RESUMO
In this white paper, experts from the Digital Pathology Association (DPA) define terminology and concepts in the emerging field of computational pathology, with a focus on its application to histology images analyzed together with their associated patient data to extract information. This review offers a historical perspective and describes the potential clinical benefits from research and applications in this field, as well as significant obstacles to adoption. Best practices for implementing computational pathology workflows are presented. These include infrastructure considerations, acquisition of training data, quality assessments, as well as regulatory, ethical, and cyber-security concerns. Recommendations are provided for regulators, vendors, and computational pathology practitioners in order to facilitate progress in the field. © 2019 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of Pathological Society of Great Britain and Ireland.
Assuntos
Inteligência Artificial/normas , Benchmarking/normas , Diagnóstico por Computador/normas , Interpretação de Imagem Assistida por Computador/normas , Patologia/normas , Formulação de Políticas , Terminologia como Assunto , Inteligência Artificial/classificação , Inteligência Artificial/ética , Benchmarking/classificação , Benchmarking/ética , Segurança Computacional , Diagnóstico por Computador/classificação , Diagnóstico por Computador/ética , Humanos , Patologia/classificação , Patologia/ética , Valor Preditivo dos Testes , Fluxo de TrabalhoRESUMO
The Food and Drug Administration (FDA or we) is classifying the whole slide imaging system into class II (special controls). The special controls that apply to the device type are identified in this order and will be part of the codified language for the whole slide imaging system's classification. We are taking this action because we have determined that classifying the device into class II (special controls) will provide a reasonable assurance of safety and effectiveness of the device. We believe this action will also enhance patients' access to beneficial innovative devices, in part by reducing regulatory burdens.
Assuntos
Diagnóstico por Computador/classificação , Diagnóstico por Computador/instrumentação , Segurança de Equipamentos/classificação , Hematologia/classificação , Hematologia/instrumentação , Microscopia/classificação , Microscopia/instrumentação , Patologia/classificação , Patologia/instrumentação , HumanosRESUMO
The Food and Drug Administration (FDA or we) is classifying the cervical intraepithelial neoplasia (CIN) test system into class II (special controls). The special controls that apply to the device type are identified in this order and will be part of the codified language for the CIN test system's classification. We are taking this action because we have determined that classifying the device into class II (special controls) will provide a reasonable assurance of safety and effectiveness of the device. We believe this action will also enhance patients' access to beneficial innovative devices, in part by reducing regulatory burdens.
Assuntos
Segurança de Equipamentos/classificação , Histologia/classificação , Histologia/instrumentação , Patologia/classificação , Patologia/instrumentação , Displasia do Colo do Útero/diagnóstico , Biomarcadores Tumorais , Feminino , Humanos , Displasia do Colo do Útero/patologiaRESUMO
Clinical terminologies are considered a key technology for capturing clinical data in a precise and standardized manner, which is critical to accurately exchange information among different applications, medical records and decision support systems. An important step to promote the real use of clinical terminologies, such as SNOMED-CT, is to facilitate the process of finding mappings between local terms of medical records and concepts of terminologies. In this paper, we propose a mapping tool to discover text-to-concept mappings in SNOMED-CT. Name-based techniques were combined with a query expansion system to generate alternative search terms, and with a strategy to analyze and take advantage of the semantic relationships of the SNOMED-CT concepts. The developed tool was evaluated and compared to the search services provided by two SNOMED-CT browsers. Our tool automatically mapped clinical terms from a Spanish glossary of procedures in pathology with 88.0% precision and 51.4% recall, providing a substantial improvement of recall (28% and 60%) over other publicly accessible mapping services. The improvements reached by the mapping tool are encouraging. Our results demonstrate the feasibility of accurately mapping clinical glossaries to SNOMED-CT concepts, by means a combination of structural, query expansion and named-based techniques. We have shown that SNOMED-CT is a great source of knowledge to infer synonyms for the medical domain. Results show that an automated query expansion system overcomes the challenge of vocabulary mismatch partially.
