ABSTRACT
Background: Pathology reports are stored as unstructured, ungrammatical, fragmented, and abbreviated free text with linguistic variability among pathologists. For this reason, tumor information extraction requires a significant human effort. Recording data in an efficient and high-quality format is essential in implementing and establishing a hospital-based-cancer registry Objective: This study aimed to describe implementing a natural language processing algorithm for oncology pathology reports. Methods: An algorithm was developed to process oncology pathology reports in Spanish to extract 20 medical descriptors. The approach is based on the successive coincidence of regular expressions. Results: The validation was performed with 140 pathological reports. The topography identification was performed manually by humans and the algorithm in all reports. The human identified morphology in 138 reports and by the algorithm in 137. The average fuzzy matching score was 68.3 for Topography and 89.5 for Morphology. Conclusions: A preliminary algorithm validation against human extraction was performed over a small set of reports with satisfactory results. This shows that a regular-expression approach can accurately and precisely extract multiple specimen attributes from free-text Spanish pathology reports. Additionally, we developed a website to facilitate collaborative validation at a larger scale which may be helpful for future research on the subject.
Introducción: Los reportes de patología están almacenados como texto libre sin estructura, gramática, fragmentados o abreviados, con variabilidad lingüística entre patólogos. Por esta razón, la extracción de información de tumores requiere un esfuerzo humano significativo. Almacenar información en un formato eficiente y de alta calidad es esencial para implementar y establecer un registro hospitalario de cáncer. Objetivo: Este estudio busca describir la implementación de un algoritmo de Procesamiento de Lenguaje Natural para reportes de patología oncológica. Métodos: Desarrollamos un algoritmo para procesar reportes de patología oncológica en Español, con el objetivo de extraer 20 descriptores médicos. El abordaje se basa en la coincidencia sucesiva de expresiones regulares. Resultados: La validación se hizo con 140 reportes de patología. La identificación topográfica se realizó por humanos y por el algoritmo en todos los reportes. La morfología fue identificada por humanos en 138 reportes y por el algoritmo en 137. El valor de coincidencias parciales (fuzzy matches) promedio fue de 68.3 para Topografía y 89.5 para Morfología. Conclusiones: Se hizo una validación preliminar del algoritmo contra extracción humana sobre un pequeño grupo de reportes, con resultados satisfactorios. Esto muestra que múltiples atributos del espécimen pueden ser extraídos de manera precisa de texto libre de reportes de patología en Español, usando un abordaje de expresiones regulares. Adicionalmente, desarrollamos una página web para facilitar la validación colaborativa a gran escala, lo que puede ser beneficioso para futuras investigaciones en el tema.
ABSTRACT
Background: Primary hepatocellular carcinoma (HCC) is a major health hazard and frequent cause of liver cancers accounting 90% of cancers of liver worldwide. It has high mortality, prevalence, and incidence rate in Sub-Saharan, South Africa, and South-east Asia. Its etiology is associated with infection, dietary habits, and lifestyle factors. Aims and Objectives: The present study was designed to discuss the various possible etiologies for high incidence of HCC in Western Arunachal Pradesh, India. Materials and Methods: Data were collected as one among 33 population-based cancer registries in India under national cancer registry program of national center for disease informatics and research, Indian Council of Medical Research between 2012 and 2014 in Tomo Riba Institute of Health and Medical Sciences, Naharlagun. Data were represented in frequency and percentage using descriptive statistics. Results: With 194 cases, HCC represented 13.5% of overall malignancies in the region. It is 3 times more common in males than in females. Age-adjusted incidence rate for men was 21.44 and for women was 7.05. Conclusion: Western Arunachal Pradesh reported high incidence of hepatocellular carcinoma in the world. This finding may be associated with high prevalence of hepatitis and alcoholism in the region and perhaps also associated with local food habits.
