Search

New approaches to evidence synthesis, which use human effort and machine automation in mutually reinforcing ways, can enhance the feasibility and sustainability of living systematic reviews. Human effort is a scarce and valuable resource, required when automation is impossible or undesirable, and includes contributions from online communities (“crowds”) as well as more conventional contributions from review authors and information specialists. Automation can assist with some systematic review tasks, including searching, eligibility assessment, identification and retrieval of full-text reports, extraction of data, and risk of bias assessment. Workflows can be developed in which human effort and machine automation can each enable the other to operate in more effective and efficient ways, offering substantial enhancement to the productivity of systematic reviews. This paper describes and discusses the potential—and limitations—of new ways of undertaking specific tasks in living systematic reviews, identifying areas where these human/machine “technologies” are already in use, and where further research and development is needed. While the context is living systematic reviews, many of these enabling technologies apply equally to standard approaches to systematic reviewing.

In the previous chapter the short survey of the philosophical inquires and psychological research in the human visual perception was outlined.

Primary objective of machine perception MU is to construct the symbolic description of the visual content of an image and using this symbolic representation to solve the perceptual problems such as interpretation of perceived images. Symbolically represented visual knowledge provides a level of abstraction at which two otherwise dissimilar domains may look more alike. For example, the concepts of a planet and a ball are quite different, but if both are represented as a circle, it may facilitate analogical retrieval, mapping and transfer. The problem of perception and interpretation of images by application of the IN-perceptual transformation (described in Chap. 6), in order to find the solution to a perceptual problem, is solved within the framework of machine understanding. The machine understanding framework is referring to the human visual system that has a highly developed capability for interpretation of the visual data and detecting many classes of patterns based on statistically significant arrangements of image elements. These classes of patterns and statistically significant arrangements of image elements are called shapes.

As it was shown in the previous Chapter visual intelligence tests belong to the category of the visual problems. Visual intelligence tests are series of tasks designed to measure the capacity to make abstractions, to learn, and to deal with novel situations.

In the previous Chapters the shape classes, the 3D object classes and the picture classes, that are part of the hierarchical categorical structure of learned knowledge needed during solving the perceptual problems, were introduced.

In the previous Chapters the shape classes, the 3D object classes and the picture classes were presented.

In the previous chapters the shape classes, the 3D object classes and the picture classes, that are part of the hierarchical categorical structure of learned knowledge needed during solving the interpretational problems, were introduced.

Perceptual abilities enable us to go beyond the data that are in the image, since we can achieve reliable identification from a small subset of the predicted correspondences and then use our knowledge to infer many properties of the scene that may not be directly supported by visual data or solve other complex perceptual problems. Without constraining influence of prior expectations, many perceptual problems would be under constrained to the extent that they could never be solved. This emphasis on world knowledge parallels developments in most other areas of machine perception, in which large amounts of problem-specific knowledge are increasingly being used both to constrain solutions and to speed the process of obtaining the solution. In machine perception MU, the knowledge that is needed during solving perceptual problems is supplied by the contextual information that is coded in the categorical structure of the shape classes, the 3D object classes, the picture classes, and the perceptual and ontological categories of visual objects.

In the previous Chapter the shape classes that are the basic visual categories used in solving perceptual problems, were presented. The shape classes are part of visual knowledge represented in the form of the categorical structure of different categories such as the 3D object classes, the picture classes and the ontological categories.

While it is important for the evidence supporting practice guidelines to be current, that is often not the case. The advent of living systematic reviews has made the concept of "living guidelines" realistic, with the promise to provide timely, up-to-date and high-quality guidance to target users. We define living guidelines as an optimization of the guideline development process to allow updating individual recommendations as soon as new relevant evidence becomes available. A major implication of that definition is that the unit of update is the individual recommendation and not the whole guideline. We then discuss when living guidelines are appropriate, the workflows required to support them, the collaboration between living systematic reviews and living guideline teams, the thresholds for changing recommendations, and potential approaches to publication and dissemination. The success and sustainability of the concept of living guideline will depend on those of its major pillar, the living systematic review. We conclude that guideline developers should both experiment with and research the process of living guidelines.

