Utility of deep neural networks in predicting gross-total resection after transsphenoidal surgery for pituitary adenoma: a pilot study

An Artificial Neural Network Model for Predicting Postoperative Facial Nerve Outcomes After Vestibular Schwannoma Surgery

BACKGROUND AND OBJECTIVES

The emergence of machine learning models has significantly improved the accuracy of surgical outcome predictions. This study aims to develop and validate an artificial neural network (ANN) model for predicting facial nerve (FN) outcomes after vestibular schwannoma (VS) surgery using the proximal-to-distal amplitude ratio (P/D) along with clinical variables.

METHODS

This retrospective study included 71 patients who underwent VS resection between 2018 and 2022. At the end of surgery, the FN was stimulated at the brainstem (proximal) and internal acoustic meatus (distal) and the P/D was calculated. Postoperative FN function was assessed using the House-Brackmann grading system at discharge (short-term) and after 9-12 months (long-term). House-Brackmann grades I-II were considered good outcome, whereas grades III-VI were considered fair/poor. An ANN model was constructed, and the performance of the model was evaluated using the area under the ROC curve for internal validation and accuracy, sensitivity, specificity, and positive and negative predictive values for external validation.

RESULTS

The short-term FN outcome was grades I-II in 57.7% and grades III-VI in 42.3% of patients. Initially, a model using P/D had an area under the curve of 0.906 (internal validation) and an accuracy of 89.1% (95% CI: 68.3%-98.8%) (external validation) for predicting good vs fair/poor short-term FN outcomes. The model was then refined to include only muscles with a P/D with a proximal latency between 6 and 8 ms. This improved the accuracy to 100% (95% CI: 79%-100%). Integrating clinical variables (patient's age, tumor size, and preoperative HB grade) in addition to P/D into the model did not significantly improve the predative value. A model was then created to predict the long-term FN outcome using P/D with latencies between 6 and 8 ms and had an accuracy of 90.9% (95% CI: 58.7%-99.8%).

CONCLUSION

ANN models incorporating P/D can be a valuable tool for predicting FN outcomes after VS surgery. Refining the model to include P/D with latencies between 6 and 8 ms further improves the model's prediction. A user-friendly interface is provided to facilitate the implementation of this model.

Current status of artificial intelligence technologies in pituitary adenoma surgery: a scoping review

PURPOSE

Pituitary adenoma surgery is a complex procedure due to critical adjacent neurovascular structures, variations in size and extensions of the lesions, and potential hormonal imbalances. The integration of artificial intelligence (AI) and machine learning (ML) has demonstrated considerable potential in assisting neurosurgeons in decision-making, optimizing surgical outcomes, and providing real-time feedback. This scoping review comprehensively summarizes the current status of AI/ML technologies in pituitary adenoma surgery, highlighting their strengths and limitations.

METHODS

PubMed, Embase, Web of Science, and Scopus were searched following the PRISMA-ScR guidelines. Studies discussing the use of AI/ML in pituitary adenoma surgery were included. Eligible studies were grouped to analyze the different outcomes of interest of current AI/ML technologies.

RESULTS

Among the 2438 identified articles, 44 studies met the inclusion criteria, with a total of seventeen different algorithms utilized across all studies. Studies were divided into two groups based on their input type: clinicopathological and imaging input. The four main outcome variables evaluated in the studies included: outcome (remission, recurrence or progression, gross-total resection, vision improvement, and hormonal recovery), complications (CSF leak, readmission, hyponatremia, and hypopituitarism), cost, and adenoma-related factors (aggressiveness, consistency, and Ki-67 labeling) prediction. Three studies focusing on workflow analysis and real-time navigation were discussed separately.

CONCLUSION

AI/ML modeling holds promise for improving pituitary adenoma surgery by enhancing preoperative planning and optimizing surgical strategies. However, addressing challenges such as algorithm selection, performance evaluation, data heterogeneity, and ethics is essential to establish robust and reliable ML models that can revolutionize neurosurgical practice and benefit patients.

Intraoperative cerebrospinal fluid leakage and residual tumors in endoscopic transsphenoidal surgery for pituitary adenoma: risk analysis and nomogram development

BACKGROUND

Endoscopic transsphenoidal surgery is the primary method used to treat pituitary adenomas (PAs) at present; however, this technique is associated with certain risks, including cerebrospinal fluid leakage (CFL) and residual tumors (RTs). In this study, we aimed to identify specific risk factors for intraoperative CFL (ioCFL) and postoperative RT in patients with pituitary adenoma and construct a corresponding nomogram for risk assessment.

