Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “knowledge graphs”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 289 records · Page 16Linked to original sources
PubMedDate not supplied

Influence of age, sex, fracture type and pre-fracture living on rehabilitation pattern after hip fracture in the elderly.

PURPOSE: To evaluate the influence of background factors on the rehabilitation pattern after a hip fracture in the elderly. METHOD: Prospective registration based on the Swedish national register for hip fracture patients called RIKSHOFT/SAHFE (Standardised Audit of Hip Fractures in Europe). The place of living was registered both before fracture and during the following four months period (120 days). Graphs were calculated and drawn based on day-to-day changes. Also influences of age, sex, fracture type and type of operation were analyzed. RESULTS: The patient's pre-fracture functional capacity as evidence by the place they were able to manage to live before the fracture was the most discriminating factor for the rehabilitation; more than sex, fracture type or type of operation. Age was also a highly discriminating factor with a pronounced influence on the rehabilitation pattern. CONCLUSIONS: These background parameters are very important factors when planning the rehabilitation of hip fracture patients. A strategy with individualized planning of the rehabilitation procedure will be highly necessary in the future, in view of the increasing amount of elderly with hip fractures prognosticated during the coming decades. The knowledge about influencing factors here presented will be useful when planning and performing the rehabilitation for this resource-consuming group of patients.

Aged↗
PubMed2002

PseudoViewer: automatic visualization of RNA pseudoknots.

MOTIVATION: Several algorithms have been developed for drawing RNA secondary structures, however none of these can be used to draw RNA pseudoknot structures. In the sense of graph theory, a drawing of RNA secondary structures is a tree, whereas a drawing of RNA pseudoknots is a graph with inner cycles within a pseudoknot as well as possible outer cycles formed between a pseudoknot and other structural elements. Thus, RNA pseudoknots are more difficult to visualize than RNA secondary structures. Since no automatic method for drawing RNA pseudoknots exists, visualizing RNA pseudoknots relies on significant amount of manual work and does not yield satisfactory results. The task of visualizing RNA pseudoknots by hand becomes more challenging as the size and complexity of the RNA pseudoknots increase. RESULTS: We have developed a new representation and an algorithm for drawing H-type pseudoknots with RNA secondary structures. Compared to existing representations of H-type pseudoknots, the new representation ensures uniform and clear drawings with no edge crossing for any H-type pseudoknots. To the best of our knowledge, this is the first algorithm for automatically drawing RNA pseudoknots with RNA secondary structures. The algorithm has been implemented in a Java program, which can be executed on any computing system. Experimental results demonstrate that the algorithm generates an aesthetically pleasing drawing of all H-type pseudoknots. The results have also shown that the drawing has high readability, enabling the user to quickly and easily recognize the whole RNA structure as well as the pseudoknots themselves.

Algorithms↗
PubMed2004

[Depth of edge influence on agriculture-forestry boundary in arid valley of upper reaches of Minjiang River, China].

By using moving split-window techniques (MSWT), this study estimated how far the edge effects penetrated the forest and agricultural fields in the arid valley of upper reaches of Minjiang River, southwestern China. Its aim was to provide general information on vegetation along edge to interior gradients in order to assist in interpretation and prediction of biological phenomena associated with agriculture-forestry boundary, and to improve current management practices in such areas. Three types of boundaries (10 transects) were investigated and sampled. The results showed that when the window width reached 6-10, the change of the SED curve on the graph tended to become stable, and one or two peaks occurred. The depth of edge influence was clearly different for different types of boundaries, and could be estimated within 50 m from the edge to interior. The depth of edge influence (DEI) on vegetation diversity almost varied between 12-30 m, mainly depending on the patch type, topography and microclimate, but seldom on slope orientation. Of the 6 forest transects in the three types of boundaries, the DEI was detected only in the forest part transects M2 and M6, but almost detectable in the agricultural part of all transects. MSWT was considered to be a useful tool for characterizing edge dynamics if enough data was available, and became a simple and powerful technique for analyzing the boundary. The results will provide further knowledge for understanding the interaction between forestry and agriculture in the arid valley.

China↗
PubMed2005

KEGG as a glycome informatics resource.

