A property concept frame representation for flexible image-content retrieval in histopathology databases.

In histopathology databases, images descriptions are collections of properties provided by experts. Image content retrieval implies comparison of such properties. The objective of this work is to enrich the traditional attribute-value representation of properties in order to take into account the polymorphism and subjectivity of properties and to manage the comparison process. In this paper we define a property concept frame (PCF) representation based on fuzzy logic to handle both representation and comparison. Seven quantifiable morphological characteristics were selected from histopathological reports to illustrate the variety of fuzzy predicates and linguistic terms in properties. The PCF representation has been tested in the context of breast pathology. It is concluded that the PCF representation provides a unification scheme to retrieve in images morphological characteristics that are described in different ways. It may enhance the relevancy of applications in various contexts such as image content-based retrieval or case-based reasoning from images.

Towards content-based image retrieval in a HIS-integrated PACS.

Development of a Picture Archiving and Communications System (PACS) with a large and diverse set of medical images will lead to large digital libraries that can be accessed to provide improved support for patient care, research and education. New representational and retrieval models for clinical images are required to address these issues. The PACS at the Georges Pompidou Hospital (GPH) is integrated in the hospital information system (HIS), and several modalities from medical imaging departments have been attached to it. The two main axes of the GPH PACS project were 1) HIS-integration to allow hospital-wide access to the images based on demographic and procedure-type information and 2) the development of content-based image retrieval to enhance the medical impact of image retrieval in daily practice.

A visual coding system in histopathology and its consensual acquisition.

Divergent descriptions of histopathologic images induce inter- and intra-observer variability in diagnosis. Even though a controlled terminology exists to describe medical imaging, pathologists do not always agree on the visual representation of the descriptive terms. The main purpose of our work is to define a methodology to build a standardized visual coding system unambiguously characterizing the terms of a microglossary. The methodology follows two steps: 1) the acquisition of experts' descriptions of images using the microglossary and 2) a consensus derivation. The procedure was applied on a set of 85 histopathological images of breast tumors described by two experts. Among the 339 objects selected in images, 176 were detected by both experts, 77% localized at the same place and 25% also identically labeled. The microglossary was enriched and illustrated via the resulting consensual descriptions. The contribution of this work supports relevant indexing of biomedical images and image-related information.

Occupational cancer in France: epidemiology, toxicology, prevention, and compensation.

This article is a description of the current situation in France with regard to occupational cancer: research, prevention, and occupation. Toxicologic experiments are carried out using (italic)in vitro(/italic) and (italic)in vivo(/italic) tests, particularly using transgenic mice. Several epidemiologic studies have been conducted over the last decades: population-based case-control studies; mortality studies and cancer incidence studies carried out in historical cohorts of workers employed in the industry; and case-control studies nested in occupational cohorts. French ethical aspects of toxicologic and epidemiologic studies are described. The results thus obtained are used to establish regulations for the prevention and the compensation of cancers attributable to occupational exposure. This French regulation for prevention of occupational cancer involves several partners: (italic)a(/italic)) the states authorities, including labor inspectors, responsible for preparing and implementing the labor legislation and for supervising its application, particularly in the fields of occupational health and safety and working conditions; (italic)b(/italic)) the Social Security Organisation for the analysis of present or potential occupational risks based on tests, visits in plants, complaints or requests from various sources, and statistics. These activities are performed within the framework of the general French policy for the prevention of occupational cancer. This organization includes the National Institute for Research and Safety, particularly involved in research in the various fields of occupational risks--animal toxicology, biologic monitoring, exposure measurements epidemiology, psychology, ergonomy, electronic systems and machineries, exposure to chemicals, noise, heat, vibration, and lighting; and (italic)c(/italic)) companies where the regulation defines the role of the plant manager, the occupational physician, and the Health, Safety and Working Conditions Committee (comprising the manager, employees' representatives, the occupational physician, and the safety department) in dealing with any problem regarding safety, occupational hygiene, and working conditions. These organizations along with medical practitioners are involved with the compensation of occupational cancers. The regulation for compensation includes the tables of occupational cancer, the possibility of recognition of a cancer case when the requirements of the tables are not met, and the postprofessional follow-up of workers exposed to a carcinogenic agent.

IDEM: a Web application of case-based reasoning in histopathology.