METHODS

We collected a range of information from 782 patients who underwent endoscopic transsphenoidal PA resection in the Department of Neurosurgery at Beijing Tiantan Hospital between 2019 and 2021. Patients were then randomly assigned to training and validation groups (in a 8:2 ratio) with R software. Univariate and multivariable logistic regression models were then used to screen variables related to ioCFL and RT. These variables were then used to construct a predictive nomogram. Finally, the accuracy of the nomogram was validated by receiver operating characteristic curve (ROC) analysis, calibration plots, and decision curve analysis (DCA).

RESULTS

Univariate and multivariable logistic regression models identified four risk factors for ioCFL (Hardy grade, tumor size, position, and consistency) and five risk factors for RT (operation time, tumor size, consistency, Knosp grade, and primary/recurrence type). The area under the ROC curve (AUC) for the ioCFL risk model was 0.666 and 0.697 for the training and validation groups, respectively. For RT, the AUCs for the two groups were 0.788 and 0.754, respectively. The calibration plots for the ioCFL and RT models showed high calibration quality and DCA analysis yielded excellent efficiency with regards to clinical decision making.

CONCLUSION

Tumor size, growth characteristics, and invasion location were identified as the main factors affecting intraoperative CFL and RT. With our novel nomogram, surgeons can identify high-risk patients according to preoperative and intraoperative tumor performance and reduce the probability of complications.

Current and Future Advances in Surgical Therapy for Pituitary Adenoma

The vital physiological role of the pituitary gland, alongside its proximity to critical neurovascular structures, means that pituitary adenomas can cause significant morbidity or mortality. While enormous advancements have been made in the surgical care of pituitary adenomas, numerous challenges remain, such as treatment failure and recurrence. To meet these clinical challenges, there has been an enormous expansion of novel medical technologies (eg, endoscopy, advanced imaging, artificial intelligence). These innovations have the potential to benefit each step of the patient's journey, and ultimately, drive improved outcomes. Earlier and more accurate diagnosis addresses this in part. Analysis of novel patient data sets, such as automated facial analysis or natural language processing of medical records holds potential in achieving an earlier diagnosis. After diagnosis, treatment decision-making and planning will benefit from radiomics and multimodal machine learning models. Surgical safety and effectiveness will be transformed by smart simulation methods for trainees. Next-generation imaging techniques and augmented reality will enhance surgical planning and intraoperative navigation. Similarly, surgical abilities will be augmented by the future operative armamentarium, including advanced optical devices, smart instruments, and surgical robotics. Intraoperative support to surgical team members will benefit from a data science approach, utilizing machine learning analysis of operative videos to improve patient safety and orientate team members to a common workflow. Postoperatively, neural networks leveraging multimodal datasets will allow early detection of individuals at risk of complications and assist in the prediction of treatment failure, thus supporting patient-specific discharge and monitoring protocols. While these advancements in pituitary surgery hold promise to enhance the quality of care, clinicians must be the gatekeepers of the translation of such technologies, ensuring systematic assessment of risk and benefit prior to clinical implementation. In doing so, the synergy between these innovations can be leveraged to drive improved outcomes for patients of the future.

Development and external validation of clinical prediction models for pituitary surgery

Introduction

Gross total resection (GTR), Biochemical Remission (BR) and restitution of a priorly disrupted hypothalamus pituitary axis (new improvement, IMP) are important factors in pituitary adenoma (PA) resection surgery. Prediction of these metrics using simple and preoperatively available data might help improve patient care and contribute to a more personalized medicine.

Research question

This study aims to develop machine learning models predicting GTR, BR, and IMP in PA resection surgery, using preoperatively available data.

Material and methods

With data from patients undergoing endoscopic transsphenoidal surgery for PAs machine learning models for prediction of GTR, BR and IMP were developed and externally validated. Development was carried out on a registry from Bologna, Italy while external validation was conducted using patient data from Zurich, Switzerland.

Results

The model development cohort consisted of 1203 patients. GTR was achieved in 207 (17.2%, 945 (78.6%) missing), BR in 173 (14.4%, 992 (82.5%) missing) and IMP in 208 (17.3%, 167 (13.9%) missing) cases. In the external validation cohort 206 patients were included and GTR was achieved in 121 (58.7%, 32 (15.5%) missing), BR in 46 (22.3%, 145 (70.4%) missing) and IMP in 42 (20.4%, 7 (3.4%) missing) cases. The AUC at external validation amounted to 0.72 (95% CI: 0.63-0.80) for GTR, 0.69 (0.52-0.83) for BR, as well as 0.82 (0.76-0.89) for IMP.