Bioinformatics approaches to carbohydrate research have recently begun using large amounts of protein and carbohydrate data. In this field called glycome informatics, the foremost necessity is a comprehensive resource for genome-scale bioinformatics analysis of glycan data. Although the accumulation of experimental data may be useful as a reference of biological and biochemical information on carbohydrates, this is insufficient for bioinformatics analysis. Thus, we have developed a glycome informatics resource (http://www.genome.jp/kegg/glycan/) in KEGG (Kyoto Encyclopedia of Genes and Genomes), an integrated knowledge base of protein networks, genomic information, and chemical information. This review describes three noteworthy features: (1) GLYCAN, a database of carbohydrate structures; (2) glycan-related pathways; and (3) Composite Structure Map (CSM), a map illustrating all possible variations of carbohydrate structures within organisms. GLYCAN includes two useful tools: an intuitive drawing tool called KegDraw, and an efficient glycan search and alignment tool called KEGG Carbohydrate Matcher (KCaM). KEGG's glycan biosynthesis and metabolism pathways, integrating carbohydrate structures, proteins, and reactions, are also a pivotal resource. CSM is constructed as a bridge between carbohydrate functions and structures. CSM is able to display, for example, expression data of glycosyltransferases in a compact manner. In all the KEGG resources, various objects including KEGG pathways, chemical compounds, as well as carbohydrate structures are commonly represented as graphs, which are widely studied and utilized in the computer science field.

Carbohydrates↗
PubMed2005

A grid layout algorithm for automatic drawing of biochemical networks.

MOTIVATION: Visualization is indispensable in the research of complex biochemical networks. Available graph layout algorithms are not adequate for satisfactorily drawing such networks. New methods are required to visualize automatically the topological architectures and facilitate the understanding of the functions of the networks. RESULTS: We propose a novel layout algorithm to draw complex biochemical networks. A network is modeled as a system of interacting nodes on squared grids. A discrete cost function between each node pair is designed based on the topological relation and the geometric positions of the two nodes. The layouts are produced by minimizing the total cost. We design a fast algorithm to minimize the discrete cost function, by which candidate layouts can be produced efficiently. A simulated annealing procedure is used to choose better candidates. Our algorithm demonstrates its ability to exhibit cluster structures clearly in relatively compact layout areas without any prior knowledge. We developed Windows software to implement the algorithm for CADLIVE. AVAILABILITY: All materials can be freely downloaded from http://kurata21.bio.kyutech.ac.jp/grid/grid_layout.htm; http://www.cadlive.jp/ SUPPLEMENTARY INFORMATION: http://kurata21.bio.kyutech.ac.jp/grid/grid_layout.htm; http://www.cadlive.jp/

Algorithms↗
PubMed1995

Associating semantic grammars with the SNOMED: processing medical language and representing clinical facts into a language-independent frame.

We describe one approach for natural language processing of medical texts that associates a semantic grammar with the SNOMED (Systematized Nomenclature of Medicine). Our research hypothesis is that the combination of the nomenclature's declarative knowledge with a formal grammar would create a scientific sublanguage embedded with medical knowledge that could be used for analyzing and formatting medical texts. This combination permitted the abstraction of templates we call "semantic patterns." These patterns represent both linguistic and medical knowledge, packed into a hybrid information format. We analyzed manually case reports described in the New England Journal of Medicine (NEJM) from 1985 to 1988 and extracted empirically a semantic grammar. Over 2,000 sentences were analyzed. About 160 structural semantic patterns were abstracted and included in the database of one parser. We tested the parser using reports from 1989 to 1990. Results show that this approach is efficient for processing, indexing, and structuring diverse parts of case reports narrative. The analyzed medical sentences are structured into a language-independent semantic frame format. We conclude that the association of semantic grammars with the SNOMED enabled the construction of a formal system for analysis and representation of clinical facts. The transformation of the structured information from its frame format into other representational schemes, like conceptual graphs, is straightforward. Another application includes the use of the formatted language-independent frame for telegraphic English-Japanese translations of medical sentences.

Language↗
PubMed2006

Development of a communication aid to facilitate risk communication in consultations with unaffected women from high risk breast cancer families: a pilot study.