A living systematic review (LSR) should keep the review current as new research evidence emerges. Any meta-analyses included in the review will also need updating as new material is identified. If the aim of the review is solely to present the best current evidence standard meta-analysis may be sufficient, provided reviewers are aware that results may change at later updates. If the review is used in a decision-making context, more caution may be needed. When using standard meta-analysis methods, the chance of incorrectly concluding that any updated meta-analysis is statistically significant when there is no effect (the type I error) increases rapidly as more updates are performed. Inaccurate estimation of any heterogeneity across studies may also lead to inappropriate conclusions. This paper considers four methods to avoid some of these statistical problems when updating meta-analyses: two methods, that is, law of the iterated logarithm and the Shuster method control primarily for inflation of type I error and two other methods, that is, trial sequential analysis and sequential meta-analysis control for type I and II errors (failing to detect a genuine effect) and take account of heterogeneity. This paper compares the methods and considers how they could be applied to LSRs.

Systematic reviews are difficult to keep up to date, but failure to do so leads to a decay in review currency, accuracy, and utility. We are developing a novel approach to systematic review updating termed "Living systematic review" (LSR): systematic reviews that are continually updated, incorporating relevant new evidence as it becomes available. LSRs may be particularly important in fields where research evidence is emerging rapidly, current evidence is uncertain, and new research may change policy or practice decisions. We hypothesize that a continual approach to updating will achieve greater currency and validity, and increase the benefits to end users, with feasible resource requirements over time.

Machine Understanding is the term introduced by authors to denote understanding by a machine and is referring to the new area of research the aim of which is to investigate the possibility of building the machine with the ability to understand. A new research area such as machine understanding needs the framework within which the problems will be formulated and solved and the machine, in order to be able to understand, needs to imitate the way in which humans understand the world and the language (text). Machine understanding that denotes understanding by the machine (shape understanding system—SUS) stresses the dependence of learning and understanding processes. In this chapter machine understanding is defined in the context of both human understanding and the existing systems that can be called understanding systems.

In this paper the method of understanding a cyclic object is presented. The cyclic object is a 2D object with holes. The cyclic object frequently appears in many fields of medical research and engineering. The method of understanding of the cyclic object is part of the research aimed at developing a shape understanding system (SUS) able to perform complex visual tasks connected with visual thinking. In SUS, in contrast to the recognition systems, ‘recognition’ is interpreted in the context of the well-defined shape classes. Understanding includes, among others, obtaining the visual concept in the process of the visual reasoning, naming and visual explanation. The result of the visual reasoning is the symbolic name that refers to the possible classes of shape. The possible classes of shape, viewed as hierarchical structures, are incorporated into the shape model. The cyclic class is defined and the processing methods characteristic for this class are described. At each stage of the reasoning process that leads to assigning an examined object to one of the possible classes, the novel processing methods are used. These methods are very efficient because they deal with the very specific classes of shapes. The main novelty of the presented method is that the cyclic object is related to the concept of the shape categories called the symbolic names. This approach makes it possible at first to focus on the processing of the cyclic object and next to interpret it as a real world object or a sign.

In this paper a method of understanding of the concave polygon object is presented. The proposed method of understanding the concave polygon object is part of the shape understanding method. The main novelty of the presented method is that the process of understanding of the concave polygon object is related to the visual concept represented as a symbolic name of the possible classes of shape. The possible classes of shape, viewed as hierarchical structures, are incorporated into the shape model. At each stage of the reasoning process that led to assigning of an examined object to one of the possible classes, the novel processing methods were used. These methods, implemented as modules of the shape understanding system (SUS) and tested on the broad classes of shapes, are very efficient because they deal with a very specific classes of shape. The system consists of different types of experts that perform different processing and reasoning tasks. The concave polygon class models and processing methods are also described.

Understanding is based on a large number of highly varied abilities called intelligence that can be measured. In this paper understanding abilities of the shape understanding system (SUS) are tested based on the methods used in intelligence tests. These tests are formulated as tasks given to the system and performance is compared with the human performance of these tasks. The main novelty of the presented method is that the process of understanding is related to the visual concept represented as a symbolic name of the possible class of shapes. The visual concept is one of the ingredients of the concept of the visual object (the phantom concept) that makes it possible to perform different tasks that are characteristic for visual understanding. The presented results are part of the research aimed at developing the shape understanding method that would be able to perform complex visual tasks connected with visual thinking. The shape understanding method is implemented as the shape understanding system. © 2005 Wiley Periodicals, Inc. Int J Int Syst 20: 799–826, 2005.

Understanding is based on a large number of highly varied abilities called intelligence that can be measured. In this paper understanding abilities of the shape understanding system (SUS) are tested based on the adoption of the intelligence tests. The SUS tests are formulated as the tasks given to the system and performance of SUS is compared with the human performance of these tasks. The main novelty of the presented method is that the process of understanding is related to the visual concept represented as a symbolic name of the possible classes of shape. The visual concept is one of the ingredients of the concept of the visual object (the phantom concept) that makes it possible to perform different tasks that are characteristic for the visual understanding. The presented results are part of the research aimed at developing the shape understanding method able to perform the complex visual tasks connected with visual thinking. The shape understanding method is implemented as the shape understanding system (SUS).