Assuntos
Automação , Codificação Clínica , Processamento de Linguagem Natural , Patologia/classificação , Systematized Nomenclature of Medicine , National Library of Medicine (U.S.) , Melhoria de Qualidade , Espanha , Estados UnidosRESUMO
OBJECTIVE: To develop a system for the automatic classification of pathology reports for Cancer Registry notifications. METHOD: A two pass approach is proposed to classify whether pathology reports are cancer notifiable or not. The first pass queries pathology HL7 messages for known report types that are received by the Queensland Cancer Registry (QCR), while the second pass aims to analyse the free text reports and identify those that are cancer notifiable. Cancer Registry business rules, natural language processing and symbolic reasoning using the SNOMED CT ontology were adopted in the system. RESULTS: The system was developed on a corpus of 500 histology and cytology reports (with 47% notifiable reports) and evaluated on an independent set of 479 reports (with 52% notifiable reports). RESULTS show that the system can reliably classify cancer notifiable reports with a sensitivity, specificity, and positive predicted value (PPV) of 0.99, 0.95, and 0.95, respectively for the development set, and 0.98, 0.96, and 0.96 for the evaluation set. High sensitivity can be achieved at a slight expense in specificity and PPV. CONCLUSION: The system demonstrates how medical free-text processing enables the classification of cancer notifiable pathology reports with high reliability for potential use by Cancer Registries and pathology laboratories.
Assuntos
Neoplasias/patologia , Patologia Clínica , Patologia/classificação , Sistema de Registros , Sistemas Computacionais , Humanos , Processamento de Linguagem Natural , QueenslandRESUMO
OBJECTIVE: To evaluate the effects of Optical Character Recognition (OCR) on the automatic cancer classification of pathology reports. METHOD: Scanned images of pathology reports were converted to electronic free-text using a commercial OCR system. A state-of-the-art cancer classification system, the Medical Text Extraction (MEDTEX) system, was used to automatically classify the OCR reports. Classifications produced by MEDTEX on the OCR versions of the reports were compared with the classification from a human amended version of the OCR reports. RESULTS: The employed OCR system was found to recognise scanned pathology reports with up to 99.12% character accuracy and up to 98.95% word accuracy. Errors in the OCR processing were found to minimally impact on the automatic classification of scanned pathology reports into notifiable groups. However, the impact of OCR errors is not negligible when considering the extraction of cancer notification items, such as primary site, histological type, etc. CONCLUSIONS: The automatic cancer classification system used in this work, MEDTEX, has proven to be robust to errors produced by the acquisition of freetext pathology reports from scanned images through OCR software. However, issues emerge when considering the extraction of cancer notification items.
Assuntos
Processos de Cópia/normas , Prontuários Médicos , Neoplasias/patologia , Patologia Clínica , Patologia/classificação , Automação , Humanos , Processamento de Linguagem NaturalRESUMO
Accurate terminology is the basis for clear communication among specialists and relies upon precise definitions, indispensable for the WHO Classification of Tumours. We identified a number of potentially misleading terms in use in the recently published WHO Classification of Tumours, 5th edition. From a list of common sources that might be consulted by specialists in the pathology field, we searched for definitions of the terms. Where at least two sources provided definitions for a term, we assessed their level of agreement using an ad hoc developed scale. We identified 26 potentially misleading terms from the 5th edition Digestive System and Breast Tumour Books, and 16 sources. The number of definitions provided by the sources ranged from no definition (for four terms) to ten (for two terms). No source had definitions for all terms. We found only 111 (27%) of a possible 416 definitions. Where two or more definitions were present for a term, the level of agreement between them was judged to be high. There was a paucity of definitions for potentially misleading terms in the sources consulted, but there was a good agreement when two or more definitions were present. In a globalized world where healthcare workers and learners in many fields may access these sources to learn about terminology with which they are unfamiliar, the lack of definitions is a hindrance to a precise understanding of classification in the speciality of pathology and to clear communication between specialist groups.
Assuntos
Neoplasias/classificação , Neoplasias/patologia , Patologia/classificação , Terminologia como Assunto , Comunicação , Compreensão , HumanosRESUMO
An algorithm for automatic coding of pathology reports using a multi-axial codification (ADICAP) is described and evaluated. It extracts << significant words >> or expressions from a corpus and records the statistical relationships between them and the modalities of the different axes. Different weighting functions are evaluated. With the best settings, in more than two cases out of three the correct modality was found among the top 5 list of candidates, except for the << organ >> axis. Several ways of improvement are discussed especially regarding the poor results on the << organ >> axis. Perspectives of a two stages assembling algorithm completing this first step are proposed.
