 17, Akash Kumar and Sergei Sawitzki discuss the design alternatives of Reed Solomon decoders, and address the problem of making optimal design decisions to obtain a high-throughput, low-power solution. Clack Abstract Bioscience computing exploits the synergy of challenges facing both computer science and biology, drawing inspiration from biology to solve computer science challenges and simultaneously using new bio-inspired adaptive software to model and simulate biological systems.
Keywords Bioscience computing, systems biology, computational simulation, morphogenesis, adaptive systems, agent based modelling, swarm agents. There has recently been a substantial increase in inter-disciplinary research interactions between computer science and the life sciences. Verhaegh et al.
The traditional role of computer science in biology e. The traditional view is giving way to a new biology, often referred to as systems biology. The rise of systems biology has caused a much closer relationship to develop between biologists and computer scientists. In systems biology, the computer science techniques are no longer merely a data service to the biologists, but are intimately involved in the formulation of biological hypotheses as biologists embrace the process-oriented world of the computer scientist.
Biologists now experiment not just in-vivo and in-vitro, but increasingly in-silico. These in-silico experiments are the basis for what we term bioscience computing. Note, however, that an in-silico experiment itself can never truly be used to test a biological hypothesis — rather, computational simulation in biology should be viewed as a process of prototyping to assist hypothesis formulation. Wet-lab experimental techniques tend to focus analytic attention on single mechanisms. By contrast, computational simulation can contribute to the activity of synthesis, of integrating many separate elements that form a network of activity.
The resultant interaction and synergy can provide a qualitatively much improved experimental framework. These in-silico results may then guide the choice of more expensive subsequent wet-lab experiments. At the mechanistic extreme there are cellular automata and agent-based simulations. Differential equations are widely used and capable of capturing detail at varying levels of abstraction.
See Figure 1. Markov chains Cellular automata Mechanistic local state adaptive Comparative spectrum of available techniques. Phenomenological models tend to focus on the global state of a system. Often they describe an a-priori given set of relations between an a-priori given set of variables [Giavitto et al. By contrast, mechanistic models provide local interaction modelling, where cells react often adaptively to a local environment, not to the state of the system as a whole thereby supporting heterogeneity.
This leads to a rich model of spatiotemporal dynamics, and offers insights into the 6 Christopher D. For example, where precise local effects due to intermolecular interactions and random molecular movement are required, a great number of equations must be generated and solved [Succi et al. In practice, the computational limits on solving a large number of related partial differential equations leads to the technique normally being applied only to abstractions of internal mechanisms and processes.
An interesting mechanistic approach is the use of cellular automata — e. The resulting system is very good at representing spatiotemporal dynamics and organisational behaviour, particularly for the simulation of adaptive behaviour. Objects and processes. Biologists in particular molecular biologists naturally focus on objects, interactions and processes. Computational simulation permits biologists to express biological systems in terms of computational objects, interactions and processes that relate directly to their biological counterparts and are therefore far easier to understand and easier to manipulate than differential equations.
Computational simulations can be expressed in terms of information networks and can use interaction-centric models e. The experience of systems biology has been that biologists have increasingly adopted the computational systems concepts of computer scientists.
This should not come as a surprise, since computer scientists have extensive experience of building, modelling, and simulating complex systems that require analysis and synthesis at many different levels of abstraction. At the lowest level, system components are lightweight agents governed by local-neighbourhood rules. The rules provide the system of dynamic interaction between agents, and from this comes the self-organising properties of the simulated organism threshold parameters may need to be derived via automatic search methods.
The emergent behaviour of the system is dependent on a combination of the competitive and co-operative interactions of the underlying localneighbourhood rules, the regulatory effects that arise from the self-organising properties of those rules, and sets of global constraints which may be derived from experimental observation. The result is a complex, dynamic system, which can itself be considered as an agent in a larger network of agents of similar complexity, each undergoing interactions according to local-neighbourhood rules at a higher level, and from which yet more complex behaviour emerges.
The dynamic emergence of hierarchies of biological complexity. While emergent behaviour has the potential for chaotic results, in a hierarchy of levels each can constrain the realisable solutions of the other 8 Christopher D. Clack levels — thus, an understanding of dynamic emergence in complex hierarchies is a fundamental step in understanding the underlying mechanisms of biology.
