Brain connectivity in body dysmorphic disorder compared with controls: a diffusion tensor imaging study.

BACKGROUND: Several neuroimaging studies have investigated brain grey matter in people with body dysmorphic disorder (BDD), showing possible abnormalities in the limbic system, orbitofrontal cortex, caudate nuclei and temporal lobes. This study takes these findings forward by investigating white matter properties in BDD compared with controls using diffusion tensor imaging. It was hypothesized that the BDD sample would have widespread significantly reduced white matter connectivity as characterized by fractional anisotropy (FA). METHOD: A total of 20 participants with BDD and 20 healthy controls matched on age, gender and handedness underwent diffusion tensor imaging. FA, a measure of water diffusion within a voxel, was compared between groups on a voxel-by-voxel basis across the brain using tract-based spatial statistics within the FSL package. RESULTS: Results showed that, compared with healthy controls, BDD patients demonstrated significantly lower FA (p < 0.05) in most major white matter tracts throughout the brain, including in the superior longitudinal fasciculus, inferior fronto-occipital fasciculus and corpus callosum. Lower FA levels could be accounted for by increased radial diffusivity as characterized by eigenvalues 2 and 3. No area of higher FA was found in BDD. CONCLUSIONS: This study provided the first evidence of compromised white matter integrity within BDD patients. This suggests that there are inefficient connections between different brain areas, which may explain the cognitive and emotion regulation deficits within BDD patients.

Protein secondary structure prediction using two-level case-based reasoning.

We have developed a two-level case-based reasoning architecture for predicting protein secondary structure. The central idea is to break the problem into two levels: (i) reasoning at the object (protein) level and using the global information from this level to focus on a more restricted problem space; (ii) decomposing objects into pieces (segments) and reasoning at the level of internal structures. As a last step to the procedure, inferences from the parts of the internal structure are synthesized into predictions about global structure. The architecture has been developed and tested on a commonly used data set with 69.5% predictive accuracy. It was then tested on a new data set with 68.2% accuracy. With additional tuning, over 70% accuracy was achieved. In addition, a series of experiments were conducted to test various aspects of the method and the results are informative.

Carcinogenicity predictions for a group of 30 chemicals undergoing rodent cancer bioassays based on rules derived from subchronic organ toxicities.

Rodent carcinogenicities for a group of 30 chemicals which form the subject of the Second NIEHS Predictive-Toxicology Evaluation Experiment are predicted based on their subchronic organ toxicities. Predictions are made by rules learned by the rule learning (RL) induction program.

A temporal analysis of QMR.

OBJECTIVE: To understand better the trade-offs of not incorporating explicit time in Quick Medical Reference (QMR), a diagnostic system in the domain of general internal medicine, along the dimensions of expressive power and diagnostic accuracy. DESIGN: The study was conducted in two phases. Phase I was a descriptive analysis of the temporal abstractions incorporated in QMR's terms. Phase II was a pseudo-prospective controlled experiment, measuring the effect of history and physical examination temporal content on the diagnostic accuracy of QMR. MEASUREMENTS: For each QMR finding that would fit our operational definition of temporal finding, several parameters describing the temporal nature of the finding were assessed, the most important ones being: temporal primitives, time units, temporal uncertainty, processes, and patterns. The history, physical examination, and initial laboratory results of 105 consecutive patients admitted to the Pittsburgh University Presbyterian Hospital were analyzed for temporal content and factors that could potentially influence diagnostic accuracy (these included: rareness of primary diagnosis, case length, uncertainty, spatial/causal information, and multiple diseases). RESULTS: 776 findings were identified as temporal. The authors developed an ontology describing the terms utilized by QMR developers to express temporal knowledge. The authors classified the temporal abstractions found in QMR in 116 temporal types, 11 temporal templates, and a temporal hierarchy. The odds of QMR's making a correct diagnosis in high temporal complexity cases is 0.7 the odds when the temporal complexity is lower, but this result is not statistically significant (95% confidence interval = 0.27-1.83). CONCLUSIONS: QMR contains extensive implicit time modeling. These results support the conclusion that the abstracted encoding of time in the medical knowledge of QMR does not induce a diagnostic performance penalty.

Designing medical informatics research and library--resource projects to increase what is learned.

