Gold Standard for Epilepsy/Tumor Surgery Coupled with Deep Learning Offers Independence to a Promising Functional Mapping Modality

Part of the SpringerBriefs in Electrical and Computer Engineering book series (BRIEFSELECTRIC)


RATIONALE: Electrocorticography-based functional language mapping (ECoG-FLM) utilizes an ECoG signal paired with simultaneous language task presentation to create functional maps of the eloquent language cortex in patients selected for resective epilepsy or tumor surgery. At present, the concordance of functional maps derived by ECoG-FLM and electrical cortical stimulation mapping (ESM) remains rather low. This impedes the transition of ECoG-FLM into an independent functional mapping modality. As ESM is considered the gold standard of functional mapping, we aimed to use it in combination with machine learning (ML) approaches (“ESM-ML guide”), to improve the accuracy of ECoG-FLM. METHODS: The ECoG data was collected from 6 patients (29.67 ± 12.5 yrs; 19–52 yrs; 3 males, 3 females). Patient ECoG activity was recorded (g.USBamp, g.tec, Austria) during administration of language tasks. For data analysis: (1) All ECoG sites were divided into ESM positive [ESM(+)] and ESM negative [ESM(−)]; (2) Features of ESM(+) and ESM(−) sites in the ECoG signal were determined by analyzing the signal in the frequency domain; (3) ML classifiers [Random Forest (RF) and Deep Learning (DL)] were trained to identify these features in language-related ECoG activity; (4) The accuracy of the ESM-ML guided classification was compared with the accuracy of the conventional ECoG-FLM. RESULTS: The conventional approach demonstrated: 58% accuracy, 22% sensitivity, and 78% specificity. The “ESM-ML guide” approach with RF classifier demonstrated: 76.2% accuracy, 73.6% sensitivity and 78.78% specificity. The DL classifier achieved the highest performances compared to all others with 83% accuracy, 84% sensitivity and 83% specificity. CONCLUSION: ECoG-FLM accuracy can be improved by using an “ESM-ML guide”, making the use of ECoG-FLM feasible as a stand-alone methodology. The long-term goal is to create a tool-box with “ready to use an ESM-ML guide” algorithm trained to provide high accuracy ECoG-FLM results by classifying between ESM(+) and ESM(−) contacts in prospective sets of language-related ECoG data and, thus, contribute towards improved surgical outcomes.


Brain computer interface (BCI) Deep learning (DL) Electrocorticography (ECoG) Epilepsy surgery Functional brain mapping High-gamma mapping Machine learning (ML) Passive mapping Language mapping 



The authors acknowledge The Central Florida Health Research (CFHR) grant (PIs: Drs. M. Korostenskaja and U. Bagci) for supporting this study. Dr. Korostenskaja also would like to express her gratitude to Drs. G. Schalk and W. Wang for their feed-back/criticism/contribution to the development of this project’s idea for submission to NIH.