Discussion and conclusion

All models showed adequate generalizability, performing similarly in training and external validation, confirming the possible potentials of machine learning in helping to adapt surgical therapy to the individual patient.

Postsurgical outcomes of nonfunctioning pituitary adenomas: a patient-level meta-analysis

BACKGROUND

Surgical resection is the main treatment for symptomatic nonfunctioning pituitary adenomas (NFPA). We aimed to analyze the impact of surgical approach, completeness of resection, and postoperative radiotherapy on long-term progression-free survival (PFS) of NFPA, using individual patient data (IPD) meta-analysis.

METHODS

An electronic literature searched was conducted on PubMed, EMBASE, and Web of Science from database inception to 6 November 2022. Studies describing the natural history of surgically resected NFPA, with provision of Kaplan-Meier curves, were included. These were digitized to obtain IPD, which was pooled in one-stage and two-stage meta-analysis to determine hazard ratios (HRs) and 95%CIs of gross total resection (GTR) versus subtotal resection (STR), and postoperative radiotherapy versus none. An indirect analysis of single-arm data between endoscopic endonasal (EES) and microscopic transsphenoidal (MTS) surgical technique was also performed.

RESULTS

Altogether, eleven studies (3941 patients) were retrieved. PFS was significantly lower in STR than GTR (shared-frailty HR 0.32, 95%CI 0.27-0.39, p < 0.001). Postoperative radiotherapy significantly improved PFS compared to no radiotherapy (shared-frailty HR 0.20, 95%CI 0.15-0.26, p < 0.001), including in the subgroup of patients with STR (shared-frailty HR 0.12, 95%CI 0.08-0.18, p < 0.001). Similar PFS was observed between EES and MTS (indirect HR 1.09, 95%CI 0.92-1.30, p = 0.301).

CONCLUSIONS

This systematic review and patient-level meta-analysis provides a robust prognostication of surgically treated NFPA. We reinforce current guidelines stating that GTR should be the standard of surgical resection. Postoperative radiotherapy is of considerable benefit, especially for patients with STR. Surgical approach does not significantly affect long-term prognosis.

REGISTRATION

PROSPERO CRD42022374034.

Machine Learning and Artificial Intelligence in Surgical Research

Machine learning, a subtype of artificial intelligence, is an emerging field of surgical research dedicated to predictive modeling. From its inception, machine learning has been of interest in medical and surgical research. Built on traditional research metrics for optimal success, avenues of research include diagnostics, prognosis, operative timing, and surgical education, in a variety of surgical subspecialties. Machine learning represents an exciting and developing future in the world of surgical research that will not only allow for more personalized and comprehensive medical care.

Machine learning models predict delayed hyponatremia post-transsphenoidal surgery using clinically available features

PURPOSE

Delayed hyponatremia (DHN), a unique complication, is the leading cause of unexpected readmission after pituitary surgery. Therefore, this study aimed to develop tools for predicting postoperative DHN in patients undergoing endoscopic transsphenoidal surgery (eTSS) for pituitary neuroendocrine tumors (PitNETs).

METHODS

This was a single-center, retrospective study involving 193 patients with PitNETs who underwent eTSS. The objective variable was DHN, defined as serum sodium levels < 135 mmol/L at ≥ 1 time between post operative days 3 and 9. We trained four machine learning models to predict this objective variable using the clinical variables available preoperatively and on the first postoperative day. The clinical variables included patient characteristics, pituitary-related hormone levels, blood test results, radiological findings, and postoperative complications.

RESULTS

The random forest (RF) model demonstrated the highest (0.759 ± 0.039) area under the curve of the receiver operating characteristic curve (ROC-AUC), followed by the support vector machine (0.747 ± 0.034), the light gradient boosting machine (LGBM: 0.738 ± 0.026), and the logistic regression (0.710 ± 0.028). The highest accuracy (0.746 ± 0.029) was observed in the LGBM model. The best-performing RF model was based on 24 features, nine of which were clinically available preoperatively.

CONCLUSIONS

The proposed machine learning models with pre- and post-resection features predicted DHN after the resection of PitNETs.

