Identification of Relevant Diffusion MRI Metrics Impacting Cognitive Functions Using a Novel Feature Selection Method

Mild Traumatic Brain Injury (mTBI) is a significant public health problem. The most troubling symptoms after mTBI are cognitive complaints. Studies show measurable differences between patients with mTBI and healthy controls with respect to tissue microstructure using diffusion MRI. However, it remains unclear which diffusion measures are the most informative with regard to cognitive functions in both the healthy state as well as after injury. In this study, we use diffusion MRI to formulate a predictive model for performance on working memory based on the most relevant MRI features. As exhaustive search is impractical, the key challenge is to identify relevant features over a large feature space with high accuracy within reasonable time-frame. To tackle this challenge, we propose a novel improvement of the best first search approach with crossover operators inspired by genetic algorithm. Compared against other heuristic feature selection algorithms, the proposed method achieves significantly more accurate predictions and yields clinically interpretable selected features (improvement of r2 in 8 of 9 cohorts and up to 0.08).

[1]  Haim Schweitzer,et al.  Optimal Column Subset Selection by A-Star Search , 2015, AAAI.

[2]  J. Doran,et al.  Experiments with the Graph Traverser program , 1966, Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences.

[3]  George McCloskey,et al.  Assessment with the WAIS-IV , 2008 .

[4]  Shervin Minaee,et al.  MTBI Identification From Diffusion MR Images Using Bag of Adversarial Visual Features , 2018, IEEE Transactions on Medical Imaging.

[5]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[6]  B. S. Manjunath,et al.  Unsupervised 3-D Feature Learning for Mild Traumatic Brain Injury , 2016, BrainLes@MICCAI.

[7]  J. Franklin,et al.  The elements of statistical learning: data mining, inference and prediction , 2005 .

[8]  Huimin Lu,et al.  Facial Emotion Recognition Based on Biorthogonal Wavelet Entropy, Fuzzy Support Vector Machine, and Stratified Cross Validation , 2016, IEEE Access.

[9]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[10]  Josef Kittler,et al.  Floating search methods in feature selection , 1994, Pattern Recognit. Lett..

[11]  Jihoon Yang,et al.  Feature Subset Selection Using a Genetic Algorithm , 1998, IEEE Intell. Syst..

[12]  M. Wald,et al.  Traumatic brain injury in the United States; emergency department visits, hospitalizations, and deaths, 2002-2006 , 2010 .

[13]  Navdeep Jaitly,et al.  Adversarial Autoencoders , 2015, ArXiv.

[14]  R. Kikinis,et al.  A review of magnetic resonance imaging and diffusion tensor imaging findings in mild traumatic brain injury , 2012, Brain Imaging and Behavior.

[15]  Y. Lui,et al.  Altered Relationship between Working Memory and Brain Microstructure after Mild Traumatic Brain Injury , 2019, American Journal of Neuroradiology.

[16]  J. Langlois,et al.  Traumatic brain injury in the United States; emergency department visits, hospitalizations, and deaths , 2006 .

[17]  Roland Bammer,et al.  Mild Traumatic Brain Injury: Is Diffusion Imaging Ready for Primetime in Forensic Medicine? , 2010, Topics in magnetic resonance imaging : TMRI.

[18]  Xin-She Yang,et al.  A wrapper approach for feature selection based on Bat Algorithm and Optimum-Path Forest , 2014, Expert Syst. Appl..

[19]  Joseph A. Helpern,et al.  White matter characterization with diffusional kurtosis imaging , 2011, NeuroImage.

[20]  Olivier Salvado,et al.  Statistical machine learning to identify traumatic brain injury (TBI) from structural disconnections of white matter networks , 2016, NeuroImage.

[21]  Suzanne Polinder,et al.  Divergent Classification Methods of Post-Concussion Syndrome after Mild Traumatic Brain Injury: Prevalence Rates, Risk Factors, and Functional Outcome. , 2018, Journal of neurotrauma.

[22]  Ron Kohavi,et al.  Wrappers for Feature Subset Selection , 1997, Artif. Intell..

[23]  Y. Lui,et al.  Prediction of longterm outcome of neuropsychological tests of MTBI patients using imaging features , 2013, 2013 IEEE Signal Processing in Medicine and Biology Symposium (SPMB).

[24]  G. Johnson,et al.  Short-term DTI predictors of cognitive dysfunction in mild traumatic brain injury , 2008, Brain injury.

[25]  Y. Lui,et al.  Classification algorithms using multiple MRI features in mild traumatic brain injury , 2014, Neurology.

[26]  Ferat Sahin,et al.  A survey on feature selection methods , 2014, Comput. Electr. Eng..

[27]  Jens H Jensen,et al.  Evaluating kurtosis‐based diffusion MRI tissue models for white matter with fiber ball imaging , 2017, NMR in biomedicine.

[28]  Yvonne W. Lui,et al.  White Matter Tract Integrity: An Indicator of Axonal Pathology after Mild Traumatic Brain Injury. , 2017, Journal of neurotrauma.