Expert systems in histopathology. IV. The management of uncertainty.

Expert systems deal with data that are categorical and conceptual, that represent elements of fuzzy sets and that often, by themselves, do not allow an unequivocal decision. The management of uncertainty in expert systems thus becomes a crucial issue. It involves defining measures of uncertainty and procedures for combining accumulating evidence in a manner that properly considers the dependence structure of diagnostic clues. Probability theory offers valuable procedures for uncertainty assessment; however, their practical application in the domain of quantitative histopathology and histopathologic diagnosis can be problematic.     

Image understanding system for histopathology

An image understanding machine vision system for histological diagnoses is based on three interacting expert systems: a diagnostic expert system utilizing terms familiar to pathologists, an interpretive expert system relating human diagnostic concepts to computable histometric features and a scene segmentation expert system which extracts the diagnostic information from the imagery. The control software for the image understanding system resides on a multiprocessor computer. This article details measures to maintain system efficiency and to accommodate the requirements of interprocess communication and processing task scheduling.     

The PARTICLE expert system for tumor grading by automated image analysis

While automated microscopic image analysis of histologic sections has been helpful in objectivizing histologic tumor grading and investigating the relationships between grading and prognosis, expert systems have the potential of linking image analysis and other data for statistical analysis and application to a wider range of tumors. One such expert system is PARTICLE, whose development was based on many years of experience in resolving histologic problems by image analysis. This paper discusses the philosophy of expert system for tumor grading and describes its implementation in the PARTICLE system. The system's structure, operations and applications are briefly presented. PARTICLE is essentially based on the evaluation of karyometric data.     

Visual Recognition Software for Binary Classification and Its Application to Spruce Pollen Identification

Discriminating between black and white spruce (Picea mariana and Picea glauca) is a difficult palynological classification problem that, if solved, would provide valuable data for paleoclimate reconstructions. We developed an open-source visual recognition software (ARLO, Automated Recognition with Layered Optimization) capable of differentiating between these two species at an accuracy on par with human experts. The system applies pattern recognition and machine learning to the analysis of pollen images and discovers general-purpose image features, defined by simple features of lines and grids of pixels taken at different dimensions, size, spacing, and resolution. It adapts to a given problem by searching for the most effective combination of both feature representation and learning strategy. This results in a powerful and flexible framework for image classification. We worked with images acquired using an automated slide scanner. We first applied a hash-based “pollen spotting” model to segment pollen grains from the slide background. We next tested ARLO’s ability to reconstruct black to white spruce pollen ratios using artificially constructed slides of known ratios. We then developed a more scalable hash-based method of image analysis that was able to distinguish between the pollen of black and white spruce with an estimated accuracy of 83.61%, comparable to human expert performance. Our results demonstrate the capability of machine learning systems to automate challenging taxonomic classifications in pollen analysis, and our success with simple image representations suggests that our approach is generalizable to many other object recognition problems.     

Using Highlighting to Train Attentional Expertise

Acquiring expertise in complex visual tasks is time consuming. To facilitate the efficient training of novices on where to look in these tasks, we propose an attentional highlighting paradigm. Highlighting involves dynamically modulating the saliency of a visual image to guide attention along the fixation path of a domain expert who had previously viewed the same image. In Experiment 1, we trained naive subjects via attentional highlighting on a fingerprint-matching task. Before and after training, we asked subjects to freely inspect images containing pairs of prints and determine whether the prints matched. Fixation sequences were automatically scored for the degree of expertise exhibited using a Bayesian discriminative model of novice and expert gaze behavior. Highlighted training causes gaze behavior to become more expert-like not only on the trained images but also on transfer images, indicating generalization of learning. In Experiment 2, to control for the possibility that the increase in expertise is due to mere exposure, we trained subjects via highlighting of fixation sequences from novices, not experts, and observed no transition toward expertise. In Experiment 3, to determine the specificity of the training effect, we trained subjects with expert fixation sequences from images other than the one being viewed, which preserves coarse-scale statistics of expert gaze but provides no information about fine-grain features. Observing at least a partial transition toward expertise, we obtain only weak evidence that the highlighting procedure facilitates the learning of critical local features. We discuss possible improvements to the highlighting procedure.     

Expert systems in medicine

The emergence of the artificial intelligence (AI) in computer technology and its application in the medical field enables the researchers to carry out such intelligent activities like image processing, medical reasoning systems, clinical decision supporting and natural language understanding, etc. A gastroenterological expert system application is briefly demonstrated in this paper. Similar expert systems can be seen to be useful in the research of gastrointestinal cytoprotection, including the plan of different compounds with cytoprotective effect, experimental and clinical medical research.     

Recent developments and present status of telepathology.