  1. 1.
    G. Schalk et al., Brain-computer interfaces (BCIs): detection instead of classification. J. Neurosci. Methods 167(1), 51–62 (2008)CrossRefGoogle Scholar
  2. 2.
    G. Schalk et al., Real-time detection of event-related brain activity. Neuroimage 43(2), 245–249 (2008)CrossRefGoogle Scholar
  3. 3.
    R. Arya et al., Presurgical language localization with visual naming associated ECoG high—gamma modulation in pediatric drug-resistant epilepsy. Epilepsia 58(4), 663–673 (2017)CrossRefGoogle Scholar
  4. 4.
    C. Kapeller et al., CortiQ-based real-time functional mapping for epilepsy surgery. J. Clin. Neurophysiol. 32(3), e12–22 (2015)CrossRefGoogle Scholar
  5. 5.
    M. Korostenskaja et al., Real-time functional mapping with electrocorticography in pediatric epilepsy: comparison with fMRI and ESM findings. Clin. EEG Neurosci. 45(3), 205–211 (2014)CrossRefGoogle Scholar
  6. 6.
    T. Kambara et al., Presurgical language mapping using event-related high-gamma activity: The Detroit procedure. Clin. Neurophysiol. 129(1), 145–154 (2018)CrossRefGoogle Scholar
  7. 7.
    Y. Tamura et al., Passive language mapping combining real-time oscillation analysis with cortico-cortical evoked potentials for awake craniotomy. J. Neurosurg. 125(6), 1580–1588 (2016)CrossRefGoogle Scholar
  8. 8.
    Y. Cho-Hisamoto et al., Cooing- and babbling-related gamma-oscillations during infancy: intracranial recording. Epilepsy Behav. 23(4), 494–496 (2012)CrossRefGoogle Scholar
  9. 9.
    R. Arya et al., Electrocorticographic language mapping in children by high-gamma synchronization during spontaneous conversation: comparison with conventional electrical cortical stimulation. Epilepsy Res. 110, 78–87 (2015)CrossRefGoogle Scholar
  10. 10.
    P.R. Bauer et al., Mismatch between electrocortical stimulation and electrocorticography frequency mapping of language. Brain Stimul. 6(4), 524–531 (2013)CrossRefGoogle Scholar
  11. 11.
    K. Kamada et al., Disconnection of the pathological connectome for multifocal epilepsy surgery. J. Neurosurg. 1–13 (2017)Google Scholar
  12. 12.
    H. Ogawa et al., Rapid and minimum invasive functional brain mapping by realtime visualization of high gamma activity during awake craniotomy. World Neurosurg. (2014)Google Scholar
  13. 13.
    N.E. Crone, A. Sinai, A. Korzeniewska, High-frequency gamma oscillations and human brain mapping with electrocorticography. Prog. Brain Res. 159, 275–295 (2006)CrossRefGoogle Scholar
  14. 14.
    M. Korostenskaja et al., Real-time functional mapping: potential tool for improving language outcome in pediatric epilepsy surgery. J. Neurosurg. Pediatr. 14(3), 287–295 (2014)CrossRefGoogle Scholar
  15. 15.
    P. Brunner et al., A practical procedure for real-time functional mapping of eloquent cortex using electrocorticographic signals in humans. Epilepsy Behav. 15(3), 278–286 (2009)CrossRefGoogle Scholar
  16. 16.
    C.D. Wray et al., Multimodality localization of the sensorimotor cortex in pediatric patients undergoing epilepsy surgery. J. Neurosurg. Pediatr. 10(1), 1–6 (2012)CrossRefGoogle Scholar
  17. 17.
    M. Korostenskaja et al., Electrocorticography-based real-time functional mapping for pediatric epilepsy surgery. J. Pediatr. Epilepsy 04(04), 184–206 (2015)CrossRefGoogle Scholar
  18. 18.
    X. Zhang et al., Surgical treatment for epilepsy involving language cortices: a combined process of electrical cortical stimulation mapping and intra-operative continuous language assessment. Seizure 22(9), 780–786 (2013)CrossRefGoogle Scholar
  19. 19.
    G. Ojemann et al., Cortical language localization in left, dominant hemisphere. An electrical stimulation mapping investigation in 117 patients. J. Neurosurg. 71(3), 316–26 (1989)CrossRefGoogle Scholar
  20. 20.
    S. Visa, A. Ralescu, Data-driven fuzzy sets for classification. Int. J. Adv. Intell. Paradigms 1(1), 3–30 (2008)CrossRefGoogle Scholar
  21. 21.
    T.M. Rutkowski et al., Multichannel spectral pattern separation—an EEG processing application. in IEEE International Conference on Acoustics, Speech and Signal Processing, 2009. ICASSP 2009 (2009)Google Scholar
  22. 22.