Organisms in nature exhibit complex adaptive behaviour that far surpasses the ability of current state-of-the-art autonomous software and robotics. Many unicell organisms exhibit complex adaptations of their shape in rapid response to environmental changes — e. We use agent-based swarm techniques combined with 3D cellular automaton CA rules to allow proteins to exist and interact with their 26 nearest neighbours in a 3D voxellated environment. The agent-based swarm technique permits the modelling and tracking of individual components and their interactions. The combination of the two techniques agent-based swarm and CA provides opportunities for optimizing computational overhead e.
The CA rules for chemical diffusion and agent interactions can be checked against current understanding of the biology. The underlying mechanism. Different cell behaviours may require different E-P maps. The following explanation of the underlying mechanism will focus on the E-P map for chemotaxis; see Figures 1. Each voxel in the cellular automaton contains one of the following units: 1.
Clack Figure 1. A generalized environment-phenotyope map. The cytoskeleton is affected by input from the environment Env via the transduction pathway TP and can affect the shape of the cell, and thereby also the environment. Figure 1. Bioscience Computing and Computational Simulation in Biology 11 3. The membrane separates the cell from the environment. Cell surface receptors are embedded in the membrane and mediate signals from the external environment to the cytoskeleton.
Membrane units containing receptors sum the concentration of C in their adjacent environment voxels. If the sum exceeds a threshold, a cascade reaction inside the cell is triggered; WASP and PIP2 are activated for the receptor and for its adjacent membrane voxels. The WASP proteins, when activated by a receptor, recruit agents nucleator and P1 actin to the membrane see below for a further explanation of recruitment. A recruited nucleator agent will switch on and recruited P1 actin changes state to P2 actin. The general rules are: 1. Diffusion: accessory proteins are represented as concentration gradients which diffuse through cytoplasm voxels.
Random movement: when not bound or stuck, an agent moves randomly. Receptors detect chemoattractant, WASP and PIP2 activate and cause the cytoskeletal behaviours shown in stages 1— 6, see text for details. It then can only move such that an S is still in its nearest neighbours. Recruitment stops if there is no S nearest neighbour. These are illustrated in Figure 1. Over time the nucleator disassociates and un-sticks from its AF and deactivates stage 1. There are three interactions affecting the cell membrane: 1. If a membrane unit has no contact with inner cellular units, it is removed becomes an environment unit ; this ensures there are no doubled-up layers of membrane.
Chemotaxis experiment. It moved towards the chemical source purely by lifetime adapation of shape: see Figure 1. Phagocytosis experiment. In nature, a single adaptive mechanism is able to provide different morphologies in response to different environmental stimuli. Clack For example, compare chemotaxis movement morphology with phagocytosis ingestion morphology : these two examples are distinct both topologically and functionally, yet are known to be controlled by the same underlying biological mechanism.
In chemotaxis, a cell detects a chemical gradient and transforms its morphology in order to follow it to the source. By contrast, phagocytosis is the process of engulfment of a foreign particle for degradation or ingestion [Castellano et al. The simulated morphology is shown in Figure 1.
Leading edge morphology top left during chemotaxis movement. Simulation of chemotaxis leading edge morphology. Phagocytic cup morphology the cup forms around the particle. Simulation of phagocytic cup morphology particle not shown. Bioscience Computing and Computational Simulation in Biology 1. An improved understanding of the internal mechanisms and organisational principles of adaptive behaviour and lifetime plasticity, especially adaptation of morphology, will provide foundational results that are applicable to many forms of adaptive response.
This includes improvements to synthetic systems such as autonomous software agents e. In the life sciences. Our initial work with the NHM was a study of the morphogenesis of diatoms single-celled algae , whose patterned cell walls are thought to be an adaptive response to their environment.
Diatoms are one of the most important groups of primary producers on the planet, which have thousands of forms and behaviours, each adapted to a different environment. The observable diatom cell wall morphologies are not explicable by the physics of diffusion alone; electron microscopy studies reveal that the 16 Christopher D. Clack cytoskeleton is intimately involved in the patterning of the cell wall and may also incorporate the use of cytoplasmic organelles as moulds for different cell wall components.
