Natural and Artificial Intelligence in Neurosurgery: A Systematic Review

Senders, Joeky T.; Arnaout, Omar; Karhade, Aditya V.; Dasenbrock, Hormuzdiyar H.; Gormley, William B.; Broekman, Marike L D; Smith, Timothy R.

doi:10.1093/neuros/nyx384

Cited by 205 publications

(148 citation statements)

References 52 publications

Supporting

Mentioning

144

Contrasting

Unclassified

Order By: Relevance

“…First, although there were some reported neurosurgical applications of ML, most of them used ML for analyzing radiological data, such as MRI images [49]. This is the first study to use ML for the prediction of TSS outcomes.…”

Section: Discussionmentioning

confidence: 99%

Prediction of Recurrence after Transsphenoidal Surgery for Cushing’s Disease: The Use of Machine Learning Algorithms

Liu

Hong

et al. 2019

Neuroendocrinology

View full text Add to dashboard Cite

Background: There are no reliable predictive models for recurrence after transsphenoidal surgery (TSS) for Cushing’s disease (CD). Objectives: This study aimed to develop machine learning (ML)-based predictive models for CD recurrence after initial TSS and to evaluate their performance. Method: A total of 354 CD patients were included in this retrospective, supervised learning, data mining study. Predictive models for recurrence were developed according to 17 variables using 7 algorithms. Models were evaluated based on the area under the receiver operating characteristic curve (AUC). Results: All patients were followed up for over 12 months (mean ± SD 43.80 ± 35.61). The recurrence rate was 13.0%. Age (p < 0.001), postoperative morning serum cortisol nadir (p = 0.002), and postoperative (p < 0.001) and preoperative (p = 0.04) morning adrenocorticotropin (ACTH) level were significantly related to recurrence. AUCs of the 7 models ranged from 0.608 to 0.781. The best performance (AUC = 0.781, 95% CI 0.706, 0.856) appeared when 8 variables were introduced to the random forest (RF) algorithm, which was much better than that of logistic regression (AUC = 0.684, p = 0.008) and that of using only postoperative morning serum cortisol (AUC = 0.635, p < 0.001). According to the feature selection algorithms, the top 3 predictors were age, postoperative serum cortisol, and postoperative ACTH. Conclusions: Using ML-based models for prediction of the recurrence after initial TSS for CD is feasible, and RF performs best. The performance of most of ML-based models was significantly better than that of some conventional models.

show abstract

Section: Discussionmentioning

confidence: 99%

Prediction of Recurrence after Transsphenoidal Surgery for Cushing’s Disease: The Use of Machine Learning Algorithms

Liu

Hong

et al. 2019

Neuroendocrinology

View full text Add to dashboard Cite

show abstract

“…Artificial intelligence techniques are being rapidly deployed in neuroradiology and radiomics [1,2], many of which may benefit from new image markers to describe and characterize regions of interest (ROI), quite apart from new computerized parameters which could quantify specific features in an objective way.…”

Section: Introductionmentioning

confidence: 99%

Advanced computational and statistical multiparametric analysis of Susceptibility-Weighted Imaging to characterize gliomas and brain metastases

Ieva

Russo

Reste³

et al. 2020

Preprint

View full text Add to dashboard Cite

contributed equally to this work.Abstract. Susceptibility-weighted imaging (SWI) is a technique useful for evaluation of the internal structures of brain tumors, including microvasculature and microbleeds. Intratumoral patterns of magnetic susceptibility can be quantified by means of fractal analysis. Here, we propose a radiomics methodological pipeline to merge advanced fractal-based computational modelling with statistical analysis to objectively characterize the fingerprint of gliomas and brain metastases. Forty-seven patients with glioma (grades II-IV, according to the WHO 2016 classification system) and fourteen with brain metastases underwent 3 Tesla MRI using a SWI protocol. All images underwent computational analysis aimed to quantify three Euclidean parameters (related to tumor and SWI volume) and five fractal-based parameters (related to the pixel distribution and geometrical complexity of the SWI patterns). Principal components analysis, linear and quadratic discriminant analysis, K-nearest neighbor and support vector machine methods were used to discriminate between tumor types. The combination of parameters offered an objective evaluation of the SWI pattern in gliomas and brain metastases. The model accurately predicted 88% of glioblastoma, according to the quantification of intratumoral SWI features, failing to discriminate the other types. SWI is not normally used to classify brain tumors, however fractal-based multi-parametric computational analysis can be used to characterize intratumoral SWI patterns to objectively quantify tumors-related features. Specific parameters still have to be identified to provide completely automatic computerized differential diagnosis.

show abstract

“…Many brain tumor patients undergoing neurosurgery face significant uncertainty regarding the outcome of surgery. Although average neurosurgical outcomes for patient cohorts can be predicted with a high degree of accuracy (Emblem et al, 2009(Emblem et al, , 2015Senders et al, 2017), the heterogeneity of brain tumors complicates predictions on an individual patient level.…”

Section: Introductionmentioning

confidence: 99%

Modeling brain dynamics after tumor resection using The Virtual Brain

Aerts

Schirner

Dhollander

et al. 2019

Preprint

View full text Add to dashboard Cite

Brain tumor patients scheduled for tumor resection often face significant uncertainty, as the outcome of neurosurgery is difficult to predict at the individual patient level. Recently, computational modeling of brain activity using so-called brain network models has been introduced as a promising tool for this purpose. However, brain network models first have to be validated, before they can be used to predict brain dynamics. In prior work, we optimized individual brain network model parameters to maximize the fit with empirical brain activity. In this study, we extend this line of research by examining the stability of fitted parameters before and after tumor resection, and compare it with baseline parameter variability using data from healthy control subjects. Based on these findings, we perform the first "virtual neurosurgery" analyses to evaluate the potential of brain network modeling in predicting brain dynamics after tumor resection.We find that brain network model parameters are relatively stable over time in brain tumor patients who underwent tumor resection, compared with baseline variability in healthy control subjects. In addition, we identify several robust associations between individually optimized model parameters, structural network topology and cognitive performance from pre-to post-operative assessment. Concerning the virtual neurosurgery analyses, we obtain promising results in some patients, whereas the predictive accuracy of the currently applied model is poor in others. These findings reveal interesting avenues for future research, as well as important limitations that warrant further investigation.

show abstract

Natural and Artificial Intelligence in Neurosurgery: A Systematic Review

Cited by 205 publications

References 52 publications

Prediction of Recurrence after Transsphenoidal Surgery for Cushing’s Disease: The Use of Machine Learning Algorithms

Prediction of Recurrence after Transsphenoidal Surgery for Cushing’s Disease: The Use of Machine Learning Algorithms

Advanced computational and statistical multiparametric analysis of Susceptibility-Weighted Imaging to characterize gliomas and brain metastases

Modeling brain dynamics after tumor resection using The Virtual Brain

Contact Info

Product

Resources

About