Neural network and logistic regression diagnostic prediction models for giant cell arteritis: development and validation

Purpose To develop and validate neural network (NN) vs logistic regression (LR) diagnostic prediction models in patients with suspected giant cell arteritis (GCA). Design: Multicenter retrospective chart review. Methods An audit of consecutive patients undergoing temporal artery biopsy (TABx) for suspected GCA was conducted at 14 international medical centers. The outcome variable was biopsy-proven GCA. The predictor variables were age, gender, headache, clinical temporal artery abnormality, jaw claudication, vision loss, diplopia, erythrocyte sedimentation rate, C-reactive protein, and platelet level. The data were divided into three groups to train, validate, and test the models. The NN model with the lowest false-negative rate was chosen. Internal and external validations were performed. Results Of 1,833 patients who underwent TABx, there was complete information on 1,201 patients, 300 (25%) of whom had a positive TABx. On multivariable LR age, platelets, jaw claudication, vision loss, log C-reactive protein, log erythrocyte sedimentation rate, headache, and clinical temporal artery abnormality were statistically significant predictors of a positive TABx (P≤0.05). The area under the receiver operating characteristic curve/Hosmer–Lemeshow P for LR was 0.867 (95% CI, 0.794, 0.917)/0.119 vs NN 0.860 (95% CI, 0.786, 0.911)/0.805, with no statistically significant difference of the area under the curves (P=0.316). The misclassification rate/false-negative rate of LR was 20.6%/47.5% vs 18.1%/30.5% for NN. Missing data analysis did not change the results. Conclusion Statistical models can aid in the triage of patients with suspected GCA. Misclassification remains a concern, but cutoff values for 95% and 99% sensitivities are provided (https://goo.gl/THCnuU).

[1]  Miguel Angel Luque-Fernandez,et al.  cvauroc: Command to compute cross-validated area under the curve for ROC analysis after predictive modeling for binary outcomes , 2019, The Stata Journal: Promoting communications on statistics and Stata.

[2]  Matthias Schonlau,et al.  Support Vector Machines and logistic regression to predict temporal artery biopsy outcomes. , 2019, Canadian journal of ophthalmology. Journal canadien d'ophtalmologie.

[3]  D. Munoz,et al.  The incidence of giant cell arteritis in Ontario, Canada. , 2019, Canadian journal of ophthalmology. Journal canadien d'ophtalmologie.

[4]  A. Kirubarajan,et al.  Systematic Review of the Yield of Temporal Artery Biopsy for Suspected Giant Cell Arteritis , 2018, Neuro-ophthalmology.

[5]  A. Aouba,et al.  Mortality causes and trends associated with giant cell arteritis: analysis of the French national death certificate database (1980–2011) , 2018, Rheumatology.

[6]  A. Gill,et al.  Value of temporal artery biopsy length in diagnosing giant cell arteritis , 2018, ANZ journal of surgery.

[7]  Edsel Ing,et al.  The Use of a Nomogram to Visually Interpret a Logistic Regression Prediction Model for Giant Cell Arteritis , 2018, Neuro-ophthalmology.

[8]  A. Iagnocco,et al.  EULAR recommendations for the use of imaging in large vessel vasculitis in clinical practice , 2018, Annals of the rheumatic diseases.

[9]  C. Pagnoux,et al.  Does herpes zoster predispose to giant cell arteritis: a geo-epidemiologic study , 2018, Clinical ophthalmology.

[10]  P. Ranalli,et al.  Lower ocular pulse amplitude with dynamic contour tonometry is associated with biopsy-proven giant cell arteritis. , 2017, Canadian journal of ophthalmology. Journal canadien d'ophtalmologie.

[11]  E. Weis,et al.  Development of a predictive model for temporal artery biopsies. , 2017, Canadian journal of ophthalmology. Journal canadien d'ophtalmologie.

[12]  John J. Chen,et al.  Multivariable prediction model for suspected giant cell arteritis: development and validation , 2017, Clinical ophthalmology.

[13]  H. Danesh-Meyer,et al.  New developments in giant cell arteritis. , 2016, Survey of ophthalmology.

[14]  Ewout W Steyerberg,et al.  Net benefit approaches to the evaluation of prediction models, molecular markers, and diagnostic tests , 2016, British Medical Journal.