Machine Learning Models to Forecast Outcomes of Pituitary Surgery: A Systematic Review in Quality of Reporting and Current Evidence

Background: The complex nature and heterogeneity involving pituitary surgery results have increased interest in machine learning (ML) applications for prediction of outcomes over the last decade. This study aims to systematically review the characteristics of ML models involving pituitary surgery outcome prediction and assess their reporting quality. Methods: We searched the PubMed, Scopus, and Web of Knowledge databases for publications on the use of ML to predict pituitary surgery outcomes. We used the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) to assess report quality. Our search strategy was based on the terms “artificial intelligence”, “machine learning”, and “pituitary”. Results: 20 studies were included in this review. The principal models reported in each article were post-surgical endocrine outcomes (n = 10), tumor management (n = 3), and intra- and postoperative complications (n = 7). Overall, the included studies adhered to a median of 65% (IQR = 60–72%) of TRIPOD criteria, ranging from 43% to 83%. The median reported AUC was 0.84 (IQR = 0.80–0.91). The most popular algorithms were support vector machine (n = 5) and random forest (n = 5). Only two studies reported external validation and adherence to any reporting guideline. Calibration methods were not reported in 15 studies. No model achieved the phase of actual clinical applicability. Conclusion: Applications of ML in the prediction of pituitary outcomes are still nascent, as evidenced by the lack of any model validated for clinical practice. Although studies have demonstrated promising results, greater transparency in model development and reporting is needed to enable their use in clinical practice. Further adherence to reporting guidelines can help increase AI’s real-world utility and improve clinical practice.

Artificial Intelligence for Neurosurgery : Current State and Future Directions

Artificial intelligence (AI) is a field of computer science that equips machines with human-like intelligence and enables them to learn, reason, and solve problems when presented with data in various formats. Neurosurgery is often at the forefront of innovative and disruptive technologies, which have similarly altered the course of acute and chronic diseases. In diagnostic imaging, such as X-rays, computed tomography, and magnetic resonance imaging, AI is used to analyze images. The use of robots in the field of neurosurgery is also increasing. In neurointensive care units, AI is used to analyze data and provide care to critically ill patients. Moreover, AI can be used to predict a patient's prognosis. Several AI applications have already been introduced in the field of neurosurgery, and many more are expected in the near future. Ultimately, it is our responsibility to keep pace with this evolution to provide meaningful outcomes and personalize each patient's care. Rather than blindly relying on AI in the future, neurosurgeons should gain a thorough understanding of it and use it to enhance their patient care.

Machine learning as a clinical decision support tool for patients with acromegaly

OBJECTIVE

To develop machine learning (ML) models that predict postoperative remission, remission at last visit, and resistance to somatostatin receptor ligands (SRL) in patients with acromegaly and to determine the clinical features associated with the prognosis.

METHODS

We studied outcomes using the area under the receiver operating characteristics (AUROC) values, which were reported as the performance metric. To determine the importance of each feature and easy interpretation, Shapley Additive explanations (SHAP) values, which help explain the outputs of ML models, are used.

RESULTS

One-hundred fifty-two patients with acromegaly were included in the final analysis. The mean AUROC values resulting from 100 independent replications were 0.728 for postoperative 3 months remission status classification, 0.879 for remission at last visit classification, and 0.753 for SRL resistance status classification. Extreme gradient boosting model demonstrated that preoperative growth hormone (GH) level, age at operation, and preoperative tumor size were the most important predictors for early remission; resistance to SRL and preoperative tumor size represented the most important predictors of remission at last visit, and postoperative 3-month insulin-like growth factor 1 (IGF1) and GH levels (random and nadir) together with the sparsely granulated somatotroph adenoma subtype served as the most important predictors of SRL resistance.

CONCLUSIONS

ML models may serve as valuable tools in the prediction of remission and SRL resistance.

Extent of Resection Research in Skull Base Neurosurgery: Previous Studies and Future Directions

Surgery is the first-line therapy for most benign and malignant skull base tumors. Extent of resection (EOR) is a metric commonly used for preoperative surgical planning and to predict risk of postoperative tumor recurrence. Therefore, understanding the evidence on EOR in skull base neurosurgery is essential to providing optimal care for each patient. Several studies from the skull base neurosurgery literature have presented investigations of various topics related to EOR, including 1) preoperative EOR scoring systems, 2) intraoperative EOR scoring systems, 3) EOR and tumor recurrence, and 4) EOR and functional outcomes. We propose that future investigations should focus on the following elements to improve EOR research in skull base neurosurgery: 1) multi-institutional collaboratives with treatment propensity matching; 2) expert consensus and mixed-methods study design; and 3) predictive analytics/machine learning. We believe that these methods offer several advantages that have been described in the literature and that they address limitations of previous studies. The aim of this review was to inform future study design and improve the overall quality of subsequent investigations on EOR in skull base neurosurgery.

