Classification of psychosocial risk factors (yellow flags) for the development of chronic low back and leg pain using artificial neural network

Due to international guidelines of treatment of acute low back pain (LBP), psychosocial risk factors like depressive mood and maladaptive pain-related coping strategies ('yellow flags') have to be assessed in the early phase of acute pain. Within this longitudinal study in patients with LBP and leg pain, we used an artificial neural network (ANN) to classify the pain intensity 6 months after the onset of treatment. Psychosocial risk factors were used as input neurons. The training of the three-layer ANN using the back-propagation algorithm yields to an accuracy of 83.1%. Further on, the complexity of this structure indicated the necessity of an early screening procedure that allows a differentiation between several high risk groups and a low risk profile in order to predict the long-term development of pain intensity.

[1]  S. Linton,et al.  Can we screen for problematic back pain? A screening questionnaire for predicting outcome in acute and subacute back pain. , 1998, The Clinical journal of pain.

[2]  M. Jayson,et al.  The Lumbar Spine and Back Pain , 1993 .

[3]  H. Weinberg,et al.  Motor cortex activity and predicting side of movement: neural network and dipole analysis of pre-movement magnetic fields , 1995, Neuroscience Letters.

[4]  D. Hallner,et al.  [Psychological mechanisms in the transition from acute to chronic pain: over- or underrated?]. , 2001, Schmerz.

[5]  P J Drew,et al.  Artificial intelligence for clinicians , 1999, Journal of the Royal Society of Medicine.

[6]  M. Hasenbring,et al.  The efficacy of a risk factor-based cognitive behavioral intervention and electromyographic biofeedback in patients with acute sciatic pain. An attempt to prevent chronicity. , 1999, Spine.

[7]  W. Baxt.  Use of an artificial neural network for the diagnosis of myocardial infarction. , 1991, Annals of internal medicine.

[8]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[9]  S J Linton,et al.  A review of psychological risk factors in back and neck pain. , 2000, Spine.

[10]  Tetsuya Takahashi.  Analyses on the temporal patterns of spikes of auditory neurons of the macaque monkey by means of an artificial neural network and tree-based models , 1997, Neuroscience Letters.

[11]  M. Hasenbring,et al.  Attentional control of pain and the process of chronification. , 2000, Progress in brain research.

[12]  M. Hasenbring,et al.  Risk Factors of Chronicity in Lumbar Disc Patients: A Prospective Investigation of Biologic, Psychologic, and Social Predictors of Therapy Outcome , 1994 .

[13]  M. Hasenbring.  Durchhaltestrategien—ein in Schmerzforschung und Therapie vernachlässigtes Phänomen? , 1993, Der Schmerz.

[14]  A. Beck,et al.  An inventory for measuring depression. , 1961, Archives of general psychiatry.

[15]  A. Silman,et al.  Short-term physical risk factors for new episodes of low back pain. Prospective evidence from the South Manchester Back Pain Study. , 1999, Spine.