Supplementary Table S5: Machine learning assisting neurosurgical care of spine patients

Authors / Year / Title / Journal / Stage within neurosurgical care / Algorithm / Input features / Application / Results/Conclusion
SEP monitoring
Fan, B., Li, H. X. and Hu, Y. / 2016 / An Intelligent Decision System for Intraoperative Somatosensory Evoked Potential Monitoring / IEEE Trans Neural SystRehabilEng / Intraoperative Assistance / Multiple (SVM, K-Means) / Multiple (MRI, SEP monitoring) / Improve specificity of SEP monitoring / The results suggest that the proposed decision system has better performance, especially in the false positive cases, and may be more effective in the trauma case.
Merzagora, A. C., Bracchi, F., Cerutti, S., Rossi, L., Gaggiani, A. and Bianchi, A. M. / 2007 / Evaluation and application of a RBF neural network for online single-sweep extraction of SEPs during scoliosis surgery / IEEE Transactions on Biomedical Engineering / Intraoperative Assistance / GMM / SEP monitoring / Improve specificity of SEP monitoring / The proposed algorithm has been proved to be particularly effective and suitable for single-sweep detection. It is able to track both sudden and smooth signal changes of both amplitude and latency and the needed computational time is moderate.
Predict decision making
Azimi, P., Mohammadi, H. R., Benzel, E. C., Shahzadi, S. and Azhari, S. / 2014 / Use of artificial neural networks to decision making in patients with lumbar spinal canal stenosis / J NeurosurgSci / Presurgical Planning / Multiple (ANN, LR) / Multiple (Clinical, MRI) / Predict surgical decision making / The ANN model displayed better accuracy rate (97.8 %), a better H--L statistic (41.1 %) which represented a good--fit calibration, and a better AUC (89.0%), compared to the LR model.
Gal, N., Stoicu-Tivadar, V., Andrei, D., Nemeş, D. I. and Nădăşan, E. / 2014 / Computer assisted treatment prediction of low back pain pathologies / Studies in health technology and informatics / Presurgical Planning / FCM / Clinical / Predict surgical decision making / The initial results are promising; there is a correlation of 0.83% between the control results and the results from the system.
Outcome prediction
Azimi, P., Benzel, E. C., Shahzadi, S., Azhari, S. and Mohammadi, H. R. / 2014 / Use of artificial neural networks to predict surgical satisfaction in patients with lumbar spinal canal stenosis: clinical article / J Neurosurg Spine / Neurosurgical outcome prediction / Multiple (ANN, LR) / Multiple (Clinical, MRI) / Predict patient satisfaction after surgery / The ANN model displayed a better accuracy rate in 96.9% of patients, a better Hosmer-Lemeshow statistic in 42.4% of patients, and a better ROC-AUC in 80% of patients, compared with the LR model.
Azimi, P., Mohammadi, H. R., Benzel, E. C., Shahzadi, S. and Azhari, S. / 2015 / Use of artificial neural networks to predict recurrent lumbar disk herniation / J Spinal Disord Tech / Neurosurgical outcome prediction / Multiple (ANN, LR) / Clinical / Predict recurrence after surgery / Compared with the logistic regression model, the ANN model was associated with superior results: accuracy rate, 94.1%; Hosmer-Lemeshow statistic, 40.2%; and area under the curve, 0.83% of patients.
Azimi, P., Benzel, E. C., Shahzadi, S., Azhari, S. and Mohammadi, H. R. / 2016 / The prediction of successful surgery outcome in lumbar disc herniation based on artificial neural networks / J NeurosurgSci / Neurosurgical outcome prediction / Multiple (ANN, LR) / Clinical / Predict symptom improvement after surgery / Compared to the LR model, the ANN model showed better results: accuracy rate, 95.8%; H-L statistic, 41.5%; and AUC, 0.82% of patients.
Hoffman, H., Lee, S. I., Garst, J. H., Lu, D. S., Li, C. H., Nagasawa, D. T., Ghalehsari, N., Jahanforouz, N., Razaghy, M., Espinal, M., Ghavamrezaii, A., Paak, B. H., Wu, I., Sarrafzadeh, M. and Lu, D. C. / 2015 / Use of multivariate linear regression and support vector regression to predict functional outcome after surgery for cervical spondylotic myelopathy / J ClinNeurosci / Neurosurgical outcome prediction / Multiple (SVM, MLR) / Multiple (Clinical, MRI) / Predict symptom improvement after surgery / With the SVR model, a combination of preoperative ODI score, preoperative mean absolute accuracy (sinusoidal function), and symptom duration yielded the best prediction of postoperative ODI (R(2)=0.932; mean absolute difference=0.0283; p=5.73 x 10(-12)).