    W. Wang et al., An electrocorticographic brain interface in an individual with tetraplegia. PLoS ONE 8(2), e55344 (2013)CrossRefGoogle Scholar
  23. 23.
    C. Kapeller et al., Single trial detection of hand poses in human ECoG using CSP based feature extraction. Conf. Proc. IEEE Eng. Med. Biol. Soc. 2014, 4599–4602 (2014)Google Scholar
  24. 24.
    M. Korostenskaja et al., Improving ECoG-based P300 speller accuracy, in Proceedings of the 6th International Brain-Computer Interface Conference 2014, vol. 088 (2014), pp. 1–4Google Scholar
  25. 25.
    G. Schalk, E.C. Leuthardt, Brain-computer interfaces using electrocorticographic signals. IEEE Rev. Biomed. Eng. 4, 140–154 (2011)CrossRefGoogle Scholar
  26. 26.
    A. Ralescu, K.H. Lee, M. Korostenskaja, Machine learning techniques provide with a new insight into pre-attentive information processing changes in pediatric intractable epilepsy, in Society for Psychophysiological Research 51st Annual Meeting 2011 (Boston, MA, USA, 2011), p. 92Google Scholar
  27. 27.
    M. Korostenskaja et al., Gold standard for epilepsy/tumor surgery coupled with deep learning offers independence to a promising functional mapping modality, in Submission for Annual International BCI2017 Award (Nominee for 2017: (2017)
  28. 28.
    M. Korostenskaja et al., Gold standard for epilepsy/tumor surgery coupled with deep learning offers independence to a promising functional mapping modality, in Poster presentation at the Annual International BCI2017 Award ceremony during the 7th Graz Brain Computer Interface Conference 2017, Graz, Austria, 18–22 September 2017Google Scholar
  29. 29.
    M. Korostenskaja et al., ESM-guided approach supported by machine learning improves accuracy of ECoG-based functional language mapping, in American Epilepsy Society Annual Meeting 2017 (Washington, DC, USA, 2017), p. Abst. 1.110Google Scholar
  30. 30.
    H. RaviPrakash et al., Automatic response assessment in regions of language cortex in epilepsy patients using ECoG-based functional mapping and machine learning, in 2017 IEEE International Conference on Systems, Man, and Cybernetics (SMC) (2017), pp. 519–524Google Scholar
  31. 31.
    M. Korostenskaja et al., Characterization of cortical motor function and imagery-related cortical activity: potential application for prehabilitation, in 2017 IEEE International Conference on Systems, Man, and Cybernetics (SMC) (2017), pp. 3014–3019Google Scholar
  32. 32.
    G. Schalk et al., BCI2000: a general-purpose brain-computer interface (BCI) system. IEEE Trans. Biomed. Eng. 51(6), 1034–1043 (2004)CrossRefGoogle Scholar
  33. 33.
    C. Kapeller et al., cortiQ—based real-time functional mapping for epilepsy surgery. J. Clin. Neurophysiol. 32(3), e12–22 (2015)CrossRefGoogle Scholar
  34. 34.
    N.A. Rahman, A course in theoretical statistics: for sixth forms, technical colleges, colleges of education, universities (Charles Griffin & Company Limited, 1968)Google Scholar
  35. 35.
    R. Arya, P.S. Horn, N.E. Crone, ECoG high-gamma modulation versus electrical stimulation for presurgical language mapping. Epilepsy Behav. 79, 26–33 (2018)CrossRefGoogle Scholar
  36. 36.
    R. Prueckl et al., Passive functional mapping guides electrical cortical stimulation for efficient determination of eloquent cortex in epilepsy patients. in IEEE Biomedical Conference 2017 Proceedings (2017)Google Scholar
  37. 37.
    M. Elsayed et al., Additive Potential of Real-Time Functional Mapping (RTFM) to Electrical Stimulation Mapping (ESM) results for epilepsy surgery candidates, in American Epilepsy Society Meeting 2014 (Seattle, Washington, USA, 2014), p. 3.276Google Scholar
  38. 38.
    R. Prueckl et al., O202 Combining the strengths of passive functional mapping and electrical cortical stimulation. Clin. Neurophysiol. 128(9), e243 (2017)CrossRefGoogle Scholar
  39. 39.
    C. Kapeller et al., Electrocorticography guides electrical cortical stimulation to identify the eloquent cortex, in American Epilepsy Society Annual Meeting 2017 (Washington, DC, USA, 2017), p. Abst. 1.123Google Scholar
  40. 40.
    M. Elsayed et al., Additive Potential of Real-Time Functional Mapping (RTFM) to Electrical Stimulation Mapping (ESM) Results for Epilepsy Surgery Candidates, in Poster for American Epilepsy Society Meeting 2014 (Seattle, Washington, USA, 2014)Google Scholar