Telepathology which is the diagnostic work of a pathologist at a distance has been developed to routine application within the last ten years. It can be classified in relation to application, technical solutions, or performance conditions. Diagnostic pathology performance distinguishes primary diagnosis (for example, frozen section statement) from secondary diagnosis (for example, expert consultation) and quality assurance (diagnostic accuracy, continuous education and training). Applications comprise (a) frozen section service; (b) expert consultations; (c) remote control measurements; and (d) education and training. The technical solutions distinguish active (remote control, live imaging) systems from passive (conventional microscope handling, static imaging), and the performance systems with interactive (on-line, live imaging) use from those with passive (offline, static imaging) practice. Intra-operative frozen section service is mainly performed with remote control systems; whereas expert consultations and education/training are commonly based upon Internet connections with static imaging in an off-line mode. The image quality, transfer rates, and screen resolution of active and passive telepathology systems are sufficient for an additional or primary judgment of histological slides and cytological smears. From the technical point of view, remote control telepathology requires a fast transfer and at least near on-line judgement of images, i.e., image acquisition, transfer and presentation can be considered one performance function. Thus, image size, line transfer rate and screen resolution define the practicability of the system. In expert consultation, the pixel resolution of images and natural color presentation are the main factors for diagnostic support, whereas the line transfer rate is of minor importance. These conditions define the technical compartments, especially size and resolution of camera and screen. The performance of commercially available systems has reached a high quality standard. Pathologists can be trained in a short time and use the systems in a routine manner. Several telepathology systems have been implemented in large Institutes of Pathology which serve for frozen section diagnosis in small hospitals located in the local area. In contrast, expert consultation is mainly performed with international connections. There is a remarkable increase of expert consultations by telepathology according to the experiences of the Armed Forces Institute of Pathology or the Department of Pathology, Thoraxklinik, Heidelberg. In expansion of these experiences, a "globalization" of telepathology can be expected. Telepathology can be used to shrink the period necessary for final diagnosis by request for diagnostic assistance to colleagues working in appropriate related time zones. Telepathology is, therefore, not a substitute of conventional diagnostic procedures but a real improvement in the world of pathology.     

A diagnostic expert system

The introduction of new technologies in the field of electronics has influenced the development of technical equipment over the last few years. The progressive miniaturization of integrated circuits makes possible an expansion of the spectrum of functions offered by this equipment. This also applies to medical technology. These more complex units call for new methods of fault detection and diagnosis. In addition to analytical redundancy, tools developed by the artificial intelligence research community, such as expert systems, are becoming more and more important for fault diagnosis. On the basis of a realized diagnosis expert system the possibilities as well as the limits of such system are discussed. Also, possible future developments of artificial intelligence, like machine learning, are considered.     

Dynamics and context dependence of visual category learning

Visual category learning by humans is observed within a paradigm of supervised learning. Mental representations for recognition are reconstructed from the observed data structures by fitting to them predicted classification data obtained from similarity-based models of recognition on the one hand and machine vision systems for image understanding on the other hand. These principles are illustrated with examples concerning the dynamics and the dependence on context of processes of category learning.     

Perspectives,potentials and trends of ex vivo and in vivo optical molecular pathology

It is pivotal for medical applications, such as noninvasive histopathologic characterization of tissue, to realize label‐free and molecule‐specific representation of morphologic and biochemical composition in real‐time with subcellular spatial resolution. This unmet clinical need requires new approaches for rapid and reliable real‐time assessment of pathologies to complement established diagnostic tools. Photonic imaging combined with digitalization offers the potential to provide the clinician the requested information both under in vivo and ex vivo conditions. This report summarizes photonic approaches and their use in combination with image processing, machine learning and augmented virtual reality that might solve current challenges in modern medicine. Details are given for pathology, intraoperative diagnosis in head and neck cancer and endoscopic diagnosis in gastroenterology.      

Medical applications of image analysis with the Magiscan 2

The previous generation of image analysis machines were capable of processing and analyzing binary, i.e., black and white images, and making measurements and decisions thereon. The Magiscan 2, one of the new generation computers, is capable of analyzing gray-level images in a variety of sophisticated ways. Its uses in the medical application of image analysis are presented, as are the techniques used to analyze images in general. The two principal current medical uses are the automatic karyotyping of chromosomes and the automatic screening of cervical smears. Other applications discussed include three-dimensional reconstruction of structures from two-dimensional sections and the possibility of developing expert systems for medical diagnostics.     

Expert systems in treating substance abuse.

Computer programs can assist humans in solving complex problems that cannot be solved by traditional computational techniques using mathematic formulas. These programs, or "expert systems," are commonly used in finance, engineering, and computer design. Although not routinely used in medicine at present, medical expert systems have been developed to assist physicians in solving many kinds of medical problems that traditionally require consultation from a physician specialist. No expert systems are available specifically for drug abuse treatment, but at least one is under development. Where access to a physician specialist in substance abuse is not available for consultation, this expert system will extend specialized substance abuse treatment expertise to nonspecialists. Medical expert systems are a developing technologic tool that can assist physicians in practicing better medicine.     