UNLABELLED: The literature on risk perception in women from high-risk breast cancer families reveals persistent over-estimation of risk, even after counseling. In this study, a communication aid was designed to facilitate discussion of risk between clinical geneticists and genetic counselors and women from this high-risk population. METHOD: Stage 1. The aid was developed by an expert panel of clinical geneticists, genetic counselors, psychologists, an epidemiologist, an oncologist, linguists and a consumer. It was guided by the international literature on risk communication and a large multi-centre Australian study of risk communication. The 13 page full-color communication aid used varying formats of words, numbers, graphs and pie-charts to address (a) the woman's subjective risk; (b) the population risk of breast cancer; c) the risk of inherited breast cancer; (d) the cumulative risk for women with BRCA1 and BRCA2 mutations; (e) family risk factors; (f) the woman's suitability for genetic testing; (h) screening and management recommendations, and (i) a re-assessment of the woman's subjective risk. Stage 2: A before-after pilot study of 38 women who were unaffected with breast cancer and were attending four Australian familial cancer clinics was undertaken. Baseline and follow-up questionnaires were completed by 27 women. Outcomes were compared to those observed in 107 similar women undergoing genetic counseling without the communication aid in 2001. RESULTS: The risk communication aid appears to be beneficial; breast cancer genetics knowledge improved in some areas and importantly, risk perceptions improved in the cohort receiving the communication aid. Psychological measures showed no difference in anxiety or depression between the group receiving the communication aid and the comparison cohort. Women and clinicians were very positive about the usefulness of the communication aid as an adjunct to the genetic counseling consultation.

Australia↗
PubMed2003

Research of the flow procedures in a pressure sewer system.

At the Institute of Land Reclamation, Hydrology and Sanitary Engineering of the University at Rostock the pressure and flow ratios are examined within a measuring section in the pressure dewatering system on the Darss peninsula. The objective of the research project is the knowledge upgrade about the highly unsteady hydraulic conditions in a pressure sewer system. This paper firstly presents the method and the dimensioning of pressure dewatering systems, which can be done using either the peak effluent method or the statistical method; the examination program will be explained. The examination includes pressure difference measuring with two pressure meters and flow data measuring via magnetic-inductive flow meters. Additionally the pump running times of 15 pumping stations, as well as the compressor action of the pressure pipe rinsing station are continuously and temporarily recorded and saved. Finally the measuring results which provide initial information about the pressure and flow conditions in a pressure dewatering system will be presented. The effects of the rinsing, the low pressure differences, the air cushions, the seasonal differences as well as the daily development graphs of the wastewater production are all clearly visible.

Engineering↗
PubMed2002

Science's signal transduction knowledge environment: the connections maps database.

Knowledge environment (KE) describes the collection of electronic networking tools that have been and continue to be developed by AAAS and Stanford University libraries. Knowledge environments use practical, production-quality tools to systematize the consensus knowledge within a scientific domain and facilitate users' access to that knowledge. Science's Signal Transduction Knowledge Environment (STKE) is the first in this new concept in electronic publishing that combines the traditional, albeit electronic, publishing of articles, such as reviews, perspectives, and protocols, with tools for organizing and collating information in the cross-disciplinary field of signal transduction. One of the major tools developed for the STKE is the Connections Map database and the software (called CMADES [Connections Maps Authority Data Entry Software]) created to facilitate data entry by Pathway Authorities. The Connections Maps are a graphical representation of a database of information about the molecules involved in cellular signaling cascades. CMADES automates many of the functions involved in adding data into the Connections Maps database, such as references and descriptors, as well as allowing the Authorities to indicate the relationships between the components in the pathway through the use of a graphing tool. CMADES and the Connections Maps represent evolving tools that assist the Authorities in systemizing information regarding a particular system at the organism- and cell-specific level and the canonical level, as well as provide the STKE user with organized and expert-supplied information about signal transduction pathways.

Biology↗
PubMedDate not supplied

Topological shape and size of peptides: identification of potential allele specific helper T cell antigenic sites.

A database of primary sequences of 28 immunogenic peptides, known to elicit T cell response, derived from five different haplotypes was compiled to identify allele specific helper T cell antigenic sites using a rule based graph-theoretical method. The prediction was based on the identification of allele specific patterns in the form of "topological shape and size" present in the peptides. Indices computed from weighted connected graph models of amino acid side chains and peptides were used in this purpose. The system was trained by 10 Ad and 10 non-Ad restricted peptide sequences, assigned actives and inactives, respectively, chosen randomly from the database, and four Ad and four non-Ad restricted sequences were kept as test peptides. This allowed the system to learn about "topological shape and size" specific for Ad restricted peptides from the differences, if any, they had with the inactive peptides in that respect. The system made 100% correct prediction for the training set peptides and misclassified only one inactive peptide of the test set. The system also identified crucial residues for lambda repressor 12-24 and insulin A-chains. This identification also shows that activity related/crucial residues could be located at varying distances from the peptide terminals. To our knowledge, the method is unique of its kind in the literature and may find application in the rational design of synthetic vaccines and other peptides of immunological importance.