  41. 41.
    K.G. Davies, G.L. Risse, J.R. Gates, Naming ability after tailored left temporal resection with extraoperative language mapping: increased risk of decline with later epilepsy onset age. Epilepsy Behav. 7(2), 273–278 (2005)CrossRefGoogle Scholar
  42. 42.
    B.P. Hermann et al., Visual confrontation naming following left anterior temporal lobectomy: a comparison of surgical approaches. Neuropsychology 13(1), 3–9 (1999)CrossRefGoogle Scholar
  43. 43.
    M.J. Hamberger et al., Brain stimulation reveals critical auditory naming cortex. Brain 128(Pt 11), 2742–2749 (2005)CrossRefGoogle Scholar
  44. 44.
    M. Genetti et al., Comparison of high gamma electrocorticography and fMRI with electrocortical stimulation for localization of somatosensory and language cortex. Clin. Neurophysiol. 126(1), 121–130 (2015)CrossRefGoogle Scholar
  45. 45.
    K. Kojima et al., Gamma activity modulated by picture and auditory naming tasks: intracranial recording in patients with focal epilepsy. Clin. Neurophysiol. 124(9), 1737–1744 (2013)CrossRefGoogle Scholar
  46. 46.
    K.J. Miller et al., Rapid online language mapping with electrocorticography. J. Neurosurg. Pediatr. 7(5), 482–490 (2011)CrossRefGoogle Scholar
  47. 47.
    A. Sinai et al., Electrocorticographic high gamma activity versus electrical cortical stimulation mapping of naming. Brain 128(Pt 7), 1556–1570 (2005)CrossRefGoogle Scholar
  48. 48.
    M.C. Cervenka et al., Language mapping in multilingual patients: electrocorticography and cortical stimulation during naming. Front. Hum. Neurosci. 5, 13 (2011)CrossRefGoogle Scholar
  49. 49.
    M. Korostenskaja et al., Predicting post-surgical language outcome with ECoG-based real-time functional mapping (RTFM) in patients with pharmaco-resistant epilepsy, in American Epilepsy Society 68th Annual Meeting 2014 (Seattle, Washington, USA, 2014), p. 2.255Google Scholar
  50. 50.
    R. Prueckl et al., Passive functional mapping guides electrical cortical stimulation for efficient determination of eloquent cortex in epilepsy patients, in 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (2017)Google Scholar
  51. 51.
    G. Hickok, D. Poeppel, The cortical organization of speech processing. Nat. Rev. Neurosci. 8(5), 393–402 (2007)CrossRefGoogle Scholar
  52. 52.
    A. Strauss et al., Alpha and theta brain oscillations index dissociable processes in spoken word recognition. Neuroimage 97, 387–395 (2014)CrossRefGoogle Scholar
  53. 53.
    E. Halgren et al., Laminar profile of spontaneous and evoked theta: rhythmic modulation of cortical processing during word integration. Neuropsychologia (2015)Google Scholar
  54. 54.
    J. Xiang et al., High frequency oscillations in pediatric epilepsy: methodology and clinical application. J. Pediatr. Epilepsy 04(04), 156–164 (2015)CrossRefGoogle Scholar
  55. 55.
    A.L. Ko et al., Identifying functional networks using endogenous connectivity in gamma band electrocorticography. Brain Connect 3(5), 491–502 (2013)CrossRefGoogle Scholar
  56. 56.
    J. Xiang et al., Multi-frequency localization of aberrant brain activity in autism spectrum disorder. Brain Dev. 38(1), 82–90 (2016)CrossRefGoogle Scholar
  57. 57.
    P. Gabriel et al., Neural correlates to automatic behavior estimations from RGB-D video in epilepsy unit. Conf. Proc. IEEE Eng. Med. Biol. Soc. 2016, 3402–3405 (2016)Google Scholar

Copyright information

© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Functional Brain Mapping and Brain Computer Interface LabOrlandoUSA
  2. 2.MEG Center, Florida Hospital for ChildrenOrlandoUSA
  3. 3.Florida Epilepsy Center, Florida HospitalOrlandoUSA
  4. 4.Center for Research in Computer Vision, University of Central FloridaOrlandoUSA
  5. 5.g.tec Medical Engineering GmbHSchiedlbergAustria
  6. 6.EECS DepartmentUniversity of CincinnatiCincinnatiUSA
  7. 7.Xiang Research Lab, Division of NeurologyCincinnati Children’s Hospital Medical CenterCincinnatiUSA
  8. 8.Harvard Medical SchoolBeth Israel Deaconess Medical CenterBostonUSA

Personalised recommendations