Shamim, M. S., Enam, S. A. and Qidwai, U. / 2009 / Fuzzy Logic in neurosurgery: predicting poor outcomes after lumbar disk surgery in 501 consecutive patients / SurgNeurol / Neurosurgical outcome prediction / FIS / Multiple (Clinical, MRI) / Predict symptom improvement after surgery / Fuzzy inference system has a sensitivity of 88% and specificity of 86% in the prediction of patients most likely to have poor outcome after lumbosacral miscrodiskectomy. The test thus has a positive predictive value of 0.36 and a negative predictive value of 0.98.
Segmentation
Preetha, J. and Selvarajan, S. / 2016 / Computer aided diagnostic system for automatic cervical disc herniation classification / Journal of Medical Imaging and Health Informatics / Presurgical Planning / SVM / MRI / Segmentation of cervical vertebrae / Experimental results show that the proposed method improves the classification accuracy.
Zhu, X., He, X., Wang, P., He, Q., Gao, D., Cheng, J. and Wu, B. / 2016 / A method of localization and segmentation of intervertebral discs in spine MRI based on Gabor filter bank / BioMedical Engineering Online / Presurgical Planning / GMM / MRI / Segmentation of cervical vertebrae / The proposed method is verified by an MRI dataset consisting of 278 intervertebral discs from 37 patients. The accuracy of localization is 98.23% and the dice similarity index for segmentation evaluation is 0.9237.
Daenzer, S., Freitag, S., von Sachsen, S., Steinke, H., Groll, M., Meixensberger, J. and Leimert, M. / 2014 / VolHOG: a volumetric object recognition approach based on bivariate histograms of oriented gradients for vertebra detection in cervical spine MRI / Med Phys / Presurgical Planning / SVM / MRI / Segmentation of cervical vertebrae / The results indicate that the proposed algorithm achieves good results despite the presence of severe image noise.
Kadoury, S., Labelle, H. and Paragios, N. / 2011 / Automatic inference of articulated spine models in CT images using high-order Markov Random Fields / Med Image Anal / Presurgical Planning / BL / CT / Segmentation of cervical vertebrae / Quantitative comparison to expert identification yields an accuracy of 1.8+/-0.7mm based on the localization of surgical landmarks.
Galbusera, F., Bassani, T., Costa, F., Brayda-Bruno, M., Zerbi, A. and Wilke, H. J. / 2016 / Artificial neural networks for the recognition of vertebral landmarks in the lumbar spine / Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization / Presurgical Planning / ANN / MRI / Segmentation of lumbar disk / The novel method proved to be able to identify vertebral landmarks, with errors and limitations which should be taken into account.
Yu, S., Tan, K. K., Sng, B. L., Li, S. and Sia, A. T. / 2015 / Lumbar Ultrasound Image Feature Extraction and Classification with Support Vector Machine / Ultrasound Med Biol / Intraoperative Assistance / ANN / Ultrasound / Segmentation of lumbar disk / A success rate of 95.0% on training set and 93.2% on test set was achieved with the proposed method. The trained SVM model was further tested on 46 off-line collected videos, and successfully identified the proper needle insertion site (interspinous region) in 45 of the cases.
Ghosh, S. and Chaudhary, V. / 2014 / Supervised methods for detection and segmentation of tissues in clinical lumbar MRI / Computerized Medical Imaging and Graphics / Presurgical Planning / RF / MRI / Segmentation of lumbar disk / RF showed very promising results for both disc detection (98% disc localization accuracy and 2.08. mm mean deviation) and sagittal MRI segmentation (dice similarity indices of 0.87 and 0.84 for the dural sac and the inter-vertebral disc, respectively).

Abbreviations: ANN: Artificial Neural Networks; BCI: Brain; Computer Interface; BL: Bayesion Learning; CT: Computer Tomography; DBS: Deep Brain Stimulation; DT: Decision Tree; EEG: Electro-EncephaloGraphy; FCM: Fuzzy C-Means; FIS: Fuzzy Inference System; GA: Genetic Algorithm; GB: Gradient Boosting; GMM: Gaussian Mixture Models; ICP: Intracranial Pressue; iEEG: Intracranial Electro-EncephaloGraphy; KNN: K-Nearest Neighbors; LDA: Linear Discriminant Analysis; LR: Logistic Regression; MERs: Micro-Electrode Recordings; MRI: Magnetic Resonance Imaging; NLP: Natural language processing; OLS: Ordinary Least Squares; PCA: Principal Component Analysis; PET: Positrion Emission Tomography; QDA: Quadratic Discriminant Analysis; RF: Random Forests; SEP: Somatosensory Evoked Potential; SVM: Support Vector Machine