Assuntos
Algoritmos , Controle de Formulários e Registros , Sistemas Computadorizados de Registros Médicos , Processamento de Linguagem Natural , Patologia/classificação , Reconhecimento Automatizado de Padrão , Vocabulário Controlado , Indexação e Redação de Resumos , Inteligência Artificial , Registros Hospitalares , Humanos , Armazenamento e Recuperação da Informação , SoftwareRESUMO
The authors present the framework of formal representation of ICD10 based on DOLCE. The goal of the work is to represent the meaning of the categories of the classification systems based on a formal top-level ontology. The ICD categories are described in the space of atomic disease concepts, while the diseases themselves are defined on a pathological basis. The anatomical entities are taken from FMA, functions, morphological abnormalities and procedures from SNOMED International, while organisms are taken form the biological taxonomy. The ontology is represented in OWL 1.0 in a modularised way. The transformation of the previous GALEN-based representation to the same conceptual system is also under way.
Assuntos
Armazenamento e Recuperação da Informação , Classificação Internacional de Doenças , Processamento de Linguagem Natural , Vocabulário Controlado , Indexação e Redação de Resumos , Algoritmos , Inteligência Artificial , Humanos , Patologia/classificação , Software , Systematized Nomenclature of MedicineRESUMO
SNOMED CT was created by the merger of SNOMED RT (Reference Terminology) and Read Codes Version 3 (also known as Clinical Terms Version 3). SNOMED CT is considered to be among the most extensive and comprehensive biomedical vocabularies available today. It is considered for use as the Reference Terminology of various institutions. We review the adequacy of SNOMED CT as a Reference Terminology and discuss the issues in its use as such. We discuss issues with content coverage of various clinical domains, data integrity and validity, and the update frequency of SNOMED CT, and why SNOMED CT alone is not adequate to serve as the Reference Terminology of a healthcare organization.
Assuntos
Systematized Nomenclature of Medicine , Vocabulário Controlado , Animais , Controle de Formulários e Registros/classificação , Humanos , Microbiologia/classificação , Patologia/classificação , Preparações Farmacêuticas/classificaçãoRESUMO
The paper assesses the utility of Medtex on automating Cancer Registry notifications from narrative histology and cytology reports from the Queensland state-wide pathology information system. A corpus of 45.3 million pathology HL7 messages (including 119,581 histology and cytology reports) from a Queensland pathology repository for the year of 2009 was analysed by Medtex for cancer notification. Reports analysed by Medtex were consolidated at a patient level and compared against patients with notifiable cancers from the Queensland Oncology Repository (QOR). A stratified random sample of 1,000 patients was manually reviewed by a cancer clinical coder to analyse agreements and discrepancies. Sensitivity of 96.5% (95% confidence interval: 94.5-97.8%), specificity of 96.5% (95.3-97.4%) and positive predictive value of 83.7% (79.6-86.8%) were achieved for identifying cancer notifiable patients. Medtex achieved high sensitivity and specificity across the breadth of cancers, report types, pathology laboratories and pathologists throughout the State of Queensland. The high sensitivity also resulted in the identification of cancer patients that were not found in the QOR. High sensitivity was at the expense of positive predictive value; however, these cases may be considered as lower priority to Cancer Registries as they can be quickly reviewed. Error analysis revealed that system errors tended to be tumour stream dependent. Medtex is proving to be a promising medical text analytic system. High value cancer information can be generated through intelligent data classification and extraction on large volumes of unstructured pathology reports.
Assuntos
Sistemas Computacionais , Neoplasias/patologia , Patologia/classificação , Sistema de Registros , Humanos , Laboratórios/normas , Programas Obrigatórios , Processamento de Linguagem Natural , Patologia Clínica , Queensland , Sensibilidade e EspecificidadeRESUMO
Automated detection methods can address delays and incompleteness in cancer case reporting. Existing automated efforts are largely dependent on complex dictionaries and coded data. Using a gold standard of manually reviewed pathology reports, we evaluated the performance of alternative input formats and decision models on a convenience sample of free-text pathology reports. Results showed that the input format significantly impacted performance, and specific algorithms yielded better results for presicion, recall and accuracy. We conclude that our approach is sufficiently accurate for practical purposes and represents a generalized process.