In this paper, the method of understanding of visual objects is presented. The main novelty of the presented method is that the process of understanding is related to the visual concept represented as a symbolic name of the possible classes of shapes. The possible classes of shapes, viewed as hierarchical structures, are incorporated into the shape model. At each stage of the reasoning process that led to assigning an examined object to one of the possible classes, novel processing methods are used. An understanding is based on interpretation of the visual object as a meaningful unit. A big advantage of the proposed method of understanding of the visual objects is that it can explain many problems connected with understanding visual forms. In this paper, the selected concept of the method of understanding of the visual objects is discussed in the context of the psychological and philosophical research. This method is implemented as a module of the shape understanding system (SUS) and tested on the broad c...

In this paper the visual reasoning that is part of visual thinking capabilities of the shape understanding system (SUS) is investigated. This research is a continuation of the authors' previous work focused on investigating understanding capabilities of the intelligent systems based on the shape understanding system. SUS is an example of the visual understanding system, where sensory information is transformed into the multilevel representation in the concept formation process that is part of the visual thinking capabilities. The visual reasoning involves transformation of the description of the object when passing consequent stages of the reasoning process and the reasoning and processing of the data are mutually dependent.

This paper presents the method of understanding objects that can be considered as thin objects. The proposed method of understanding thin objects is part of the shape understanding method developed by the author. The main novelty of the presented method is that the process of understanding thin objects is related to the visual concept represented as a symbolic name of the possible class of shapes. The possible classes of shape, viewed as hierarchical structures, are incorporated into the shape model. At each stage of the reasoning process that led to assigning of an examined object to one of the possible classes, novel processing methods are used. These methods are very efficient because they deal with a very specific class of shapes. In this paper, the 2-D objects that are classified as thin objects are regarded as geometrical objects without any reference to the real world objects. However, the shape under standing method is designed to understand an object at many levels of interpretation, such as the topological level, the linguistic level and the real world reference level. This approach influences the way in which the system of shape understanding is designed. The system consists of different types of experts that perform different processing and reasoning tasks.

The shape taxonomy based on the concept of possible classes of shape is proposed. Possible classes of shape are based on the shape models and viewed as a hierarchical structure of the level of description. At each level of description the different aspects of shape such as geometrical properties of shape perceptual properties of figure and meaningful properties of visual form are incorporated in the shape model. As a part of the hierarchical structure of shape the general classes, the base classes and the specific classes of shape arc proposed. Classes derived from the general classes, namely, the convex classes and the concave classes are described. Possible classes of shape are developed as a part of the shape understanding method that is implemented as the Shape Understanding System (SUS).

The shape understanding system (SUS) that is able to perform different tasks of shape analysis and recognition, based on the ability of the system to understand the different concepts of shape at the different levels of cognition, is proposed. The proposed method of shape understanding is based on the concept of possible classes of shapes. Shape understanding is carried out in several stages of reasoning process that involves transformation of the object by applying image transformations. The process of shape understanding includes creating the concept of shape based on the data acquired in the processing stage. The SUS consists of different types of experts that perform different processing and reasoning tasks.

In the previous Chapters, the knowledge implementation was defined in the context of both human learning and machine learning. The knowledge implementation is focused on learning of the knowledge and skills by machine (SUS) and is concerned with two main aspects of human learning: learning of the visual knowledge in the context of the categorical structure of the learned categories of the visual objects and learning of the knowledge that is connected with understanding of the content of the text. As it was described in Chapter 3, SUS operates in two main modes, learning and understanding mode. SUS ability to understand depends on the effectiveness of learning process and learning of the new knowledge depends on SUS ability to understand.

In Chapter 3, the knowledge implementation is defined in the context of both human learning and machine learning. The term the knowledge implementation introduced in this book is refering to learning of the knowledge and skills by the machine, the shape understanding system (SUS). It is concerned with two main aspects of human learning: learning of visual knowledge in the context of the categorical structure of the learned categories of the visual objects and learning of the knowledge that is connected with understanding of the content of the text. The knowledge implementation is part of the shape understanding method, developed by authors [1] that stresses the importance of the knowledge acquisition in understanding and thinking processes. In this Chapter some aspects of the shape understanding method, that allow for better understanding of the knowledge implementation approach, are presented.