ABSTRACT
Background: Stomach cancer is among the most frequent, is a leading cause of mortality in low- and middle-income countries. Assessing its survival is important to guide evidence-based health policies. Aims: To estimate stomach cancer survival in Colombia (2014-2019) with data from the National Cancer Information System (NCIS) and in Cali with data from the Cali Population Cancer Registry (RPCC) (1998-2017). Methods: NCIS estimated the overall 3-year net survival for 8,549 people, while RPCC estimated 5-year net survival for 6,776 people. Results: The 3-year net survival was 36.8% (95% CI: 35.5-38.1). Net survival was higher in people with special insurance (61.7%; 95% CI: 44.8-74.8) or third payer (40.5%; 95% CI: 38.7-42.3) than state insurance (30.7%; 95% CI: 28.7-32.8). It was also higher in women and people diagnosed at early stages. Multivariable analysis showed consistency with survival estimations with a higher risk of death in men, people with state insurance, and diagnosed at advanced stages. In Cali, the 5-year net survival remained stable in men during the last 20 years. In women the 5-year net survival in women increased 8.60 percentage points, equivalent to a 50% increase compared to the 1998-2002 period. For 2013-17, it was 19.1% (95%CI: 16.2-22.2) in men, and 24.8% (95% CI: 20.4-29.3) in women. Conclusions: Population survival estimates from the RPCC were lower than those observed in the NCIS. The differences in their methods and scope can explain variability. Nevertheless, our findings could be complementary to improve cancer control planning in the country.
Antecedentes: El cáncer de estómago se encuentra entre los más frecuentes y es una de las principales causas de mortalidad en los países de ingresos bajos y medianos. Evaluar su supervivencia es importante para orientar las políticas de salud basadas en la evidencia. Objetivos: Estimar la supervivencia del cáncer de estómago en Colombia (2014-2019) con datos del Sistema Nacional de Información del Cáncer (NCIS) y en Cali con datos del Registro Poblacional de Cáncer de Cali (RPCC) (1998-2017). Métodos: El NCIS estimó la supervivencia neta a tres años para 8,549 personas y el RPCC la calculó a 5 años para 6,776 personas registradas en sus bases de datos. Resultados: La supervivencia neta a tres años en Colombia fue del 36.8% (IC 95%: 35.5-38.1). La supervivencia neta fue mayor en personas con seguro especial (61.7%; IC 95%: 44.8-74.8) o tercer pagador (40.5%; IC 95%: 38.7-42.3) que el seguro estatal (30.7%; IC 95%: 28.7-32.8). También fue mayor en mujeres y personas diagnosticadas en etapas tempranas. El análisis multivariable mostró consistencia con la estimación de supervivencia con mayor riesgo de muerte en hombres, personas con seguro estatal y diagnosticados en estadios avanzados. En Cali, la supervivencia neta a 5 años se mantuvo estable en los hombres durante los últimos 20 años. En las mujeres aumentó 8.60 puntos porcentuales, equivalente a un aumento del 50% en comparación con el período 1998-2002. Para el período 2013-17 fue 19.1% (IC 95%: 16.2-22.2) en los hombres y 24.8% (IC 95%: 20.4-29.3) en las mujeres. Conclusiones: Las estimaciones de supervivencia del RPCC fueron más bajas que las obtenidas por el NCIS. Las diferencias en sus métodos y alcance pueden explicar la variabilidad. Sin embargo, nuestros hallazgos pueden ser complementarios para mejorar la planificación del control del cáncer en el país..
ABSTRACT
There is a constant rise in cancer cases, but the trends and pattern vary according to the geographical region. The aim of this brief research was to present an update of all cancer incidences through age-adjusted rates and their changes in different regions of the country. The data for this study were obtained from published reports of 28 population-based cancer registries (2005–2014) in India. Among males, currently, East/Northeast region ranked first, on the basis of incidence of cancer cases. Out of 28 registries, 11 registries encountered lung cancer as the leading site. Currently, East/Northeast regions were ranked first on the basis of incidence of cancer cases among females. Our study showed that 20 registries among the 28 had breast cancer as the leading one. Thus, the present overview revealed that all cancers in both males and females are consistent and had a high incidence in East/Northeast region of the country.
