Bevan Koopman1, Guido Zuccon2, Anthony Nguyen3, Anton Bergheim4, Narelle Grayson5. 1. The Australian e-Health Research Centre, CSIRO, Brisbane, Australia. Electronic address: bevan.koopman@csiro.au. 2. Queensland University of Technology, Brisbane, Australia. Electronic address: g.zuccon@qut.edu.au. 3. The Australian e-Health Research Centre, CSIRO, Brisbane, Australia. Electronic address: Anthony.Nguyen@csiro.au. 4. Cancer Institute NSW, Sydney, Australia. Electronic address: anton.bergheim@cancerinstitute.org.au. 5. Cancer Institute NSW, Sydney, Australia. Electronic address: Narelle.Grayson@cancerinstitute.org.au.
Abstract
OBJECTIVE: Death certificates provide an invaluable source for cancer mortality statistics; however, this value can only be realised if accurate, quantitative data can be extracted from certificates--an aim hampered by both the volume and variable nature of certificates written in natural language. This paper proposes an automatic classification system for identifying cancer related causes of death from death certificates. METHODS: Detailed features, including terms, n-grams and SNOMED CT concepts were extracted from a collection of 447,336 death certificates. These features were used to train Support Vector Machine classifiers (one classifier for each cancer type). The classifiers were deployed in a cascaded architecture: the first level identified the presence of cancer (i.e., binary cancer/nocancer) and the second level identified the type of cancer (according to the ICD-10 classification system). A held-out test set was used to evaluate the effectiveness of the classifiers according to precision, recall and F-measure. In addition, detailed feature analysis was performed to reveal the characteristics of a successful cancer classification model. RESULTS: The system was highly effective at identifying cancer as the underlying cause of death (F-measure 0.94). The system was also effective at determining the type of cancer for common cancers (F-measure 0.7). Rare cancers, for which there was little training data, were difficult to classify accurately (F-measure 0.12). Factors influencing performance were the amount of training data and certain ambiguous cancers (e.g., those in the stomach region). The feature analysis revealed a combination of features were important for cancer type classification, with SNOMED CT concept and oncology specific morphology features proving the most valuable. CONCLUSION: The system proposed in this study provides automatic identification and characterisation of cancers from large collections of free-text death certificates. This allows organisations such as Cancer Registries to monitor and report on cancer mortality in a timely and accurate manner. In addition, the methods and findings are generally applicable beyond cancer classification and to other sources of medical text besides death certificates.
OBJECTIVE: Death certificates provide an invaluable source for cancer mortality statistics; however, this value can only be realised if accurate, quantitative data can be extracted from certificates--an aim hampered by both the volume and variable nature of certificates written in natural language. This paper proposes an automatic classification system for identifying cancer related causes of death from death certificates. METHODS: Detailed features, including terms, n-grams and SNOMED CT concepts were extracted from a collection of 447,336 death certificates. These features were used to train Support Vector Machine classifiers (one classifier for each cancer type). The classifiers were deployed in a cascaded architecture: the first level identified the presence of cancer (i.e., binary cancer/nocancer) and the second level identified the type of cancer (according to the ICD-10 classification system). A held-out test set was used to evaluate the effectiveness of the classifiers according to precision, recall and F-measure. In addition, detailed feature analysis was performed to reveal the characteristics of a successful cancer classification model. RESULTS: The system was highly effective at identifying cancer as the underlying cause of death (F-measure 0.94). The system was also effective at determining the type of cancer for common cancers (F-measure 0.7). Rare cancers, for which there was little training data, were difficult to classify accurately (F-measure 0.12). Factors influencing performance were the amount of training data and certain ambiguous cancers (e.g., those in the stomach region). The feature analysis revealed a combination of features were important for cancer type classification, with SNOMED CT concept and oncology specific morphology features proving the most valuable. CONCLUSION: The system proposed in this study provides automatic identification and characterisation of cancers from large collections of free-text death certificates. This allows organisations such as Cancer Registries to monitor and report on cancer mortality in a timely and accurate manner. In addition, the methods and findings are generally applicable beyond cancer classification and to other sources of medical text besides death certificates.
Authors: Anthony N Nguyen; Donna Truran; Madonna Kemp; Bevan Koopman; David Conlan; John O'Dwyer; Ming Zhang; Sarvnaz Karimi; Hamed Hassanzadeh; Michael J Lawley; Damian Green Journal: AMIA Annu Symp Proc Date: 2018-12-05
Authors: José Carlos Ferrão; Mónica Duarte Oliveira; Filipe Janela; Henrique M G Martins; Daniel Gartner Journal: Health Syst (Basingstoke) Date: 2020-03-01
Authors: Markus Kreuzthaler; Bastian Pfeifer; Jose Antonio Vera Ramos; Diether Kramer; Victor Grogger; Sylvia Bredenfeldt; Markus Pedevilla; Peter Krisper; Stefan Schulz Journal: BMC Med Inform Decis Mak Date: 2019-04-04 Impact factor: 2.796