Expert systems in histopathology. I. Introduction and overview

An introduction to, and overview of, expert systems is presented, along with some preliminary comments on their application in diagnostic and analytical histopathology and cytopathology. The terminology common to expert systems is defined, and the nature of expert systems is discussed. In particular, the differences between expert systems and other types of computer programs (e.g., algorithms) or means of solving problems are explored. The rationale for their use and the types of tasks for which they are appropriate are also discussed.     

Color image analysis for quantifying renal tumor angiogenesis

OBJECTIVE: To segment and quantify microvessels in renal tumor angiogenesis based on a color image analysis method and to improve the accuracy and reproducibility of quantifying microvessel density. STUDY DESIGN: The segmentation task was based on a supervised learning scheme. First, 12 color features (RGB, HSI, I1I2I3 and L*a*b*) were extracted from a training set. The feature selection procedure selected I2L*S features as the best color feature vector. Then we segmented microvessels using the discriminant function made using the minimum error rate classification rule of Bayesian decision theory. In the quantification step, after applying a connected component-labeling algorithm, microvessels with discontinuities were connected and touching microvessels separated. We tested the proposed method on 23 images. RESULTS: The results were evaluated by comparing them with manual quantification of the same images. The comparison revealed that our computerized microvessel counting correlated highly with manual counting by an expert (r = 0.95754). The association between the number of microvessels after the initial segmentation and manual quantification was also assessed using Pearson's correlation coefficient (r = 0.71187). The results indicate that our method is better than conventional computerized image analysis methods. CONCLUSION: Our method correlated highly with quantification by an expert and could become a way to improve the accuracy, feasibility and reproducibility of quantifying microvessel density. We anticipate that it will become a useful diagnostic tool for angiogenesis studies.     

Automatically detecting and tracking free‐ranging Japanese macaques in video recordings with deep learning and particle filters

Recently, automated observation systems for animals using artificial intelligence have been proposed. In the wild, animals are difficult to detect and track automatically because of lamination and occlusions. Our study proposes a new approach to automatically detect and track wild Japanese macaques (Macaca fuscata) using deep learning and a particle filter algorithm. Macaque likelihood is derived through deep learning and used as an observation model in a particle filter to predict the macaques’ position and size in an image. By using deep learning as an observation model, it is possible to simplify the observation model and improve the accuracy of the classifier. We investigated whether the algorithm could find body regions of macaques in video recordings of free‐ranging groups at Katsuyama, Japan to evaluate our model. Experimental results showed that our method with deep learning as an observation model had higher tracking accuracy than a method that uses a support vector machine. More generally, our study will help researchers to develop automatic observation systems for animals in the wild.     

上海滩涂植被资源遥感分析

利用2003年8月2日L andsat5-TM多光谱遥感影像,运用遥感处理软件ERDA S Im ag ine 8.6,经几何校正分幅裁剪等图像预处理后,采用监督分类和目视解译相结合对上海市滩涂植被进行解译分析。结合全球定位系统(GPS)样点定位,对分类结果进行全面的野外核实和修正,同时利用地理信息系统(G IS)对解译结果进行数据合成,统计出滩涂各类植被的分布区域及面积等数据。实际调查及其分析统计显示,上海滩涂植物群落总面积为21302.1hm2,主要植被组成为芦苇、海三棱草及互花米草三大群落,滩涂植物群落具有明显的高程梯度分布规律。大尺度的上海市滩涂植被的空间分布现状及其数量调查为上海市滩涂资源的合理规划、生物多样性保护和可持续开发利用提供科学依据。     

Tracking Plasticity: Effects of Long-Term Rehearsal in Expert Dancers Encoding Music to Movement

Our knowledge of neural plasticity suggests that neural networks show adaptation to environmental and intrinsic change. In particular, studies investigating the neuroplastic changes associated with learning and practicing motor tasks have shown that practicing such tasks results in an increase in neural activation in several specific brain regions. However, studies comparing experts and non-experts suggest that experts employ less neuronal activation than non-experts when performing a familiar motor task. Here, we aimed to determine the long-term changes in neural networks associated with learning a new dance in professional ballet dancers over 34 weeks. Subjects visualized dance movements to music while undergoing fMRI scanning at four time points over 34-weeks. Results demonstrated that initial learning and performance at seven weeks led to increases in activation in cortical regions during visualization compared to the first week. However, at 34 weeks, the cortical networks showed reduced activation compared to week seven. Specifically, motor learning and performance over the 34 weeks showed the typical inverted-U-shaped function of learning. Further, our result demonstrate that learning of a motor sequence of dance movements to music in the real world can be visualized by expert dancers using fMRI and capture highly significant modeled fits of the brain network variance of BOLD signals from early learning to expert level performance.     

Search for promoter sites of prokaryotic DNA using learning techniques

Using learning techniques previously described in this journal, we have built an expert system able to point to the start DNA point of a sequence and therefore to recognize a promoter. However, to build this system, we have focused on the TATA box and its environment. We have used this expert system to look for new promoters and also to construct new promoters. The results obtained are discussed.