ABSTRACT
Actualmente se realiza un diagnóstico anual de 650.000 nuevos casos de carcinoma escamoso de cabeza y cuello en el mundo, siendo el carcinoma escamoso de laringe una patología neoplásica que compete al otorrinolaringólogo. La incidencia mundial del cáncer escamoso de laringe se estima en 3,9 por cada 100.000 habitantes con una mortalidad general de 2,0 por cada 100.000 habitantes. En Chile el registro de cáncer se realiza en base a los cinco registros poblacionales de cáncer que existen. No se tienen datos exactos respecto a incidencia y mortalidad por carcinoma escamoso de laringe, siendo la estimación de la incidencia de 1,2 casos por cada 100.000 habitantes y la estimación de mortalidad ajustada por edad de 0,7 casos por cada 100.000 habitantes. Se han descrito diversos factores de riesgo ambientales y estilos de vida para este cáncer, por lo tanto, las estrategias de prevención primaria en salud son claves a la hora de generar un impacto en la incidencia del carcinoma escamoso de laringe.
The annual diagnosis of head and neck squamous cell carcinoma is 650,000 new cases. The laryngeal carcinoma is a malignant disease that should include an otolaryngologist in its evaluation. The global incidence of laryngeal carcinoma is estimated at 3.9 per 100,000 inhabitants with an overall mortality rate of 2.0 per 100,000 inhabitants. In Chile the cancer registry is based on the five population cancer registries that exist. There is no accurate data on incidence and mortality from laryngeal carcinoma, being an estimated incidence of 1.2 cases per 100,000 inhabitants and an age-adjusted mortality of 0.7 cases per 100.00 inhabitants. There have been described various environmental risk factors and lifestyles for this cancer, therefore, primary prevention strategies are key to generate an impact on the incidence of larynx carcinoma.
Subject(s)
Humans , Carcinoma, Squamous Cell/mortality , Laryngeal Neoplasms/mortality , Carcinoma, Squamous Cell/epidemiology , Carcinoma, Squamous Cell/prevention & control , Chile/epidemiology , Diseases Registries , Incidence , Laryngeal Neoplasms/epidemiology , Laryngeal Neoplasms/prevention & control , Risk FactorsABSTRACT
CONTEXT: Within India, the incidence of gallbladder cancer (GBC) is characterized by marked geographical variation; however, the reasons for these differences are unclear. AIMS: To evaluate the role of place of birth, length of residence, and effect of migration from high‑ to low‑risk region on GBC development. SETTINGS AND DESIGN: Population‑based cancer registries (PBCRs); case–control study. SUBJECTS AND METHODS: Data of PBCRs were used to demonstrate geographical variation in GBC incidence rates. A case–control study data examined the role of birth place, residence length, and effect of migration in etiology of GBC. STATISTICAL ANALYSIS: Rate ratios for different PBCRs were estimated using Chennai Cancer Registry as the reference population. Odds ratios (ORs) for developing GBC in a high‑risk region compared to a low‑risk region and associated 95% confidence interval (CI) were estimated through unconditional logistic regression models using case–control study. RESULTS: GBC shows marked variation in incidence with risk highest in Northeast regions and lowest in South India. OR of 4.82 (95% CI: 3.87–5.99) was observed for developing GBC for individuals born in a high‑risk region compared to those born in a low‑risk region after adjusting for confounders. A dose–response relationship with increased risk with increased length of residence in a high‑risk region was observed (OR lifetime 5.58 [95% CI: 4.42–7.05]; Ptrend ≤ 0.001). The risk persisted even if study participant migrated from high‑ to low‑risk region (OR = 1.36; 95% CI: 1.02–1.82). CONCLUSIONS: The present study signifies the importance of place of birth, length of stay, and effect of migration from high‑ to low‑risk region in the development of GBC. The data indicate role of environmental and genetic factors in etiology of disease.