Different software engineering and artificial intelligence methods can be used to design Internet retrieval of prototypical medical images. We used the case-based reasoning (CBR) approach to provide an 'intelligent' access to a collection of illustrated medical cases through the Internet. This paper presents a Web interface for the CBR system IDEM (image and diagnosis from examples in medicine) in the domain of breast pathology. Thanks to the definition of a similarity measure between the descriptions of cases we propose a flexible querying of the case-base and a quantitative browsing among cases through similarity links. The resemblance rates provided by the system argue for the quality and the relevancy of the retrieved data. The flexibility of the querying process is robust to missing information and could be adapted to a daily practice. The CBR approach is a promising method for a clinical relevant and an efficient retrieval of reference images and diagnosis clues through Internet.

Lung cancer risk in hard-metal workers.

An industry-wide mortality study on the association between lung cancer and occupational exposure to cobalt and tungsten carbide was carried out in the French hard-metal industry. This case-control study was nested in the historical cohort of workers ever employed in this industry's 10 facilities, most of which are located in eastern France. Workers were followed up from 1968 to 1991. Occupational exposure was assessed using a job-exposure matrix that provided semiquantitative scores for 320 job periods. These scores were significantly correlated with the levels of cobalt measured in 744 historical air samples. In this cohort, which comprised 5,777 males and 1,682 females, the death rate from lung cancer was significant (63 deaths, standardized mortality ratio=1.30, 95% confidence interval (CI) 1.00-1.66) when compared with national death rates. Sixty-one cases and 180 controls were included in the study. When the exposures during the last 10 years were ignored, a twofold lung cancer risk was observed among workers simultaneously exposed to cobalt and tungsten carbide (odds ratio (OR)=1.93, 95% CI 1.03-3.62) adjusted for other cobalt exposure (OR=2.21, 95% CI 0.99-4.90). The odds ratios increased with cumulative exposure (first quartile, OR=1.00; second quartile, OR=2.64; third quartile, OR=2.59; fourth quartile, OR=4.13) and, to a lesser degree, with duration of exposure (one decade, OR=1.00; two decades, OR=1.61; three decades, OR=2.77; four decades, OR=2.03). Adjustments for smoking and for exposures to known or suspected carcinogens did not change the results, yet the odds ratio for smoking (3.38) was lower than expected, suggesting the possibility of some misclassification. Occupational risk was highest among smokers. This study supports the hypothesis that workers who manufacture hard metals have an increased mortality from lung cancer due to simultaneous exposure to cobalt and tungsten carbide.

Case based diagnosis in histopathology of breast tumours.

Relevant knowledge and decision making process in histopathology is mostly included in typical pathological cases encountered by the expert. In this article we address the issue of exploiting this knowledge in the diagnosis process. We present the first steps of a Case-Based-Reasoning (CBR) system that uses the previously resolved pathological cases in order to facilitate decision making and diagnosis formulation of a new case. The work has been performed in two phases. Firstly, an object-oriented model of the domain was developed and 35 pathological cases of breast tumours were represented within this model. Secondly, the functional architecture of the CBR system was designed and the main procedure, the selection of similar cases, was achieved. The selection procedure is based on an original similarity measure that takes into account both semantic and structural resemblances and differences between the cases. A first evaluation of the system was performed on several cases of the data base. The interest of the CBR approach in situations where heuristic rules cannot be clearly defined is discussed.

[Development of a job-exposure matrix in the heavy-metal industry in France]. / Elaboration d'une matrice emplois-expositions dans l'industrie productrice de métaux durs en France.

A job exposure matrix (JEM) was developed by a committee of experts using the DELPHI method, in the French hard metal industry, in order to assess occupational exposures to cobalt along with tungsten carbide resulting from the industrial process. This JEM is part of a nested case-control study, carried out within the historical cohort of workers ever employed in these factories, aimed at assessing lung cancer risk. The committee included 8 experts: hygienists, chemical engineers, occupational physicians and epidemiologists. The JEM was developed in four stages: (i) visit of factories, (ii) definition of lines (job-periods) and columns (exposures) of the JEM, (iii) definition of coding procedures, (iv) coding the cells of the JEM. This last stage used a method derived from the DELPHI method. Throughout the study period 1945-1994. 320 job-periods and 21 agents were defined. A quantitative assignment (level 0 to level 9) along with a frequency code (1 to 3) was attempted for 4 agents, whereas only a qualitative assignment (non exposed/exposed, i.e. 0/l) was done for the other agents. An additional probability code (1 to 3) was assigned to all agents. This procedure led to 46 columns and 14,720 cells in the JEM. When applying the DELPHI method, the consensus of the committee was obtained for 85% of all cells after the first individual assignment of experts, 88% after the second individual assignment and 100% after the third assignment by the experts all together. In order to validate the JEM, these expert assignments will be brought together with the results of exposure measurements that were performed in some workplaces of these factories. The JEM will also be linked with the data base of the case-control study for the exposure assessment of cases and controls.