Alleles↗
PubMed2006

Detailed polybrominated diphenyl ether (PBDE) congener composition of the widely used penta-, octa-, and deca-PBDE technical flame-retardant mixtures.

Polybrominated diphenyl ethers (PBDEs) have been widely used to flame-retard products common in homes and the workplace, and subsequently, they have become widely dispersed in the environment. Detailed compositional knowledge of these complex PBDE mixtures is crucial to a fuller understanding of their toxicological potencies and environmental fate due to selective congener biomagnification, degradation, and transport. Utilizing recenttechnical enhancements and newly available commercial standards, we developed a method capable of analyzing a larger suite of mono- through deca-BDEs. We then characterized the congener composition of six common technical flame-retardant mixtures: two penta-BDE products (DE-71 and Bromkal 70-5DE) two octa-BDE products (DE-79 and Bromkal 79-8DE) and two deca-BDE products (Saytex 102E and Bromkal 82-0DE). PBDEs were analyzed by gas chromatography/mass spectrometry (GC/MS). Structural conformations based on fragmentation patterns and molecular ions were established by electron-capture negative ionization (ECNI) and electron ionization (El). Sixty-four commercially available PBDE standards were chromato-graphed on two GC columns (DB-1HT and DB-5HT) and relative retention indexes (RRI) calculated. Thirty-nine PBDEs were identified in these products, 29 at concentrations >0.02% by weight. Of these, 12 previously unreported congeners have been confirmed as commercial mixture components. Four of these congeners were detected >0.02% w/w (BDE-144, -171, -180, and -201) and three (BDE-75, -184, and -194) at <0.02%. Five other congeners (four <0.02% by weight) were tentatively identified based on their molecular ion and ECNI fragmentation in the absence of corresponding analytical standards.

Flame Retardants↗
PubMed2000

Complexity IX. Closure over the organization of a scientific truth.

The specificity of human knowledge allow one to construct specific truths about human behavior. A structural notation and language for describing a complex hierarchically organized biological system was developed for the explicit purpose of analyzing the origins of health and disease. A specific application of these concepts to a specific patient (such as an individual suffering from the heritable disease, sickle cell anemia) requires a systematic formulation of a scientific truth. No universal law is applicable. The value of a clinical truth for the patient, as well as for the physician and society, is substantial. This value has moral, ethical, and legal weight. Both physics and chemistry use a universal external invariant reference system. Human beings and other living organisms function by an internal reference system that is neither invariant nor universal. In order to address the complexity of scientific truths within living systems, a mathematical graph is constructed from observations, descriptions, and symbolizations of the relevant human scientific activities and is placed in mutual coreference with three philosophical theories of truth. Consistency within the referencing relations of the graphic object creates an image of complex truths. When mapped over degrees of internal organization, the structural consistencies can form hierarchically transitive relations (many-to-one) creating redundancies that confirm one-another. Both structural and dynamic information can be composed within the graphic framework. The redundancies intrinsic to the degrees of organization notation, semantics, and syntax augment one another in the search for scientific truth. The degree of certitude emerging from structural implications increases in relation to the number of hierarchical degrees of organization invoked to represent the various behaviors of complex systems (as illustrated by the sickle cell anemia example). The successful synthesis of the complex image (or complex simple) of a scientific truth approaches the Heideggerian notion of identity in the sense that A = A and A is A.

Behavior↗
PubMed1991

Thought police. Healthcare executives must carefully guard intellectual property developed in their facilities.

Healthcare executives should have a working knowledge of intellectual property law--the legal principle that, for a limited period, treats intangible ideas and concepts as if they were products or property. Not only should administrators be aware of how to protect their facilities' patents, trademarks, copyrights, and trade secrets, but they must also be sure that employees in their own facilities do not infringe on the intellectual property of others. A patent is granted when an inventor proposes a useful new process, machine, "article of manufacture," or composition of matter, including any new and useful improvement on existing items in these same categories. A trademark is a name, symbol, device, or combination thereof adopted and used by an institution to identify its goods or services and to distinguish them from its competitors' goods or services. Expressions of ideas and thoughts set forth in words, sentences, paragraphs, sketches, pictures, graphs, or any other means of conveying ideas or concepts commonly understood to be works of authorship--both published and unpublished--may be protected by a copyright. Trade secret law applies to those who might misappropriate information that has been given to them because of their special relationship with the holder of the trade secret (i.e., through employment, contract, or other fiduciary or trust relationship).