ABSTRACT
Background: There is limited information on population-based cancer survival data in Latin America. Objective: To obtain estimates of survival for some cancers recognized as a public health priority in Colombia using data from the Cancer Registry of Cali for 1995-2004. Methods: All cancer cases for residents of Cali were included for the following sites: breast (3,984), cervix uteri (2,469), prostate (3,999), stomach (3,442) and lung (2,170). Five-year relative survival estimates were calculated using the approach described by Estève. Results: Five-year relative survival was 79% in patients with prostate cancer and 68% and 60% in women with breast or cervix uteri cancer, respectively. The cure fraction was 6% in subjects with lung cancer and 15% in those with stomach cancer. The probability of dying from breast or prostate cancer in people in the lower socio-economic strata (SES) was 1.8 and 2.6 times greater, respectively, when compared to upper SES, p <0.001. Excess mortality associated with cancer was independent of age in prostate or breast cancer. After adjusting for age, sex and SES, the risk of dying from breast, cervix uteri, prostate and lung cancer during the 2000-2004 period decreased 19%, 13%, 48% and 16%, respectively, when compared with the period of 1995-1999. There was no change in the prognosis for patients with stomach cancer. Conclusions: Survival for some kinds of cancer improved through the 1995-2004 period, however health care programs for cancer patients in Cali are inequitable. People from lower SES are the most vulnerable and the least likely to survive.
Antecedentes: En Latinoamérica existe poca información de estimaciones de supervivencia al cáncer basadas en estudios de población. Objetivo: Obtener estimaciones de supervivencia relativa (SR) para algunos tipos de cáncer reconocidos como prioridad de salud pública en Colombia con la información del Registro Poblacional de Cáncer de Cali. Métodos: Se incluyeron todos los casos de cáncer invasivo ocurridos en residentes de Cali durante el periodo 1994-2005 para las siguientes localizaciones: mama (3,984), cérvix (2,469), próstata (3,999), estómago (3,442) y pulmón (2,170). Las estimaciones de supervivencia relativa a cinco años se calcularon utilizando el método descrito por Estève. Resultados: La SR a cinco años fue 79% en pacientes con cáncer de próstata y 68% y 60% en mujeres con cáncer de mama y cérvix. La fracción de curación fue 6% en sujetos con cáncer de pulmón y 15% en aquellos con cáncer de estómago; en estos pacientes; la SR a cinco años fue 17%. La probabilidad de morir por cáncer de mama o próstata en personas de los ESE más bajos fue de 1.8 y 2.6 veces más, respectivamente, en comparación con los ESE altos, p <0.001. Después de ajustar por edad, sexo y ESE, el riesgo de morir por cáncer de mama, cérvix, próstata o pulmón en el período 2000-2004 se redujo 19%, 13% 52% y 16%, respectivamente, en comparación con el período 1995-1999. No hubo cambio en el pronóstico para los pacientes con cáncer de estómago. Conclusión: La supervivencia para algunos tipos de cáncer ha mejorado a través de los años 1995-2004, pero los programas de atención para los pacientes con cáncer en Cali son inequitativos. Las personas de ESE bajos son más vulnerables y tienen menos probabilidad de sobrevivir al cáncer.
Subject(s)
Aged , Female , Humans , Male , Middle Aged , Neoplasms/epidemiology , Population Surveillance , Colombia/epidemiology , Prognosis , Registries , Survival RateABSTRACT
OBJETIVO : revisar los avances en materia de registros poblacionales de cáncer en Colombia, Chile y Brasil. METODOLOGIA: búsqueda de literatura en español, portugués e inglés a través de lilacs, bireme, scielo, medline, isi, y sitios web especializados. RESULTADOS Colombia cuenta con 8 registros, 4 indexados ante la iacr (10% población nacional), y 4 aún sin indexar (18% población nacional); el registro de Cali con 50 años de existencia es referente regional. Tres registros chilenos indexados cubren el 6,2% de la población nacional y dos sin indexación abarcan un 7,2% de población adicional; el Registro Nacional de Cáncer Infantil de Chile cubre el 100% de la población menor de 15 años y opera desde el año 2006. Brasil tiene registros poblacionales en 20 ciudades que cubren el 19% de la población total, y operan en integración con los registros hospitalarios. DISCUSION Y CONCLUSION la cobertura poblacional en la región es baja en comparación con Europa y Norteamérica. Se evidencian avances en los tres países, en especial por la existencia de respaldo político y normativo. Es necesario ampliar la cobertura poblacional y continuar mejorando la calidad de los datos. El uso de tecnologías informáticas favorece el desarrollo y la integración de los registros.
