Save
Moving the neurosurgical frontiers with artificial intelligence: the era of machine learning
EANS Academy. Golubovic J. 09/26/19; 275954; EP12055
Jagos Golubovic
Jagos Golubovic

Access to this content is reserved for EANS members and attendees of this event. Click here to become an EANS member and gain your access to the full content of the EANS Academy


Abstract
Discussion Forum (0)
Rate & Comment (0)
Background: Classical statistical methods have been utilized for long in developing guidelines in the era of evidence based neurosurgery. Despite limitations in these methods (human error, nonlinear variables or large amount of data) one still uses them to strive toward optimal surgical decision-making. Proper identification of diagnosis or/and patient group is essential to achieve optimal result. Machine learning (ML) is an emerging tool that enables computers to analyse and learn from massive data sets via complex algorithms in order to make accurate further predictions. In the previous years ML has been widely used to help improve neurosurgical outcome or to make a diagnosis. This systematical review /comparison has a goal to compare ML methods of neurosurgical decision making to the classical ones by externally validating outcome/diagnosis in each of four biggest neurosurgical fields (oncology, vascular, spine and trauma).
Methods: A systematic search (keywords: machine learning and neurosurgery) in the PubMed database was performed to identify all potentially relevant studies in the past 5 years (up to January 31st 2019). Four ML algorithms have been then externally validated (using author´s databases) and compared to traditional prediction models.
Results: 308 studies found included ML algorithm models for outcome prediction or diagnosis assumption in patients undergoing neurosurgical treatment for oncology, spine pathology, cerebrovascular diseases, traumatic brain injury and functional area of neurosurgery. Based on prediction tasks and features included, ML models had high outcome/positive diagnostic prediction. ML models when compared to traditional regression methods had significantly better performance and showed improvement in accuracy in all fields of neurosurgical research which was also confirmed by external validation on our own data sets.
Conclusions: ML demonstrated superb outcome prediction/diagnostic accuracy in almost every neurosurgical field of expertise. ML, as a developing technique has a potential to replace classical statistical models and consecutively improve neurosurgical outcome.
Background: Classical statistical methods have been utilized for long in developing guidelines in the era of evidence based neurosurgery. Despite limitations in these methods (human error, nonlinear variables or large amount of data) one still uses them to strive toward optimal surgical decision-making. Proper identification of diagnosis or/and patient group is essential to achieve optimal result. Machine learning (ML) is an emerging tool that enables computers to analyse and learn from massive data sets via complex algorithms in order to make accurate further predictions. In the previous years ML has been widely used to help improve neurosurgical outcome or to make a diagnosis. This systematical review /comparison has a goal to compare ML methods of neurosurgical decision making to the classical ones by externally validating outcome/diagnosis in each of four biggest neurosurgical fields (oncology, vascular, spine and trauma).
Methods: A systematic search (keywords: machine learning and neurosurgery) in the PubMed database was performed to identify all potentially relevant studies in the past 5 years (up to January 31st 2019). Four ML algorithms have been then externally validated (using author´s databases) and compared to traditional prediction models.
Results: 308 studies found included ML algorithm models for outcome prediction or diagnosis assumption in patients undergoing neurosurgical treatment for oncology, spine pathology, cerebrovascular diseases, traumatic brain injury and functional area of neurosurgery. Based on prediction tasks and features included, ML models had high outcome/positive diagnostic prediction. ML models when compared to traditional regression methods had significantly better performance and showed improvement in accuracy in all fields of neurosurgical research which was also confirmed by external validation on our own data sets.
Conclusions: ML demonstrated superb outcome prediction/diagnostic accuracy in almost every neurosurgical field of expertise. ML, as a developing technique has a potential to replace classical statistical models and consecutively improve neurosurgical outcome.
Code of conduct/disclaimer available in General Terms & Conditions
Anonymous User Privacy Preferences

Strictly Necessary Cookies (Always Active)

MULTILEARNING platforms and tools hereinafter referred as “MLG SOFTWARE” are provided to you as pure educational platforms/services requiring cookies to operate. In the case of the MLG SOFTWARE, cookies are essential for the Platform to function properly for the provision of education. If these cookies are disabled, a large subset of the functionality provided by the Platform will either be unavailable or cease to work as expected. The MLG SOFTWARE do not capture non-essential activities such as menu items and listings you click on or pages viewed.


Performance Cookies

Performance cookies are used to analyse how visitors use a website in order to provide a better user experience.



Google Analytics is used for user behavior tracking/reporting. Google Analytics works in parallel and independently from MLG’s features. Google Analytics relies on cookies and these cookies can be used by Google to track users across different platforms/services.


Save Settings