Refining DICOM for pathology--progress from the IHE and DICOM pathology working groups.

For making medical decisions, healthcare professionals require that all necessary information is both correct and easily available. We address the issue of integrating anatomical pathology department information into the electronic healthcare enterprise. The pathology workflow from order to report, including specimen processing and image acquisition was modelled. An integration profile - pathology general workflow- was created in the framework of the Integrating the Healthcare Enterprise (IHE). This Integration Profile relies on 8 transactions based on HL7 or DICOM standards. An important issue was to define information entities (order, imaging study and report) and real-world objects (specimen, tissue sample, slide, etc.). Joint efforts between IHE and DICOM WG26 has resulted in a proposed common model for "specimen" usable for both HL7 and DICOM transactions related to anatomic pathology.

The role of local terminologies in electronic health records. The HEGP experience.

Despite decades of work, there is no universally accepted standard medical terminology and no generally usable terminological tools have yet emerged. The local dictionary of concepts of the Georges Pompidou European Hospital (HEGP) is a Terminological System (TS) designed to support clinical data entry. It covers 93 data entry forms and contains definitions and synonyms of more than 5000 concepts, sometimes linked to reference terminologies such as ICD-10. In this article, we evaluate to which extend SNOMED CT could fully replace or rather be mapped to the local terminology system. We first describe the local dictionary of concepts of HEGP according to some published TS characterization framework. Then we discuss the specific role that a local terminology system plays with regards to reference terminologies.

Integrating anatomical pathology to the healthcare enterprise.

For medical decisions, healthcare professionals need that all required information is both correct and easily available. We address the issue of integrating anatomical pathology department to the healthcare enterprise. The pathology workflow from order to report, including specimen process and image acquisition was modeled. Corresponding integration profiles were addressed by expansion of the IHE (Integrating the Healthcare Enterprise) initiative. Implementation using respectively DICOM Structured Report (SR) and DICOM Slide-Coordinate Microscopy (SM) was tested. The two main integration profiles--pathology general workflow and pathology image workflow--rely on 13 transactions based on HL7 or DICOM standard. We propose a model of the case in anatomical pathology and of other information entities (orders, image folders and reports) and real-world objects (specimen, tissue samples, slides, etc). Cases representation in XML schemas, based on DICOM specification, allows producing DICOM image files and reports to be stored into a PACS (Picture Archiving and Communication System.

Computation of semantic similarity within an ontology of breast pathology to assist inter-observer consensus.

Computer-assisted consensus in medical imaging involves automatic comparison of morphological abnormalities observed by physicians in images. We built an ontology of morphological abnormalities in breast pathology to assist inter-observer consensus. Concepts of morphological abnormalities extracted from existing terminologies, published grading systems and medical reports were organized in an taxonomic hierarchy and furthermore linked by the relation "is a diagnostic criterion of" according to diagnostic meaning. We implemented position-based, content-based and mixed semantic similarity measures between concepts in this ontology and compared the results with experts' judgment. The position-based similarity measure using both taxonomic and non-taxonomic relations performed as well as the other measures and was used for automatic comparison of morphological abnormalities within the IDEM computer-assisted consensus platform.

Integration of multiple ontologies in breast cancer pathology.

The diagnostic variability in pathology, widely reported in the literature, is partly due to the use of different classification systems by pathologists. The descriptions of morphological characteristics on the same image within different classification systems can be considered as different points of view of pathologists. Our aim is to represent the points of view of the experts in pathology during image interpretation and to propose a method ological and technical solution in order to implement interoperability between these points of view. According to the hybrid ontology approach, we developed a system in three stages consisting in 1) the representation of the various points of view in local ontologies 2) the realization of a shared vocabulary and the development of a mapping tool used to allow the matching of local ontologies and shared vocabulary 3) the development of a transcoding algorithm for the translation of a case description from one point of view to another. A first evaluation of the transcoding algorithm was conducted for 33 cases of breast pathology. Our results show that the pathologists generally produce descriptions of the cases which do not follow rigorously the interpretation rules corresponding to the point of view they assert to adopt. While most of the concepts of local ontologies can be transcoded from a local ontology to another one (varying from 62.5 % to 100% according to the local ontology), the transcoding of a description which is valid according to a certain point of view, often results in a description which is not rigorously in accordance with the new point of view. These results underline the differences of interpretation rules existing in the different points of view.