Assuntos
Sistemas de Apoio a Decisões Clínicas/organização & administração , Diagnóstico por Computador/métodos , Registros Eletrônicos de Saúde/classificação , Aprendizado de Máquina , Neoplasias/diagnóstico , Patologia/classificação , Interpretação Estatística de Dados , Mineração de Dados/métodos , Humanos , Processamento de Linguagem Natural , Neoplasias/patologiaRESUMO
This work develops an automated classifier of pathology reports which infers the topography and the morphology classes of a tumor using codes from the International Classification of Diseases for Oncology (ICD-O). Data from 94,980 patients of the A.C. Camargo Cancer Center was used for training and validation of Naive Bayes classifiers, evaluated by the F1-score. Measures greater than 74% in the topographic group and 61% in the morphologic group are reported. Our work provides a successful baseline for future research for the classification of medical documents written in Portuguese and in other domains.
Assuntos
Mineração de Dados/métodos , Diagnóstico por Computador/métodos , Processamento de Linguagem Natural , Neoplasias/diagnóstico , Neoplasias/patologia , Patologia/classificação , Brasil/epidemiologia , Sistemas de Apoio a Decisões Clínicas , Classificação Internacional de Doenças , Neoplasias/epidemiologia , Reconhecimento Automatizado de Padrão/métodos , Prevalência , Reprodutibilidade dos Testes , Sensibilidade e EspecificidadeRESUMO
The Food and Drug Administration (FDA) is reclassifying the automated differential cell counter (ADCC) from class III (premarket approval) into class II (special controls). FDA is also identifying the guidance document entitled "Class II Special Controls Guidance Document: Premarket Notifications for Automated Differential Cell Counters for Immature or Abnormal Blood Cells; Final Guidance for Industry and FDA" as the special control that the agency believes will reasonably ensure the safety and effectiveness of the device. This reclassification is being undertaken based on new information submitted in a reclassification petition from the International Society for Laboratory Hematology (ISLH), under the Federal Food, Drug, and Cosmetic Act (the act), as amended by the Safe Medical Devices Act of 1990 and the FDA Modernization Act of 1997.
Assuntos
Contagem de Células Sanguíneas/classificação , Hematologia/classificação , Patologia/classificação , United States Food and Drug Administration , Contagem de Células Sanguíneas/instrumentação , Aprovação de Equipamentos , Desenho de Equipamento , Segurança de Equipamentos , Hematologia/instrumentação , Humanos , Patologia/instrumentação , Estados UnidosRESUMO
The Food and Drug Administration (FDA) is classifying the Factor V Leiden deoxyribonucleic acid (DNA) mutation detections systems device into class II (special controls). The special control that will apply to the device is the guidance document entitled "Class II Special Controls Guidance Document: Factor V Leiden DNA Mutation Detection Systems." The agency is taking this action in response to a petition submitted under the Federal Food, Drug, and Cosmetic Act (the act) as amended by the Medical Device Amendments of 1976 (the 1976 amendments), the Safe Medical Devices Act of 1990 (SMDA), the Food and Drug Administration Modernization Act of 1997 (FDAMA), and the Medical Device User Fee and Modernization Act of 2002. The agency is classifying this device into class II (special controls) in order to provide a reasonable assurance of safety and effectiveness of the device. Elsewhere in this issue of the Federal Register, FDA is publishing a notice of availability of a guidance document that is the special control for this device.
Assuntos
Segurança de Equipamentos/classificação , Fator V/genética , Técnicas Genéticas/instrumentação , Hematologia/instrumentação , Patologia/instrumentação , Mutação Puntual/genética , Trombofilia/genética , Aprovação de Equipamentos/legislação & jurisprudência , Desenho de Equipamento , Fator V/classificação , Técnicas Genéticas/classificação , Testes Genéticos/classificação , Testes Genéticos/legislação & jurisprudência , Hematologia/classificação , Humanos , Patologia/classificação , Kit de Reagentes para Diagnóstico/classificação , Trombofilia/sangue , Estados Unidos , United States Food and Drug AdministrationRESUMO
The necessity of the study of technogenic stimuli effects on humans resulting in ecological pathology and ecological nosology is emphasized. Clinicomorphological aspects of latent health disturbances, their structural basis (ecological pathology), i.e. qualitative and quantitative spectrum of initial alterations in the subcellular structures and cells are considered. The examples of ecological diseases produced by specific chemical stimuli are presented. It is suggested to name ecological diseases with terms at the basis of which are etiological stimuli. As illustration statistical data are provided (clinical and pathological) on the role of occupational xenobiotics in the onset of coronary heart disease at young age and its sequelae.