Confidentiality↗
PubMed1995

"Hyperstat": an educational and working tool in epidemiology.

The work of a researcher in epidemiology is based on studying literature, planning studies, gathering data, analyzing data and writing results. Therefore he has need for performing, more or less, simple calculations, the need for consulting or quoting literature, the need for consulting textbooks about certain issues or procedures, and the need for looking at a specific formula. There are no programs conceived as a workstation to assist the different aspects of researcher work in an integrated fashion. A hypertextual system was developed which supports different stages of the epidemiologist's work. It combines database management, statistical analysis or planning, and literature searches. The software was developed on Apple Macintosh by using Hypercard 2.1 as a database and HyperTalk as a programming language. The program is structured in 7 "stacks" or files: Procedures; Statistical Tables; Graphs; References; Text; Formulas; Help. Each stack has its own management system with an automated Table of Contents. Stacks contain "cards" which make up the databases and carry executable programs. The programs are of four kinds: association; statistical procedure; formatting (input/output); database management. The system performs general statistical procedures, procedures applicable to epidemiological studies only (follow-up and case-control), and procedures for clinical trials. All commands are given by clicking the mouse on self-explanatory "buttons". In order to perform calculations, the user only needs to enter the data into the appropriate cells and then click on the selected procedure's button. The system has a hypertextual structure. The user can go from a procedure to other cards following the preferred order of succession and according to built-in associations. The user can access different levels of knowledge or information from any stack he is consulting or operating. From every card, the user can go to a selected procedure to perform statistical calculations, to the reference database management system, to the textbook in which all procedures and issues are discussed in detail, to the database of statistical formulas with automated table of contents, to statistical tables with automated table of contents, or to the help module. he program has a very user-friendly interface and leaves the user free to use the same format he would use on paper. The interface does not require special skills. It reflects the Macintosh philosophy of using windows, buttons and mouse. This allows the user to perform complicated calculations without losing the "feel" of data, weight alternatives, and simulations. This program shares many features in common with hypertexts. It has an underlying network database where the nodes consist of text, graphics, executable procedures, and combinations of these; the nodes in the database correspond to windows on the screen; the links between the nodes in the database are visible as "active" text or icons in the windows; the text is read by following links and opening new windows. The program is especially useful as an educational tool, directed to medical and epidemiology students. The combination of computing capabilities with a textbook and databases of formulas and literature references, makes the program versatile and attractive as a learning tool. The program is also helpful in the work done at the desk, where the researcher examines results, consults literature, explores different analytic approaches, plans new studies, or writes grant proposals or scientific articles.

Computer-Assisted Instruction↗
PubMed2006

Improving the specificity of the [13C]mixed triacylglycerol breath test by estimating carbon dioxide production from heart rate.

BACKGROUND: The [13C]mixed triacylglycerol (MTG) breath test is a non-invasive measure of fat digestion and can be used to assess the need for enzyme replacement therapy in children with cystic fibrosis (CF). However, it lacks specificity. Quantitation of cumulative percent dose recovered (cPDR) requires knowledge of carbon dioxide production rate (VCO2). A resting value is assumed, but children are unlikely to remain at rest during the test. OBJECTIVE: To improve the specificity and therefore the positive predictive value (PPV) of the MTG breath test using calibrated heart rate monitors to estimate non-resting VCO2. DESIGN: Proof of concept study. SUBJECTS: Six children with CF, 10 healthy children and eight healthy adults performed [13C]MTG breath tests. METHODS: Heart rate monitors were worn throughout the test. Non-resting VCO2 was estimated continuously from heart rate. Percentage dose recovered was calculated using predicted resting VCO2, measured resting VCO2 and non-resting VCO2 estimated from heart rate. Physical activity level (PAL) was taken as cPDR calculated using non-resting VCO2 divided by cPDR calculated using measured resting VCO2. The cutoff point was determined using two graph-receiver operator characteristics. RESULTS: Use of calibrated heart rate monitors to estimate non-resting VCO2 improved the specificity of the test. The PPV increased from 0.67 to 0.99. PAL was 1.3 in adults and children who performed the test in hospital, and 1.7 in children who performed the test at home. CONCLUSION: Individually calibrated heart rate monitors are useful tools to estimate non-resting VCO2 during the [13C]MTG breath test.

Adolescent↗
PubMed2005

Being around and knowing the players: networks of influence in health policy.