OBJECTIVE: to review the progress in the area of population-based cancer registries in Colombia, Chile and Brazil. METHODOLOGY an ethnographic study whose participants were 20 mothers of children with Down's syndrome. Data were collected using semi-structured interviews and field observations. Similarly, this study included a categorical analysis and a literature review. RESULTS : Colombia has 8 records, 4 of them are indexed by the iacr (10% of the national population), and 4 are still not indexed (18% of the national population). The registry of Cali city is 50 years old and is considered a regional reference. Likewise, three Chilean registries are indexed and cover 6.2% of the national population. In addition, two unindexed registries cover a further 7.2% of the Chilean population. Also, the National Childhood Cancer Registry of Chile covers 100% of the population under 15 years and operates since 2006. Brazil, in turn, has population-based registries for 20 cities which cover 19% of the total population. These registries operate jointly with hospital registries. DISCUSSION AND CONCLUSION: population coverage in the region is low compared to Europe and North America. There is evidence of progress in all three countries, particularly the existence of political and regulatory support. It is necessary to expand population coverage and further improve data quality. The use of information technologies promotes the development and integration of registries.
Subject(s)
Humans , Breast FeedingABSTRACT
BACKGROUND: Cancer registries play a fundamental role in cancer control and multicenter collaborative research. Recently, the need for reassessment of cancer registry criteria has arisen due to the newly released 2010 World Health Organization (WHO) classification. Accordingly, development of new coding guidelines for cancer is necessary to improve the quality of cancer registries, as well as to prevent conflicts that may arise when seeking medical insurance compensation. METHODS: With funding from the Management Center for Health Promotion, 35 members of the Gastrointestinal Pathology Study Group and the Cancer Registration Committee of the Korean Society of Pathologists (KSP) participated in a second workshop for gastrointestinal tumor registration in Korea. RESULTS: The topics of gastric epithelial tumor, colonic intramucosal carcinoma, neuroendocrine tumor (NET), gastrointestinal stromal tumor (GIST) and appendiceal mucinous tumor were discussed for new coding guidelines. A survey was then conducted among 208 members of the KSP for a consensus of the guidelines proposed in the workshop. CONCLUSIONS: Although a few issues were set aside for further discussion, such as coding for non-gastric GIST and some types of NET, the members agreed upon most of the proposed guidelines. Therefore, we suggest using the newly revised International Classification of Diseases for Oncology, 3rd edition (ICD-O-3) coding guidelines for registering gastrointestinal tumors in Korea.
Subject(s)
Carcinoma, Neuroendocrine , Clinical Coding , Colon , Consensus , Financial Management , Gastrointestinal Neoplasms , Gastrointestinal Stromal Tumors , Health Promotion , Insurance , International Classification of Diseases , Korea , Mucins , Registries , World Health OrganizationABSTRACT
BACKGROUND: Cancer registration in Korea has a longer than 30-years of history, during which time cancer registration has improved and become well-organized. Cancer registries are fundamental for cancer control and multi-center collaborative research. However, there have been discrepancies in assigning behavior codes. Thus, we intend to propose appropriate behavior codes for the International Classification of Disease Oncology, 3rd edition (ICD-O-3) for microinvasive tumors of the ovary and breast not only to improve the quality of the cancer registry but also to prevent conflicts. METHODS: As in series I, two pathology study groups and the Cancer Registration Committee of the Korean Society of Pathologists (KSP) participated. To prepare a questionnaire on provisional behavior code, the relevant subjects were discussed in the workshop, and consensus was obtained by convergence of opinion from members of KSP. RESULTS: Microinvasive tumor of the breast should be designated as a microinvasive carcinoma which was proposed as malignant tumor (/3). Serous borderline tumor with microinvasion of the ovary was proposed as borderline tumor (/1), and mucinous borderline tumor with microinvasion of the ovary as either borderline (/1) or carcinoma (/3) according to the tumor cell nature. CONCLUSIONS: Some issues should be elucidated with the accumulation of more experience and knowledge. Here, however, we present our second proposal.