Careful study of medical informatics research and library-resource projects is necessary to increase the productivity of the research and development enterprise. Medical informatics research projects can present unique problems with respect to evaluation. It is not always possible to adapt directly the evaluation methods that are commonly employed in the natural and social sciences. Problems in evaluating medical informatics projects may be overcome by formulating system development work in terms of a testable hypothesis; subdividing complex projects into modules, each of which can be developed, tested and evaluated rigorously; and utilizing qualitative studies in situations where more definitive quantitative studies are impractical.

The potential of organ specific toxicity for predicting rodent carcinogenicity.

Relationships between organ specific toxicity (specifications of the presence or absence of 43 morphological effects in 32 organs) observed from 13-week subchronic studies and rodent carcinogenicity were investigated by manually measuring the concordance of each feature and also automatically using the RL (Rule Learner) induction program. Of the 32 organs, the presence or absence of any effect in liver or kidney was found very relevant to rodent carcinogenicity. While the concordance of Salmonella genotoxicity with rodent carcinogenicity was only 60%, the battery of liver and kidney was 74% accurate with 75% sensitivity and 71% specificity. Further, using the RL program, rule sets based on organ specific toxicity together with the default predictions based on Salmonella mutagenicity were on average 80% accurate with 83% sensitivity and 82% specificity.

Learning rules to predict rodent carcinogenicity of non-genotoxic chemicals.

The results of short-term assays (induction of chromosomal aberrations and sister-chromatid exchanges, oncogenic transformations and cellular toxicity) together with MTD (maximum tolerated dose) values and physical chemical properties of non-genotoxic (i.e. Salmonella non-mutagens) carcinogens and non-carcinogens were submitted to RL, an inductive learning program. RL was able to learn rules that correctly predicted between 70 and 80% of non-genotoxic chemicals. This is a marked improvement over current predictions using only the results of short-term assays and exceeds the predictions of human experts that used the whole spectrum of acute and subchronic toxicity results as well as human knowledge and intuition.

Designing computer-based frameworks that facilitate doctor--patient collaboration.

A current trend in medicine involves establishing collaborative problem solving between patients and physicians in order to involve patients more in their own care. Neither diagnosis nor therapy can be completely successful unless the patient and the doctor understand each other and collaborate with each other in an effort to gauge the other's requests, needs and concerns. This is made even more difficult by the fact that there is often a big difference between the doctors and patients in terms of expectations, vocabulary used, and other factors. For diagnosis of many disorders, a detailed description of the problem and of the patient's history are required. For therapy, patients must understand how and when to take prescribed drugs, what changes to make in diet, exercise, or lifestyle and why they are important. This paper describes a model of asynchronous collaboration between people with very different knowledge of medicine in which a computer framework attempts to mediate between patients and physicians and reduce some of the differences in communication. It allows patients to pace themselves in familiarizing themselves with the relevant domain terms, some of the medical factors underlying the conditions under question, and the justifications and implications of the prescribed treatment plan. It also allows physicians to request more information of patients and gives patients contextual information to help them understand the underlying reasons for the questions. This framework has been partially implemented in the domain of migraines. As described in the paper, not only is the system designed to cooperate with the patient, but using the system also results in better mutual understanding between the doctor and the patient, thus leading to better collaboration between them.

An intelligent interactive system for delivering individualized information to patients.

This paper is a report on the first phase of a long-term, interdisciplinary project whose goal is to increase the overall effectiveness of physicians' time, and thus the quality of health care, by improving the information exchange between physicians and patients in clinical settings. We are focusing on patients with long-term and chronic conditions, initially on migraine patients, who require periodic interaction with their physicians for effective management of their condition. We are using medical informatics to focus on the information needs of patients, as well as of physicians, and to address problems of information exchange. This requires understanding patients' concerns to design an appropriate system, and using state-of-the-art artificial intelligence techniques to build an interactive explanation system. In contrast to many other knowledge-based systems, our system's design is based on empirical data on actual information needs. We used ethnographic techniques to observe explanations actually given in clinic settings, and to conduct interviews with migraine sufferers and physicians. Our system has an extensive knowledge base that contains both general medical terminology and specific knowledge about migraine, such as common trigger factors and symptoms of migraine, the common therapies, and the most common effects and side effects of those therapies. The system consists of two main components: (a) an interactive history-taking module that collects information from patients prior to each visit, builds a patient model, and summarizes the patients' status for their physicians; and (b) an intelligent explanation module that produces an interactive information sheet containing explanations in everyday language that are tailored to individual patients, and responds intelligently to follow-up questions about topics covered in the information sheet.