Our model generated representations of diatom cell walls that were, at each stage of development, consistent with empirical observations and exhibited some of the functions of diatom cell walls. More importantly, understanding of the mechanism of the cytoskeleton during morphogenesis was improved; e. We are currently seeking funding to conduct a further experiment to aid the NHM in the understanding of diatom colony behaviour.
Certain species of diatom have developed a complex set of interactions during morphogenesis, which allows them to form and disband colonies, triggered by environmental cues and giving them a greater chance of survival e. Diatom colony formation is an explicit and interesting example of morphological adaptation to environmental changes; it is a type of cyclomorphosis where adaptation cycles through two or more forms. There has been a large amount of speculation as to how and why certain species of diatom form colonies; it contributes to current understanding within diatom research, and also provides a good model to improve understanding of the hierarchical adaptive systems that underlie morphological plasticity.
Biologists have a computational, process-oriented understanding of their subject — they think in terms of objects, interactions and processes: processes are more important than the end result; dynamic behaviour is more impor- Bioscience Computing and Computational Simulation in Biology 17 tant than equilibrium; and behaviour and interactions of individual objects are important.
The process-oriented approach to simulating biological complexity leads to an increased understanding of dynamic emergence and regulatory interaction and control: this is a fundamental step towards a future theory of biology. References Alberts, B. Bray, J. Lewis, M. Raff, K. Roberts, and J. Watson . Molecular Biology of The Cell. Garland Publishing, 3rd edition. Anderson, A. Chaplain . Continuous and discrete mathematical models of tumour-induced angiogenesis. Bulletin of Mathematical Biology, — Araujo, R.
McElwain . A history of the study of solid tumour growth: The contribution of mathematical modelling. Bentley, K. Clack . Morphological plasticity: Environmentally driven morphogenesis. Cox, and P. Bentley . Nanoscience and Nanotechnology Journal, 5 1 — Castellano, F. Chavrier, and E. Caron . Actin dynamics during phagocytosis. Seminars in Immunology, — Condeelis, J. How is actin polymerization nucleated in vivo? Davey, M. Crawford . Filament formation in the diatom melosira granulata.
Journal of Phycology, — Gatenby, R. Maini . Mathematical oncology: Cancer summed up. Nature, — Clack Giavitto, J-L. Godin, O. Michel, and P. Prusunkiewicz . Glazier, J. Graner . Simulation of the differential adhesion driven rearrangement of biological cells. Physical Review E, 47 3 — Holt, M. Koffer . Cell motility: Proline-rich proteins promote protrusions. Ideker, T. Lauffenburger . Building with a scaffold: emerging stategies for highto low-level cellular modelling. Trends in Biotechnology, 21 6 — Kirkwood, T. Boys, C. Gillespie, C. Proctor, D. Shanley, and D. Wilkinson . Towards an e-biology of ageing: Integrating theory and data.
Nature Reviews, Molecular Cell Biology, — Noble, D. The rise of computational biology. Patel, M. Internal report. Priami C. Scalerandi, M. Capogrosso Sansone, C. Benati, and C. Condat . Competition effects in the dynamics of tumor cords. Physical Review E, 65 5 Pt 1 Succi, S. Karlin, and H. Chen . Colloquium: Role of the h theorem in lattice boltzmann hydrodynamics simulations. Reviews of Modern Physics, — Table 1.
Glossary of biological terms. The escape of substances from a cell to its environment. Migration of cells along a concentration gradient of an attractant. The self-replicating genetic structure of cells containing the cellular DNA that contains the linear array of genes.
Journal of Biomedical Informatics
The contents of a cell but not including the nucleus. A system of molecules within eukaryotic cells providing shape, internal spatial organization, and motility, and may assist in communication with other cells. Microscopic algae with cell walls made of silicon and of two separating halves.
A cell or organism with a membrane-bound nucleus and other subcellular compartments. Includes all organisms except viruses, bacteria, and bluegreen algae. Any material produced by cells and secreted into the surrounding medium. The properties of the ECM determine the properties of the tissue e. A cell found in most tissues of the body, involved in wound repair and closure; they migrate towards the wound site via chemotaxis.