Assuntos
Saúde Ambiental/tendências , Patologia/tendências , Terminologia como Assunto , Doença das Coronárias/induzido quimicamente , Humanos , Doenças Profissionais/induzido quimicamente , Patologia/classificação , Xenobióticos/efeitos adversosRESUMO
We report on basic design decisions and novel annotation procedures underlying the development of PathoJen, a corpus of Medline abstracts annotated for pathological phenomena, including diseases as a proper subclass. This named entity type is known to be hard to delineate and capture by annotation guidelines. We here propose a two-category encoding schema where we distinguish short from long mention spans, the first covering standardized terminology (e.g. diseases), the latter accounting for less structured descriptive statements about norm-deviant states, as well as criteria and observations that might signal pathologies. The second design decision relates to the way annotation instances are sampled. Here we subscribe to an Active Learning-based approach which is known to save annotation costs without sacrificing annotation quality by means of a sample bias. By design, Active Learning picks up 'hard' to annotate instances for human annotators, whereas 'easier' ones are passed over to the automatic classifier whose models already incorporate and gradually improve with previous annotation experience.
Assuntos
Algoritmos , Inteligência Artificial , Patologia/classificação , Aprendizagem Baseada em Problemas , Humanos , MEDLINERESUMO
OBJECTIVE: Our study aimed to construct and evaluate functions called "classifiers", produced by supervised machine learning techniques, in order to categorize automatically pathology reports using solely their content. METHODS: Patients from the Poitou-Charentes Cancer Registry having at least one pathology report and a single non-metastatic invasive neoplasm were included. A descriptor weighting function accounting for the distribution of terms among targeted classes was developed and compared to classic methods based on inverse document frequencies. The classification was performed with support vector machine (SVM) and Naive Bayes classifiers. Two levels of granularity were tested for both the topographical and the morphological axes of the ICD-O3 code. The ability to correctly attribute a precise ICD-O3 code and the ability to attribute the broad category defined by the International Agency for Research on Cancer (IARC) for the multiple primary cancer registration rules were evaluated using F1-measures. RESULTS: 5121 pathology reports produced by 35 pathologists were selected. The best performance was achieved by our class-weighted descriptor, associated with a SVM classifier. Using this method, the pathology reports were properly classified in the IARC categories with F1-measures of 0.967 for both topography and morphology. The ICD-O3 code attribution had lower performance with a 0.715 F1-measure for topography and 0.854 for morphology. CONCLUSION: These results suggest that free-text pathology reports could be useful as a data source for automated systems in order to identify and notify new cases of cancer. Future work is needed to evaluate the improvement in performance obtained from the use of natural language processing, including the case of multiple tumor description and possible incorporation of other medical documents such as surgical reports.
Assuntos
Informática Médica/organização & administração , Neoplasias/patologia , Patologia/classificação , Sistema de Registros , Inteligência Artificial , França/epidemiologia , Humanos , Classificação Internacional de Doenças , Neoplasias/epidemiologia , SemânticaRESUMO
La Tomografía por emisión de positrones/tomografía computada (PET/CT) se ha vuelto fundamental para la evaluación oncológica. En los últimos años se ha hecho evidente su utilidad para evaluar otras patologías inflamatorias no neoplásicas, las cuales pueden presentar aumento del metabolismo medible. El PET/CT tiene la ventaja de poder detectar enfermedades incluso cuando no tienen un correlato en las imágenes morfológicas, permitiendo además localizar de manera precisa estas alteraciones. Entre estas patologías se encuentran el estudio de fiebre de origen desconocido, enfermedades inflamatorias, enfermedades del tejido conectivo, vasculitis y también en el seguimiento y diagnóstico de algunas patologías infecciosas. Se realizará una revisión en la literatura de la utilidad del PET/CT en estas patologías complementada con casos clínicos.