An evaluation of machine-learning methods for predicting pneumonia mortality.

This paper describes the application of eight statistical and machine-learning methods to derive computer models for predicting mortality of hospital patients with pneumonia from their findings at initial presentation. The eight models were each constructed based on 9847 patient cases and they were each evaluated on 4352 additional cases. The primary evaluation metric was the error in predicted survival as a function of the fraction of patients predicted to survive. This metric is useful in assessing a model's potential to assist a clinician in deciding whether to treat a given patient in the hospital or at home. We examined the error rates of the models when predicting that a given fraction of patients will survive. We examined survival fractions between 0.1 and 0.6. Over this range, each model's predictive error rate was within 1% of the error rate of every other model. When predicting that approximately 30% of the patients will survive, all the models have an error rate of less than 1.5%. The models are distinguished more by the number of variables and parameters that they contain than by their error rates; these differences suggest which models may be the most amenable to future implementation as paper-based guidelines.

Representing and developing temporally abstracted knowledge as a means towards facilitating time modeling in medical decision-support systems.

The utilization of the appropriate level of temporal abstraction is an important aspect of time modeling. We discuss some aspects of the relation of temporal abstraction to important knowledge engineering parameters such as model correctness, ease of model specification, knowledge availability, query completeness, inference tractability, and semantic clarity. We propose that versatile and efficient time-modeling formalisms should encompass ways to represent and reason at more than one level of abstraction, and we discuss such a hybrid formalism. Although many research efforts have concentrated on the automation of specific temporal abstractions, much research needs to be done in understanding and developing provably optimal abstractions. We provide an initial framework for studying this problem in a manner that is independent of the particular problem domain and knowledge representation, and suggest several research challenges that appear worth pursuing.

Broadening our approach to evaluating medical information systems.

Evaluation in medical informatics tends to follow the paradigm of controlled clinical trials. This model carries with it a number of assumptions whose implications for medical informatics deserve examination. In this paper, we describe the conventional wisdom on evaluation, pointing out some of its underlying assumptions and suggesting that these assumptions are problematic when applied to some aspects of evaluation. In particular, we believe that these assumptions contribute to the problem of user acceptance. We then suggest a broader approach to evaluation, offering some conceptual and methodological distinctions that we believe will be of use to the medical informatics community in rethinking this issue.

The use of physician domain knowledge to improve the learning of rule-based models for decision-support.

This paper describes a study testing the hypothesis that the learning of a decision-support model by a computer learning algorithm from clinical data can be improved by the addition of domain knowledge from practicing physicians. The domain of the experiment is community-acquired pneumonia. The overall design of the study compares a computer learning algorithm given clinical data to one given clinical data plus domain knowledge added by physician subjects. This study showed that the performance of the computer-generated models augmented with knowledge added by physician subjects were significantly better than the computer-generated models generated without added knowledge using a two-stage rule induction algorithm in the domain of community-acquired pneumonia. This result was highly significant and shows that the addition of domain knowledge may be beneficial to the learning of clinical decision-support models, especially in domains where data is limited.

Protein secondary structure prediction using two-level case-based reasoning.

We have developed a two-level case-based reasoning architecture for predicting protein secondary structure. The central idea is to break the problem into two levels: first, reasoning at the object (protein) level, and using the global information from this level to focus on a more restricted problem space; second, decomposing objects into pieces (segments), and reasoning at the internal structures level; finally, synthesizing the pieces back to the objects. The architecture has been implemented and tested on a commonly used data set with 69.3% predictive accuracy. It was then tested on a new data set with 67.3% accuracy. Additional experiments were conducted to determine the effects of using different similarity matrices.

Intelligent aids for parallel experiment planning and macromolecular crystallization.