The set of different gene alleles existing in an organism. A molecule composed of a very large number of atoms. Includes proteins, starches and nucleic acids e. The development and adaptation of the shape and form of an organism. A membrane-bound structure in a eukaryotic cell that partitions the cell into regions which carry out different cellular functions.
An inward folding of the cell membrane creating an interior pocket, formed by an actin dependent process during phagocytosis. The engulfment of a particle or a microorganism by leukocytes. The physical characteristics of an organism. The functional behavior of the physiological state of an individual or species, describing the physiological dynamics of the normal intact organism.
An organism with eukaryotic cells that is neither plant nor animal nor fungi. Wiskott-Aldrich syndrome protein. Regulates the formation of actin chains. DNA computing uses properties of biomolecules and techniques from molecular biology to perform computations, instead of using the traditional silicon-based computer technologies. To date experiments have been performed both in-vitro and in-vivo. Keywords DNA computing, aqueous computing, molecular computing.
Most prominently, genomics and proteomics have greatly improved our knowledge of the components of biological systems at the molecular level. Scientists have elucidated the complete gene sequences of several model organisms and provided general understanding of the molecular machinery involved in gene expression.
Now, all these advances have also facilitated a change in attitude. So the topic is to use biology with an engineering approach: to compute with molecules or to synthesize new reactions and organisms with the available biological knowledge. In this chapter we will 21 Wim F. We give an overview of DNA computing in Section 2. In Section 2. In , Adleman demonstrated a proof-of-concept use of DNA as form of computation that was used to solve the Hamiltonian path problem.
Since the initial Adleman experiments, DNA computing has made advances and has shown to have potential as a means to solve several large-scale combinatorial search problems. There has been research over onedimensional lengths, two-dimensional tiles, and even three-dimensional DNA graphs processing, self-assembling DNA graphs [Sa-Ardyen et al. A new term, natural computing, has been introduced to describe computing going on in nature and computing inspired by nature [Brauer et al. A P system is a computing model which abstracts from the way the alive cells process chemical compounds in their compartmental structure.
Benenson et al. Here we decided to present only a cross section of approaches to DNA computing: molecular computing, aqueous computing, and Turing machines. He showed how to solve a seven-node instance of the Hamiltonian Path problem, an NP-Complete problem similar to the traveling salesman problem. As the number of cities increases, the solution run time grows exponentially relative to the number of cities at which point the problem requires brute force search methods.
HPPs with a large number of cities quickly become computationally expensive, making them less than feasible to solve on even the latest super- or grid- computer. Adlemans demonstration only involves seven cities, making it in some sense a trivial problem. It is an example of computation at a molecular level, potentially a size limit that may never be reached by the semiconductor industry. In an innovative way the DNA is used as a data structure to encode symbols.
This is important because current hard disk drives have a capacity of GB. In research, Seagate has reached densities of 50 terabits Tb per square. In 50 terabits we can store over 3. The computing machinery works at molecular levels with the use only of DNA strands and enzymes. It demonstrated the possibility for massively parallel computation, as many enzymes can work on many DNA molecules simultaneously. Consider the example of Figure 2. For this example, the molecular solution is as follows.
Step 1. Represent the cities in the graph i. Generate all possible connections represented by edges in the graph using DNA hybridization. As shown in Figure 2.
Intelligent Algorithms in Ambient and Biomedical Computing (Philips Research Book Series)
Louis Figure 2. Phoenix An example HP graph. In this manner, all the different connections can be encoded. Step 3. Select itineraries that start and end with the correct city. In this step, the goal is to copy and amplify paths that start with Boston and end with Phoenix. To achieve this, polymerase chain reaction PCR is used, with primers that are complimentary to Boston and Phoenix. Step 4. Select itineraries with the correct number of cities.
- Supportive Care in Cancer Therapy?
- Table from Intelligent Algorithms in Ambient and Biomedical Computing - Semantic Scholar.
- Applied Microbiology.
- How To Control Your Brain At Will.
- Stability Analysis and Design of Structures.
- Intelligent Algorithms in Ambient and Biomedical Computing | SpringerLink?
