Prediction of cause of death from forensic autopsy reports using text classification techniques: A comparative study. (July 2018)

• Record Type: Journal Article
• Authors: Mujtaba, Ghulam; Shuib, Liyana; Raj, Ram Gopal; Rajandram, Retnagowri; Shaikh, Khairunisa
• Is Part Of: Journal of forensic and legal medicine. Volume 57(2018)
• Page Start: 41
• Page End: 50
• Publication Date: 2018-07
• Subjects: Forensic autopsy reports -- Postmortem reports -- Machine learning -- Feature engineering -- Automated text classification; Medical jurisprudence -- Periodicals; Forensic sciences -- Periodicals; Forensic Medicine -- Periodicals; Médecine légale -- Périodiques; Electronic journals; 614.1
• DOI: 10.1016/j.jflm.2017.07.001
• Languages: English
• ISSNs: 1752-928X

Abstract

Abstract: Objectives: Automatic text classification techniques are useful for classifying plaintext medical documents. This study aims to automatically predict the cause of death from free text forensic autopsy reports by comparing various schemes for feature extraction, term weighing or feature value representation, text classification, and feature reduction. Methods: For experiments, the autopsy reports belonging to eight different causes of death were collected, preprocessed and converted into 43 master feature vectors using various schemes for feature extraction, representation, and reduction. The six different text classification techniques were applied on these 43 master feature vectors to construct a classification model that can predict the cause of death. Finally, classification model performance was evaluated using four performance measures i.e. overall accuracy, macro precision, macro-F-measure, and macro recall. Results: From experiments, it was found that that unigram features obtained the highest performance compared to bigram, trigram, and hybrid-gram features. Furthermore, in feature representation schemes, term frequency, and term frequency with inverse document frequency obtained similar and better results when compared with binary frequency, and normalized term frequency with inverse document frequency. Furthermore, the chi-square feature reduction approach outperformed Pearson correlation, and information gain approaches. Finally, in text classification algorithms, support vector machine classifier outperforms random forest, Naive Bayes, k-nearest neighbor, decision tree, and ensemble-voted classifier. Conclusion: Our results and comparisons hold practical importance and serve as references for future works. Moreover, the comparison outputs will act as state-of-art techniques to compare future proposals with existing automated text classification techniques. Highlights: Compare various ATC techniques to determine CoD from forensic autopsy reports. Unigram feature extraction scheme outperformed bigram, trigram and hybrid schemes. For feature ranking chi-square outperformed information gain and Pearson Correlation. SVM outperformed kNN, J48, RF, NB and ensemble voted classifiers. Our findings could stimulate researchers to propose new techniques in this domain.