The accumulation and use of power is crucial to the health policy process. This paper examines the power of the medical profession in the health policy arena, by analysing which actors are perceived as influential, and how influence is structured in health policy. It combines an analysis of policy networks and social networks, to examine positional and personal influence in health policy in the state of Victoria, Australia. In the sub-graph of the influence network examined here, those most widely regarded as influential are academics, medically qualified and male. Positional actors (the top politician, political advisor and bureaucrat in health and the top nursing official) form part of a core group within this network structure. A second central group consists of medical influentials working in academia, research institutes and health-related NGOs. In this network locale overall, medical academics appear to combine positional and personal influence, and play significant intermediary roles across the network. While many claim that the medical profession has lost power in health policy and politics, this analysis yields few signs that the power of medicine to shape the health policy process has been greatly diminished in Victoria. Medical expertise is a potent embedded resource connecting actors through ties of association, making it difficult for actors with other resources and different knowledge to be considered influential. The network concepts and analytical techniques used here provide a novel means for uncovering different types of influence in health policy.

Administrative Personnel↗
PubMed1993

Can nutrition information increase sales of healthful vended snacks?

Snack selections from unrefrigerated vending machines were studied in relation to increasing availability of nutrient-dense snack options and providing nutrient information at four selected vending sites on a large university campus. Only four of 133 different snacks available for unrefrigerated vending met the criterion for nutrient-dense snacks. When snack proportion was changed to increase availability of nutrient-dense snacks, sales dropped. When nutrition information in bar graph form was posted on the machines, sales increased but not back to the original level. Although snack sales increased after graphic nutrition information was posted, sales were primarily for the least nutrient-dense, perhaps because few well-liked, nutrient-dense snacks were available for unrefrigerated vending. Schools concerned about providing a selection of vended snacks in agreement with the U.S. Dietary Guidelines might be advised to maintain refrigerated vending or to pursue a "healthy snack" machine concept.

Adolescent↗
PubMed2003

Computerized diagnostic decision support system for the classification of preinvasive cervical squamous lesions.

Previous studies have revealed considerable interobserver and intraobserver variation in the histological classification of preinvasive cervical squamous lesions. The aim of the present study was to develop a decision support system (DSS) for the histological interpretation of these lesions. Knowledge and uncertainty were represented in the form of a Bayesian belief network that permitted the storage of diagnostic knowledge and, for a given case, the collection of evidence in a cumulative manner that provided a final probability for the possible diagnostic outcomes. The network comprised 8 diagnostic histological features (evidence nodes) that were each independently linked to the diagnosis (decision node) by a conditional probability matrix. Diagnostic outcomes comprised normal; koilocytosis; and cervical intraepithelial neoplasia (CIN) I, CIN II, and CIN III. For each evidence feature, a set of images was recorded that represented the full spectrum of change for that feature. The system was designed to be interactive in that the histopathologist was prompted to enter evidence into the network via a specifically designed graphical user interface (i-Path Diagnostics, Belfast, Northern Ireland). Membership functions were used to derive the relative likelihoods for the alternative feature outcomes, the likelihood vector was entered into the network, and the updated diagnostic belief was computed for the diagnostic outcomes and displayed. A cumulative probability graph was generated throughout the diagnostic process and presented on screen. The network was tested on 50 cervical colposcopic biopsy specimens, comprising 10 cases each of normal, koilocytosis, CIN I, CIN II, and CIN III. These had been preselected by a consultant gynecological pathologist. Using conventional morphological assessment, the cases were classified on 2 separate occasions by 2 consultant and 2 junior pathologists. The cases were also then classified using the DSS on 2 occasions by the 4 pathologists and by 2 medical students with no experience in cervical histology. Interobserver and intraobserver agreement using morphology and using the DSS was calculated with kappa statistics. Intraobserver reproducibility using conventional unaided diagnosis was reasonably good (kappa range, 0.688 to 0.861), but interobserver agreement was poor (kappa range, 0.347 to 0.747). Using the DSS improved overall reproducibility between individuals. Using the DSS, however, did not enhance the diagnostic performance of junior pathologists when comparing their DSS-based diagnosis against an experienced consultant. However, the generation of a cumulative probability graph also allowed a comparison of individual performance, how individual features were assessed in the same case, and how this contributed to diagnostic disagreement between individuals. Diagnostic features such as nuclear pleomorphism were shown to be particularly problematic and poorly reproducible. DSSs such as this therefore not only have a role to play in enhancing decision making but also in the study of diagnostic protocol, education, self-assessment, and quality control.

Bayes Theorem↗