Objective: To examine the applicability of a neural network classification strategy to examine the independent contribution of psychomotor disturbance (PMD) and endogeneity symptoms to the DSM-III-R definition of melancholia.
Method: We studied 407 depressed patients with the clinical dataset comprising 17 endogeneity symptoms and the 18-item CORE measure of behaviourally rated PMD. A multilayer perceptron neural network was used to fit non-linear models of varying complexity. A linear discriminant function analysis was also used to generate a model for comparison with the non-linear models.
Results: Models (linear and non-linear) using PMD items only and endogeneity symptoms only had similar rates of successful classification, while non-linear models combining both PMD and symptom scores achieved the best classifications.
Conclusions: Our current non-linear model was superior to a linear analysis, a finding which may have wider application to psychiatric classification. Our non-linear analysis of depressive subtypes supports the binary view that melancholic and non-melancholic depression are separate clinical disorders rather than different forms of the same entity. This study illustrates how non-linear modelling with neural networks is a potentially fruitful approach to the study of the diagnostic taxonomy of psychiatric disorders and to clinical decision-making.
ParkerGHadzi-PavlovicDBoycePClassifying depression by mental state signs. British Journal of Psychiatry1990; 157:55–65.
2.
ParkerGHadzi-PavlovicDWilhelmKDefining melancholia: properties of a refined sign-based measure. British Journal of Psychiatry1994; 164:316–326.
3.
ParkerGHadzi-PavlovicD.Melancholia. A disorder of movement and mood. New York: Cambridge University Press, 1996.
4.
ParkerGHadzi-PavlovicDAustinM-PSub-typing depression. I. Is psychomotor disturbance necessary and sufficient to the definition of melancholia?Psychological Medicine1995; 25:815–823.
5.
RipleyBDNeural networks and related methods for classification. Journal of the Royal Statistical Society1994; 56:409–156.
6.
WeissSMKulikowskiCAComputer systems that learn. San Mateo, CA: Morgan Kaufman, 1991.
7.
ChengBTitteringtonDMNeural networks: a review from a statistical perspective. Statistical Science1994; 9:2–54.
8.
SarleWSNeural networks and statistical models. In: Proceedings of the Nineteenth Annual SAS Users Group International Conference, April 1994. Cary, NC: SAS Institute Inc., 1994.
9.
BishopCMNeural networks for pattern recognition. Oxford: Oxford University Press, 1995.
10.
FlorioTEinfeldSLLevyF.Neural networks and psychiatry: candidate application in clinical decision making. Australian and New Zealand Journal of Psychiatry1994; 28:651–666.
11.
CrossSSHarrisonRFKennedyRLIntroduction to neural networks. Lancet1995; 346:1075–1079.
12.
BaxtWGUse of an artificial neural network for data analysis in clinical decision-making: the diagnosis of acute coronary occlusion. Neural Computation1990; 2:480–489.
13.
AndererPSaletuBKloppelBSemlitschHWernerH.Discrimination between demented patients and normals based on topographic slow wave activity: comparison between z statistics, discriminant analysis and artificial neural network classifiers. Electroencephalography and Clinical Neurophysiology1994; 91:108–117.
14.
AkayMWelkowitzW.Accoustical detection of coronary occlusions using neural networks. Journal of Biomedical Engineering1993; 15:469–473.
15.
KippenhamJBarkerWPascalSNagelJDuaraR.Evaluation of a neural network classifier for PET scans of normal and Alzheimer's disease subjects. Journal of Nuclear Medicine1992; 33:1459–1467.
16.
CohenILSudhalterVLandon-JimenezDKeoghM.A neural network approach to the classification of autism. Journal of Autism and Developmental Disabilities1993; 3:443–466.
17.
CicchettiDVVolkmarFKlinAShowalterD.Diagnosing autism using ICD-10 criteria: a comparison of neural network and standard multivariate procedures. Child Neuropsychology1995; 1:26–37.
18.
American Psychiatric Association.Diagnostic and statistical manual of mental disorders. 3rd ed.Washington, DC: American Psychiatric Association, 1987.
19.
EfronBTibshiraniRJAn introduction to the bootstrap. New York: Chapman and Hall, 1993.
20.
McLachlanG.Discrimination and classification. Brisbane: University of Queensland Press, 1990.
21.
GoodmanP.NevProp2 [software package]. Reno, NV: University of Nevada Centre for Biomedical Modelling Research, 1994.
22.
ParkerGHadzi-PavlovicDHickieISub-typing depression III. Development of a clinical algorithm for melancholia and comparision with other diagnostic measures. Psychological Medicine1995; 25:833–840.
23.
DawesRMCorriganB.Linear models in decision making. Psychological Bulletin 81974; 1:95–106.
24.
WainerH.Estimating coefficients in linear models: it don't make no nevermind. Psychological Bulletin1976; 83:213–217.
25.
CohenJDServan-SchrieberD.A neural network model of disturbances in processing of context in schizophrenia. Psychiatric Annals1992; 22:131–136.
