A case study of using artificial neural networks for classifying cause of death from verbal autopsy

Boulle, Andrew; Chandramohan, Daniel; Weller, Peter

doi:10.1093/ije/30.3.515

Cited by 42 publications

(42 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Contrary to prior reports, our findings suggest that the intra-rater reliability for classifying cause of death using PCVA is high [7,33]. Reliability of 3 out of 4 physicians was classified as ‘substantial’, and repeat CSMF estimates for common causes of death were similar to the original estimates.…”

Section: Discussioncontrasting

confidence: 96%

Verbal Autopsy: Evaluation of Methods to Certify Causes of Death in Uganda

et al. 2015

Self Cite

View full text Add to dashboard Cite

To assess different methods for determining cause of death from verbal autopsy (VA) questionnaire data, the intra-rater reliability of Physician-Certified Verbal Autopsy (PCVA) and the accuracy of PCVA, expert-derived (non-hierarchical) and data-driven (hierarchal) algorithms were assessed for determining common causes of death in Ugandan children. A verbal autopsy validation study was conducted from 2008-2009 in three different sites in Uganda. The dataset included 104 neonatal deaths (0-27 days) and 615 childhood deaths (1-59 months) with the cause(s) of death classified by PCVA and physician review of hospital medical records (the ‘reference standard’). Of the original 719 questionnaires, 141 (20%) were selected for a second review by the same physicians; the repeat cause(s) of death were compared to the original,and agreement assessed using the Kappa statistic.Physician reviewers’ refined non-hierarchical algorithms for common causes of death from existing expert algorithms, from which, hierarchal algorithms were developed. The accuracy of PCVA, non-hierarchical, and hierarchical algorithms for determining cause(s) of death from all 719 VA questionnaires was determined using the reference standard. Overall, intra-rater repeatability was high (83% agreement, Kappa 0.79 [95% CI 0.76-0.82]). PCVA performed well, with high specificity for determining cause of neonatal (>67%), and childhood (>83%) deaths, resulting in fairly accurate cause-specific mortality fraction (CSMF) estimates. For most causes of death in children, non-hierarchical algorithms had higher sensitivity, but correspondingly lower specificity, than PCVA and hierarchical algorithms, resulting in inaccurate CSMF estimates. Hierarchical algorithms were specific for most causes of death, and CSMF estimates were comparable to the reference standard and PCVA. Inter-rater reliability of PCVA was high, and overall PCVA performed well. Hierarchical algorithms performed better than non-hierarchical algorithms due to higher specificity and more accurate CSMF estimates. Use of PCVA to determine cause of death from VA questionnaire data is reasonable while automated data-driven algorithms are improved.

show abstract

Section: Discussioncontrasting

confidence: 96%

Verbal Autopsy: Evaluation of Methods to Certify Causes of Death in Uganda

et al. 2015

Self Cite

View full text Add to dashboard Cite

show abstract

“…Interpretation has largely relied on either expert assessment of the VA interviews by physicians, or the application of predetermined algorithms, often based on a decision-tree approach (4). Some researchers have reported attempts to use other methods, such as neural networks (5).…”

Section: Introductionmentioning

confidence: 99%

A probabilistic approach to interpreting verbal autopsies: methodology and preliminary validation in Vietnam

Byass

Huong

Minh

2003

Scand J Public Health

114

View full text Add to dashboard Cite

Scand J Public Health 2003; 31(Suppl. 62): 32-37 Aims: Verbal autopsy (VA) has become an important tool in the past 20 years for determining cause of death in communities where there is no routine registration. In many cases, expert physicians have been used to interpret the VA findings and so assign individual causes of death. However, this is time consuming and not always repeatable. Other approaches such as algorithms and neural networks have been developed in some settings. This paper aims to develop a method that is simple, reliable and consistent, which could represent an advance in VA interpretation. Methods: This paper describes the development of a Bayesian probability model for VA interpretation as an attempt to find a better approach. This methodology and a preliminary implementation are described, with an evaluation based on VA material from rural Vietnam. Results: The new model was tested against a series of 189 VA interviews from a rural community in Vietnam. Using this very basic model, over 70% of individual causes of death corresponded with those determined by two physicians, increasing to over 80% if those cases ascribed to old age or as being indeterminate by the physicians were excluded. Discussion: Although there is a clear need to improve the preliminary model and to test more extensively with larger and more varied datasets, these preliminary results suggest that there may be good potential in this probabilistic approach.

show abstract

“…The difference spectrum of a test sample (observed À simulated) can give information about the PCs used for simulation of that sample which in turn provide information about the changes in biochemical composition of the tissue. ANN consists of an input layer, one or more hidden layers and an output layer with variable number of processing elements (artificial neurons or nodes) in each layer [26][27][28]. Each node, with the exception of the input neurons, receives multiple weighted inputs and produces an output which is usually a nonlinear function of the inputs.…”

Section: Discussionmentioning

confidence: 99%

“…Three-layer networks are sufficient to design any nonlinear network [21,[26][27][28][29][30][31][32][33][34][35][36][37][38]. Therefore, in the present study we have used three-layered feed-forward network with one input layer, one hidden and one output layer.…”

Section: Back-propagation Algorithm Of Annmentioning

confidence: 99%

See 1 more Smart Citation

A pilot study on colonic mucosal tissues by fluorescence spectroscopy technique: Discrimination by principal component analysis (PCA) and artificial neural network (ANN) analysis

Kamath

D'souza

Mathew

et al. 2008

Journal of Chemometrics

View full text Add to dashboard Cite

Pulsed laser-induced autofluorescence spectra of pathologically certified normal and malignant colonic mucosal tissues were recorded at 325 nm excitation. The spectra were analysed using three different methods for discrimination purposes. First, all the spectra were subjected to the principal component analysis (PCA) and the discrimination between normal and malignant cases were achieved using parameters like, spectral residuals, Mahalanobis distance and scores of factors. Second, to understand the changes in tissue composition between the two classes (normal, and malignant), difference spectrum was constructed by subtracting mean spectrum of calibration set samples from simulated mean of all spectra of any one class (normal/malignant) and in third, artificial neural network (ANN) analysis was carried out on the same set of spectral data by training the network with spectral features like, mean, median, spectral residual, energy, standard deviation, number of peaks for different thresholds (100, 250 and 500) after carrying out 1st-order differentiation of the training set samples and discrimination between normal and malignant conditions were achieved. The specificity and sensitivity were determined in PCA and ANN analyses and they were found to be 100 and 91.3% in PCA, and 100 and 93.47% in ANN, respectively.

show abstract

A case study of using artificial neural networks for classifying cause of death from verbal autopsy

Cited by 42 publications

References 7 publications

Verbal Autopsy: Evaluation of Methods to Certify Causes of Death in Uganda

Verbal Autopsy: Evaluation of Methods to Certify Causes of Death in Uganda

A probabilistic approach to interpreting verbal autopsies: methodology and preliminary validation in Vietnam

A pilot study on colonic mucosal tissues by fluorescence spectroscopy technique: Discrimination by principal component analysis (PCA) and artificial neural network (ANN) analysis

Contact Info

Product

Resources

About