ABSTRACT
BACKGROUND: Cancer registration in Korea has a longer than 30-years of history, during which time cancer registration has improved and become well-organized. Cancer registries are fundamental for cancer control and multi-center collaborative research. However, there have been discrepancies in assigning behavior codes. Thus, we intend to propose appropriate behavior codes for the International Classification of Disease Oncology, 3rd edition (ICD-O-3) for microinvasive tumors of the ovary and breast not only to improve the quality of the cancer registry but also to prevent conflicts. METHODS: As in series I, two pathology study groups and the Cancer Registration Committee of the Korean Society of Pathologists (KSP) participated. To prepare a questionnaire on provisional behavior code, the relevant subjects were discussed in the workshop, and consensus was obtained by convergence of opinion from members of KSP. RESULTS: Microinvasive tumor of the breast should be designated as a microinvasive carcinoma which was proposed as malignant tumor (/3). Serous borderline tumor with microinvasion of the ovary was proposed as borderline tumor (/1), and mucinous borderline tumor with microinvasion of the ovary as either borderline (/1) or carcinoma (/3) according to the tumor cell nature. CONCLUSIONS: Some issues should be elucidated with the accumulation of more experience and knowledge. Here, however, we present our second proposal.
ABSTRACT
OBJECTIVES: An accurate estimation of cancer patients is the basis of epidemiological studies and health services. However in Korea, cancer patients visiting out-patient clinics are usually ruled out of such studies and so these studies are suspected of underestimating the cancer patient population. The purpose of this study is to construct a more complete, hospital-based cancer patient registry using multiple sources of medical information. METHODS: We constructed a cancer patient detection algorithm using records from various sources that were obtained from both the in-patients and out-patients seen at Asan Medical Center (AMC) for any reason. The medical data from the potentially incident cancer patients was reviewed four months after first being detected by the algorithm to determine whether these patients actually did or did not have cancer. RESULTS: Besides the traditional practice of reviewing the charts of in-patients upon their discharge, five more sources of information were added for this algorithm, i.e., pathology reports, the national severe disease registry, the reason for treatment, prescriptions of chemotherapeutic agents and radiation therapy reports. The constructed algorithm was observed to have a PPV of 87.04%. Compared to the results of traditional practice, 36.8% of registry failures were avoided using the AMC algorithm. CONCLUSIONS: To minimize loss in the cancer registry, various data sources should be utilized, and the AMC algorithm can be a successful model for this. Further research will be required in order to apply novel and innovative technology to the electronic medical records system in order to generate new signals from data that has not been previously used.
Subject(s)
Adult , Female , Humans , Male , Middle Aged , Hospitals , Medical Records , Neoplasms/diagnosis , Organizational Case Studies , Program Development , Registries , Republic of KoreaABSTRACT
BACKGROUND: Cancer registries are fundamental for cancer control and multicenter collaborative research. However, there have been discrepancies among pathologists in classifying cancer and assigning the codes according to the International Classification of Disease Oncology 3 (ICD-O3). To improve the quality of cancer registries as well as to prevent the conflict with medical insurance compensation, a guideline for the coding of cancer is mandatory. METHODS AND RESULTS: Funded by the Management Center for Health Promotion, 40 members of the Gastrointestinal Pathology Study Group and the Cancer Registration Committee of the Korean Society of Pathologists participated in the 1st workshop for gastrointestinal tumor registration. The subjects of gastric epithelial tumor, intramucosal carcinoma of the colon, carcinoid tumor, gastrointestinal stromal tumor and appendiceal mucinous tumor were discussed to create a guideline. A survey to obtain consensus for the guideline proposed by the workshop was carried out by the members of the Korean Society of Pathologists and 240 members completed the questionnaire. CONCLUSION: Although there are some issues to be discussed further, such as coding of high grade dysplasia/adenoma and intramucosal carcinoma of stomach and colon, the members agreed upon most parts of the proposed guideline. Therefore, we suggest using the ICD-O3 coding guideline for gastrointestinal tumor.