This paper presents a framework called Parallel Experiment Planning (PEP) that is based on an abstraction of how experiments are performed in the domain of macromolecular crystallization. The goal in this domain is to obtain a good quality crystal of a protein or other macromolecule that can be X-ray diffracted to determine three-dimensional structure. This domain presents problems encountered in real-world situations, such as a parallel and dynamic environment, insufficient resources and expensive tasks. The PEP framework comprises of two types of components: (1) an information management system for keeping track of sets of experiments, resources and costs; and (2) knowledge-based methods for providing intelligent assistance to decision-making. The significance of the developed PEP framework is three-fold--(a) the framework can be used for PEP even without one of its major intelligent aids that simulates experiments, simply by collecting real experimental data; (b) the framework with a simulator can provide intelligent assistance for experiment design by utilizing existing domain theories; and (c) the framework can help provide strategic assessment of different types of parallel experimentation plans that involve different tradeoffs.

Heuristic refinement method for the derivation of protein solution structures: validation on cytochrome b562.

A method is described for determining the family of protein structures compatible with solution data obtained primarily from nuclear magnetic resonance (NMR) spectroscopy. Starting with all possible conformations, the method systematically excludes conformations until the remaining structures are only those compatible with the data. The apparent computational intractability of this approach is reduced by assembling the protein in pieces, by considering the protein at several levels of abstraction, by utilizing constraint satisfaction methods to consider only a few atoms at a time, and by utilizing artificial intelligence methods of heuristic control to decide which actions will exclude the most conformations. Example results are presented for simulated NMR data from the known crystal structure of cytochrome b562 (103 residues). For 10 sample backbones an average root-mean-square deviation from the crystal of 4.1 A was found for all alpha-carbon atoms and 2.8 A for helix alpha-carbons alone. The 10 backbones define the family of all structures compatible with the data and provide nearly correct starting structures for adjustment by any of the current structure determination methods.

Induction of rules for biological macromolecule crystallization.

X-ray crystallography is the method of choice for determining the 3-D structure of large macromolecules at a high enough resolution. The rate limiting step in structure determination is the crystallization itself. It takes anywhere between a few weeks to several years to obtain macromolecular crystals that yield good diffraction patterns. The theory of forces that promote and maintain crystal growth is preliminary, and crystallographers systematically search a large parameter space of experimental settings to grow good crystals. There is a wealth of experimental data on crystal growth most of which is in paper laboratory notebooks. Some of the data has been gathered in electronic form, e.g., the Biological Macromolecular Crystallization Database (BMCD) which is a repository of successful experimental conditions for growing over 800 different macromolecules (Gilliland 1987). Crystallographers are in need of computational tools to gather and analyze past data to design new crystal growth trails. We are building the Crystallographer's Assistant (CA) to help crystallographers record and maintain experimental context in electronic form, offer suggestions on experimental conditions that are likely to be successful, and provide explanations for failed experiments. As an initial step in this project, we have applied RL, an inductive learning program, to the BMCD. In this paper we report initial experiments and findings in applying RL to the BMCD. From the point of view of crystallography, we have discovered possibly significant new empirical relationships in crystal growth. From the point of view of machine learning, our work suggests refinements of existing methods for incorporating detailed domain knowledge into inductive analysis techniques.

Validation of the first step of the heuristic refinement method for the derivation of solution structures of proteins from NMR data.

A new method for the analysis of NMR data in terms of the solution structure of proteins has been developed. The method consists of two steps: first a systematic search of the conformational space to define the region allowed by the initial set of experimental constraints, and second, the narrowing of this region by the introduction of additional constraints and optional refinement procedures. The search of the conformational space is guided by heuristics to make it computationally feasible. The method is therefore called the heuristic refinement method and is coded in an expert system called PROTEAN. The paper describes the validation of the first step of the method using an artificial NMR data set generated from the known crystal structure of sperm whale carbon monoxymyoglobin. It is shown that the initial search procedure yields a low-resolution structure of the myoglobin molecule, accurately reproducing its main topological features, and that the precision of the structure depends on the quality of the initial data set.

Involving patients in health care: explanation in the clinical setting.

The long-term goal of our research is to improve the overall effectiveness of physicians' time, by improving the information exchange between physicians and chronic-care patients, initially migraine patients. The computer system we are constructing has a partial knowledge base about migraines, common therapies, and common side effects of those therapies. The system consists of two main programs: data collection and explanation. The design of our system is based on empirical data concerning patients' information needs.