[15]  Gary S Collins,et al.  Sample size considerations for the external validation of a multivariable prognostic model: a resampling study , 2015, Statistics in medicine.

[16]  M. El-Dairi,et al.  Diagnostic Algorithm for Patients With Suspected Giant Cell Arteritis , 2015, Journal of neuro-ophthalmology : the official journal of the North American Neuro-Ophthalmology Society.

[17]  C. Kaufmann,et al.  Ocular pulse amplitude as a diagnostic adjunct in giant cell arteritis , 2015, Eye.

[18]  B. van Calster,et al.  Calibration of Risk Prediction Models , 2015, Medical decision making : an international journal of the Society for Medical Decision Making.

[19]  G. Collins,et al.  Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD Statement , 2015, BMC Medicine.

[20]  G. Rebolleda,et al.  A calculator for temporal artery biopsy result prediction in giant cell arteritis suspects. , 2014, European journal of internal medicine.

[21]  M. Sarossy,et al.  The use of statistical modeling to predict temporal artery biopsy outcome from presenting symptoms and laboratory results , 2014 .

[22]  A. Godra,et al.  Polyangiitis overlap syndrome with granulomatosis with polyangiitis (Wegener's) and giant cell arteritis. , 2013, Canadian journal of ophthalmology. Journal canadien d'ophtalmologie.

[23]  A. Karthikesalingam,et al.  Importance of specimen length during temporal artery biopsy , 2011, The British journal of surgery.

[24]  K. Barraclough,et al.  BSR and BHPR guidelines for the management of giant cell arteritis. , 2010, Rheumatology.

[25]  R. Nesher,et al.  Rate of Discordant Findings in Bilateral Temporal Artery Biopsy to Diagnose Giant Cell Arteritis , 2009, The Journal of Rheumatology.

[26]  S. Scherer,et al.  Race and ancestry in biomedical research: exploring the challenges , 2009, Genome Medicine.

[27]  H. Raspe,et al.  EULAR recommendations for the management of large vessel vasculitis , 2008, Annals of the rheumatic diseases.

[28]  R. L. Font,et al.  Histological parameters helpful in recognising steroid-treated temporal arteritis: an analysis of 35 cases , 2006, British Journal of Ophthalmology.

[29]  J Coste,et al.  Temporal artery biopsy for diagnosing giant cell arteritis: the longer, the better? , 2006, Annals of the rheumatic diseases.

[30]  Patrick Royston,et al.  The cost of dichotomising continuous variables , 2006, BMJ : British Medical Journal.

[31]  Ryan D Niederkohr,et al.  Management of the patient with suspected temporal arteritis a decision-analytic approach. , 2005, Ophthalmology.

[32]  Yvonne Vergouwe,et al.  Substantial effective sample sizes were required for external validation studies of predictive logistic regression models. , 2005, Journal of clinical epidemiology.

[33]  R. Cooper,et al.  Commentary: considerations for use of racial/ethnic classification in etiologic research. , 2001, American journal of epidemiology.

[34]  R. Malhotra,et al.  Efficacy of unilateral versus bilateral temporal artery biopsies for the diagnosis of giant cell arteritis. , 2000, American journal of ophthalmology.

[35]  S. Hayreh,et al.  Occult giant cell arteritis: ocular manifestations. , 1998, American journal of ophthalmology.

[36]  J. Leavitt,et al.  Systemic amyloidosis with temporal artery involvement mimicking temporal arteritis. , 1997, Ophthalmic surgery and lasers.

[37]  J V Tu,et al.  Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes. , 1996, Journal of clinical epidemiology.

[38]  M L Astion,et al.  Application of neural networks to the classification of giant cell arteritis. , 1994, Arthritis and rheumatism.

[39]  Javier Martín,et al.  Genetic component of giant cell arteritis. , 2014, Rheumatology.

[40]  Oguzhan Alagoz,et al.  Informatics in radiology: comparison of logistic regression and artificial neural network models in breast cancer risk estimation. , 2010, Radiographics : a review publication of the Radiological Society of North America, Inc.

[41]  E. L. Kelly Clinical versus statistical prediction: A theoretical analysis and review of the evidence. , 1955 .