Specifications and implementation of a new exchange format to support computerized consensus in pathology.

In the pathology domain, consensus sessions around multi-headed microscopes enhance reproducibility and can reduce inter- and intra-observer variability. Computerized tools and Web technology could facilitate the organization of consensus sessions and assist pathologists to agree on features that are relevant to diagnosis. In the context of the IDEM project, whose aim is to achieve a computerized platform to allow pathologists to derive consensual diagnostic during Internet-based collaborative sessions, we propose a new extension of the existing TELESLIDE format. This new extended format enables the storage and the exchange of multi-experts descriptions that will be processed by the IDEM consensus engine to produce consensual descriptions. We describe this new format and its implementation in the IDEM teleconsensus platform.

A collaborative platform for consensus sessions in pathology over Internet.

The design of valid databases in pathology faces the problem of diagnostic disagreement between pathologists. Organizing consensus sessions between experts to reduce the variability is a difficult task. The TRIDEM platform addresses the issue to organize consensus sessions in pathology over the Internet. In this paper, we present the basis to achieve such collaborative platform. On the one hand, the platform integrates the functionalities of the IDEM consensus module that alleviates the consensus task by presenting to pathologists preliminary computed consensus through ergonomic interfaces (automatic step). On the other hand, a set of lightweight interaction tools such as vocal annotations are implemented to ease the communication between experts as they discuss a case (interactive step). The architecture of the TRIDEM platform is based on a Java-Server-Page web server that communicate with the ObjectStore PSE/PRO database used for the object storage. The HTML pages generated by the web server run Java applets to perform the different steps (automatic and interactive) of the consensus. The current limitations of the platform is to only handle a synchronous process. Moreover, improvements like re-writing the consensus workflow with a protocol such as BPML are already forecast.

New terminology services based on term comparison using semantic definitions and similarity computation.

As medical information can be encoded within different terminological systems, terms comparison is an important issue to allow communication between applications. In description logics, terms are compared by the means of semantic definitions and subsumption relations. Similarity is also a convenient method for term comparison but is not supported by terminology servers which implement subsumption relations. We present new terminology services built on a semantic distance that could help for semantic mediation between medical applications ranging from semi-automatic encoders to data mining tools. These services are 1) comparison between terms 2) k-nearest-neighbors 3) support for concept coding 4) automatic generation of similarity tables 5) distance based queries 6) support for clustering.

The HEGP component-based clinical information system.

OBJECTIVES: The opening of the Georges Pompidou University Hospital (HEGP) in southwest Paris from the merging of three aging facilities was the opportunity to conceive and deploy a brand new clinical information system (CIS) based on a component-based approach. This paper describes the process of selection of the business components, the main functions currently in use, and the technical infrastructure that proved necessary. METHODS: The HEGP CIS features generic and healthcare-related components. The generic components include a reference manager, a security manager, a document manager, a Corba bus, and various mediation and supervision tools. The healthcare-related components include the patient, healthcare record, act management, and resource scheduler components. RESULTS: Major functions of CIS were operational at the opening of the hospital in July 2000. Two years later, the unique patient record and the provider order entry system were used in 96% of the concerned healthcare units. Sixty-five percent of the biological orders and 55% of the imaging orders were directly entered by the physicians. Access to investigation results including on-line availability of images is used by physicians in 100% of the units. DISCUSSION AND CONCLUSION: A component-based approach was found to be high-performing and cost-effective for the design and deployment of HEGP CIS.

A customizable similarity measure between histological cases.

IDEM, a computerized environment dedicated to pathologists, includes a Case Based Reasoning (CBR) procedure to retrieve similar histological cases in the database. The relevancy of a retrieved case strongly depends on the similarity measure comparing case descriptions. The present work deals with the definition of a similarity measure in the context of IDEM. In a first step, a theoretical measure (relational, numerical and informed), based on the domain constraints, was selected. In a second step, the theoretical measure is optimized according to the current case base. Results are presented for a database of 53 cases of breast tumors. The contribution of this work is to give to pathologists an interactive environment that optimizes the similarity measure between histological cases. This work is also a contribution to the CBR cycle life since the similarity measure can be adapted while new cases are added to the base.