In this process, so-called gel electrophoresis is used to select the chains with six cities. Step 5. Select itineraries that contain each city only once. Phoenix, as shown in Figure 2. We should note here that if there is more than one solution then all the solutions will be attracted. Step 6. He also introduced schemes for reading information from these molecules. Boston City and connection representation with DNA sequences. The wet-lab procedure for the aqueous computing has been carried out in the laboratory of Susannah Gal. This work is part of a world-wide search for information storage techniques and computational procedures that take advantage of the vast parallelism of biomolecular operations.
All known sequential solutions of the SAT problem, which is NP-complete, require a number of steps that grows exponentially in the size of the problem instance. The number of steps required by the aqueous algorithm grows only linearly. Another application of aqueous computing is the cardinality of a maximal independent subset of a graph has been computed and reported by Head et al.
The cardinality of a minimal dominating subset of a graph has also been computed by the same group. The aqueous approach suggests a convenient way to carry out computations in the style introduced by Lipton . Partition is the action of fast replication of the entire memory, i. We will call these crucial locations on the molecule the stations of the molecule. We should note here that this kind of abstract computation can be implemented in different ways.
Next, we discuss how the operations Partition, Unite, and Write can be orchestrated to provide a useful computation.
Arguing the Pros and Cons of Artificial Intelligence in Healthcare
We will do this for the maximum independent set problem. Let m be a molecule that possesses six stations, each admitting a write operation. We designate these six stations with the vertices a, b, c, d, e, f , respectively. With each memory molecule, m, we associate a non-negative integer P m. At each phase of the computation, the value of P m will be the number of stations that remain in their original condition, i.
We also suppose that the molecules can be sorted by this parameter and that the value of the parameter can be determined for each of the classes resulting from the sort. Next, we introduce the procedure. Partition T0 into tubes denoted as T1 and T2. In T1 write at station a; in T2 write at station b.
Partition T0 into tubes T1 and T2. In T1 write at station b; in T2 write at station c. Partition the contents of T0 into tubes T1 and T2. In T1 write at station c; in T2 write at station d. In T1 write at station d; in T2 write at station e. Sort the memory molecules m remaining in T0 according to their values P m. In the present case, the largest parameter value is expected to be attained by the molecules representing One can think of the above program as follows: a FOR loop that is traversed once for each edge of the graph G; followed by a Sort treated as a single step. Of course one limiting factor for graphs with large number of edges 28 Nevenka Dimitrova would be the number of molecules that are required to provide the solution.
The paradigm is to use DNA as software, and enzymes as hardware. The way in which these molecules undergo chemical reactions with each other allows simple operations to be performed as a byproduct of the reactions. The devices can be controlled by the composition of the DNA software molecules.
Of course this is a completely different approach as compared to pushing electrons around a dry circuit in a conventional computer. Turing machines. Alan Turing and Alonzo Church independently arrived at equivalent conclusions. In fact, a Turing machine TM is a very simple machine. Yet, a TM has the power of any digital computing machinery.
Special Issue: Ambient Intelligence for Health Environments (2016)
On curating multimodal sensory data for personalized health and wellness platforms. BioMedical Engineering OnLine , vol. Design, implementation and validation of a novel open framework for agile development of mobile health applications. S2:S6, pp. Mining Minds: an innovative framework for personalized health and wellness support. Multi-sensor fusion based on asymmetric decision weighting for robust activity recognition. Neural Processing Letters , vol. PhysioDroid: combining wearable health sensors and smartphones for a ubiquitous, continuous and personal monitoring.
The Scientific World Journal , vol. Dealing with the effects of sensor displacement in wearable activity recognition. Window size impact in activity recognition. Human activity recognition based on a sensor weighting hierarchical classifier. Soft Computing , vol. Amft, O. Evaluation of inertial sensor displacement effects in activity recognition systems. Damas, M. Improving wearable activity recognition via fusion of multiple equally-sized data subwindows.
Konsolakis, K. Human Behaviour Analysis Through Smartphones. Nijeweme-d'Hollosy, W. A study on the perceptions of autistic adolescents towards mainstream emotion recognition technologies. Moreno, S. Ramos-Monteon, J. Carrillo-Perez, F. Recher, F. Optimizing activity recognition in stroke survivors for wearable exoskeletons. Enabling remote assessment of cognitive behaviour through mobile experience sampling.