Purpose: To compare the classification of psychiatric disorders using multinomial logistic regression (MLR) and artificial neural networks (ANNs).

Methods: This study collects data from hospitals in the cities of Salamanca and Zamora (Spain) for three years, configuring a file with psychiatric data with 572 cases and 52 variables, 11 of them referred to psychiatric disorders. The comparison between MLR and ANN is carried out by classification in 4 levels, with level 2 being the reference. We also rely on ANOVA to study the homogeneity of the classification for MLR and ANN.

Results: For study group A, in general, higher percentages of correct classification are obtained through ANNs, the same does not occur for groups B and C. The application of ANOVA for a 95% confidence, informs that there are no significant differences among the percentage averages obtained in the five levels for the MLR not being so for the percentage averages obtained through neural networks

Conclusions: The study shows that, by both techniques, women are better classified, that ANN classify the first group of psychiatric disorders (Group A) better than MLR and that the homogeneous classification is obtained by MLR.

psychiatric disorders, forensic psychiatry, multinomial logistic regression, artificial neural network, ANOVA.

Received: August 21, 2021; Accepted: September 29, 2021; Published:  October 8, 2021

How to cite this article: Elena Martín Pérez, Amaya Caldero Alonso and Quintín Martín Martín, Classification of psychiatric disorders using multinomial logistic regression versus artificial neural network, JP Journal of Biostatistics 18(3) (2021), 395-408. DOI: 10.17654/BS018010395

This Open Access Article is Licensed under Creative Commons Attribution 4.0 International License

Reference:

[1] T. Ayer, J. Chhatwal, O. Alagoz, C. E. Kahn, Jr., R. W. Woods and E. S. Burnside, Informatics in radiology: comparison of logistic regression and artificial neural network models in breast cancer risk estimation, Radiographics 30(1) (2010), 13-22.
[2] M. Barni, P. Faila and R. Lazzareti, Privacy-preserving ECG, classification with branching programs and neural networks, IEEE Transactions on Information Forensics and Security 6(2) (2011), 452-468.
[3] E. Bartfay, W. J. Mackillop and J. L. Pater, Comparing the predictive value of neural network models to logistic regression models on the risk of death for small-cell lung cancer patients, European Journal of Cancer Care (England) 15(2) (2006), 115-124.
[4] M. Behrman, R. Linder, A. H. Assadi, B. R. Stacey and M. M. Backonja, Classification of patients with pain based on neuropathic pain symptoms: comparison of an artificial neural network against an established scoring system, European Journal of Pain 11(4) (2007), 370-376.
[5] G. Caocci et al. Comparison between an artificial neural network and logistic regression in predicting acute graft-vs-host disease after unrelated donor hematopoietic stem cell transplantation in thalassemia patients, Experimental Hematology 38(5) (2010), 426-433.
[6] S. T. Chen et al. Comparative analysis of logistic regression, support vector machine and artificial neural network for the differential diagnosis of benign and malignant solid breast tumors by the use of three-dimensional power Doppler imaging, Korean Journal of Radiology 10(5) (2009), 464-471.
[7] P. du Jardin, J. Ponsaille, V. Alunni-Perret and G. Quatrehomme, A comparison between neural network and other metric methods to determine sex from the upper femur in a modern French population, Forensic Science International 192 (2009), 127.e1-127.e6.
[8] Elham Shafiei, Esmaeil Fakharian, Abdollah Omidi, Hossein Akbari, Ali Delpisheh and Arash Nademi, Comparison of artificial neural network and logistic regression models for prediction of psychological symptom six months after mild traumatic brain injury, Iran. J. Psychiatry Behav. Sci. 11 (2017), e5849.
[9] M. B. Fonseca et al., Bipolar and Schizophrenia disorders diagnosis using artificial neural network, Neuroscience & Medicine 9 (2018), 209-220.
[10] Filippo Amato, Alberto López, Eladia María Peña-Méndez, Petr Vaňhara, Aleš Hampl and Josef Havel, Artificial neural networks in medical diagnosis, Journal of Applied Biomedicine 11 (2013), 47-58.
[11] Guy C. M. Skinner and David P. Farrington, Antisocial personality versus GP reported and self-reported health outcomes, The Journal of Forensic Psychiatry & Psychology 32(4) (2021), 506-519.
[12] R. M. Harris, Using Artificial Neural Networks for Forensic File Type Identification, Center for Education and Research in Information Assurance and Security, Purdue University, West Lafayette, 2007.
[13] Hilleke Hulshoff Pol and Edward Bullmoreb, Neural networks in psychiatry, European Neuropsychopharmacology 23 (2013), 1-6.
[14] Hui Chen, Jing Zhang, Yan Xu, Budong Chen and Kuan Zhang, Performance comparison of artificial neural network and logistic regression model for differentiating lung nodules on CT scans, Expert Systems with Applications 39 (2012), 11503-11509.
[15] IBM SPSS 26 Manual “Neural Networks”, 2021.
[16] A. Kazemnejad, Z. Batvandi and J. Faradmal, Comparison of artificial neural network and binary logistic regression for determination of impaired glucose tolerance/diabetes, Eastern Mediterranean Health Journal 16(6) (2010), 615-620.
[17] C. C. Lin, Y. K. Ou, S. H. Chen, Y. C. Liu and J. Lin, Comparison of artificial neural network and logistic regression models for predicting mortality in elderly patients with hip fracture, Injury 41(8) (2010), 869-873.
[18] Manual of Diagnosis and Statistics of Mental Disorders (DSM-5), 5ª ed., Arlington, VA, American Psychiatric Association, 2014.
[19] M. A. Mazurowski, P. A. Habas, J. M. Zurada, J. Y. Lo, J. A. Baker and G. D. Tourassi, Training neural network classifiers for medical decision making: the effects of imbalanced datasets on classification performance, International Joint Conference on Neural Networks IJCNN’07, Orlando, Florida, USA, Vol. 21, 2008, pp. 427-436.
[20] Niels Peek, Carlo Combib, Roque Marin and Riccardo Bellazzi, Thirty years of artificial intelligence in medicine (AIME) conferences: a review of research themes, Artificial Intelligence in Medicine 65 (2015), 61-73.
[21] D. R. Pai, K. D. Lawrence, R. K. Klimberg and S. M. Lawrence, Experimental comparison of parametric, non-parametric, and hybrid multigroup classification, Expert Systems with Applications 39(10) (2012), 8593-8603.
[22] Tanya R. Peckmanna and Brooke Fisherb, Sex estimation from the patella in an African American population, Journal of Forensic and Legal Medicine 54 (2018), 1-7.
[23] Torgyn Shaikhina and Natalia A. Khovanova, Handling limited datasets with neural networks in medical applications: a small-data approach, Artificial Intelligence in Medicine 75 (2017), 51-63.
[24] Turker Ince, Serkan Kiranyaz, Jenni Pulkkinen and Moncef Gabbouj, Evaluation of global and local training techniques over feed-forward neural network architecture spaces for computer-aided medical diagnosis, Expert Systems with Applications 37 (2010), 8450-8461.
[25] T. W. Way et al., Computer-aided diagnosis of pulmonary nodules on CT scans: improvement of classification performance with nodule surface features, Medical Physics 36(7) (2009), 3086-3098.
[26] Jack V. Tu, Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes, Journal of Clinical Epidemiology 11 (1996), 1225-1231.
[27] Erhard Lang, Lawrence Pitts, Susan Damron and Robert Rutledge, Outcome after severe head injury: an analysis of prediction based upon comparison of neural network versus logistic regression analysis, Neurological Research 23(4) (1997), 274-280.
[28] N. Terrin, C. H. Schmid, J. L. Griffith, R. B. D’agostino and H. P. Selker, External validity of predictive models: a comparison of logistic regression, classification trees, and neural networks, Journal of Clinical Epidemiology 56 (2003), 721-729.
[29] Jae H Song, Santosh S. Venkatesh, Emily A. Conant, Peter H. Arger and Chandra M. Sehgal, Comparative analysis of logistic regression and artificial neural network for computer-aided diagnosis of breast masses, Academic Radiology 12(4) (2005), 487-495.
[30] Srinivasa Pai, T. N. Nagabhushana and Raj B. K. N. Rao, Tool condition monitoring using acoustic emission, surface roughness and growing cell structures neural network, Machining Science and Technology 16(4) (2012), 653-676.
[31] K. J. Ottenbacher, R. T. Linn, P. M. Smith, S. B. Illig, M. Mancuso and C. V. Granger, Comparison of logistic regression and neural network analysis applied to predicting living setting after hip fracture, Annals of Epidemiology 14(8) (2004), 551-559.
[32] Kenneth D. Lawrence and Ronald K. Klimberg, Advances in Business and Management Forecasting, Emerald Publishing Limited, 2017.
[33] B. Eftekhar et al., Comparison of artificial neural network and logistic regression models for prediction of mortality in head trauma based on initial clinical data, BMC Medical Informatics and Decision Making 5(3) (2005). https://doi.org/10.1186/1472-6947-5-3.
[34] F. Jaimes, J. Farbiarz, D. Alvarez and C. Martínez, Comparison between logistic regression and neural networks to predict death in patients with suspected sepsis in the emergency room, Critical Care 9 (2005), R150-R156.
[35] E. Lin, Y. Hwang, S. C. Wang, Z. J. Gu and E. Y. Chen, An artificial neural network approach to the drug efficacy of interfere on treatments, Pharmacogenomics 7(7) (2006), 1017-1024.