Machine learning-based clinical outcome prediction in surgery for acromegaly

PURPOSE

Biochemical remission (BR), gross total resection (GTR), and intraoperative cerebrospinal fluid (CSF) leaks are important metrics in transsphenoidal surgery for acromegaly, and prediction of their likelihood using machine learning would be clinically advantageous. We aim to develop and externally validate clinical prediction models for outcomes after transsphenoidal surgery for acromegaly.

METHODS

Using data from two registries, we develop and externally validate machine learning models for GTR, BR, and CSF leaks after endoscopic transsphenoidal surgery in acromegalic patients. For the model development a registry from Bologna, Italy was used. External validation was then performed using data from Zurich, Switzerland. Gender, age, prior surgery, as well as Hardy and Knosp classification were used as input features. Discrimination and calibration metrics were assessed.

RESULTS

The derivation cohort consisted of 307 patients (43.3% male; mean [SD] age, 47.2 [12.7] years). GTR was achieved in 226 (73.6%) and BR in 245 (79.8%) patients. In the external validation cohort with 46 patients, 31 (75.6%) achieved GTR and 31 (77.5%) achieved BR. Area under the curve (AUC) at external validation was 0.75 (95% confidence interval: 0.59-0.88) for GTR, 0.63 (0.40-0.82) for BR, as well as 0.77 (0.62-0.91) for intraoperative CSF leaks. While prior surgery was the most important variable for prediction of GTR, age, and Hardy grading contributed most to the predictions of BR and CSF leaks, respectively.

CONCLUSIONS

Gross total resection, biochemical remission, and CSF leaks remain hard to predict, but machine learning offers potential in helping to tailor surgical therapy. We demonstrate the feasibility of developing and externally validating clinical prediction models for these outcomes after surgery for acromegaly and lay the groundwork for development of a multicenter model with more robust generalization.

Interpretable Machine Learning-Based Prediction of Intraoperative Cerebrospinal Fluid Leakage in Endoscopic Transsphenoidal Pituitary Surgery: A Pilot Study

Purpose  Transsphenoidal surgery (TSS) for pituitary adenomas can be complicated by the occurrence of intraoperative cerebrospinal fluid (CSF) leakage (IOL). IOL significantly affects the course of surgery predisposing to the development of postoperative CSF leakage, a major source of morbidity and mortality in the postoperative period. The authors trained and internally validated the Random Forest (RF) prediction model to preoperatively identify patients at high risk for IOL. A locally interpretable model-agnostic explanations (LIME) algorithm is employed to elucidate the main drivers behind each machine learning (ML) model prediction. Methods  The data of 210 patients who underwent TSS were collected; first, risk factors for IOL were identified via conventional statistical methods (multivariable logistic regression). Then, the authors trained, optimized, and audited a RF prediction model. Results  IOL reported in 45 patients (21.5%). The recursive feature selection algorithm identified the following variables as the most significant determinants of IOL: Knosp's grade, sellar Hardy's grade, suprasellar Hardy's grade, tumor diameter (on X, Y, and Z axes), intercarotid distance, and secreting status (nonfunctioning and growth hormone [GH] secreting). Leveraging the predictive values of these variables, the RF prediction model achieved an area under the curve (AUC) of 0.83 (95% confidence interval [CI]: 0.78; 0.86), significantly outperforming the multivariable logistic regression model (AUC = 0.63). Conclusion  A RF model that reliably identifies patients at risk for IOL was successfully trained and internally validated. ML-based prediction models can predict events that were previously judged nearly unpredictable; their deployment in clinical practice may result in improved patient care and reduced postoperative morbidity and healthcare costs.

The Application of Artificial Intelligence and Machine Learning in Pituitary Adenomas

Pituitary adenomas (PAs) are a group of tumors with complex and heterogeneous clinical manifestations. Early accurate diagnosis, individualized management, and precise prediction of the treatment response and prognosis of patients with PA are urgently needed. Artificial intelligence (AI) and machine learning (ML) have garnered increasing attention to quantitatively analyze complex medical data to improve individualized care for patients with PAs. Therefore, we critically examined the current use of AI and ML in the management of patients with PAs, and we propose improvements for future uses of AI and ML in patients with PAs. AI and ML can automatically extract many quantitative features based on massive medical data; moreover, related diagnosis and prediction models can be developed through quantitative analysis. Previous studies have suggested that AI and ML have wide applications in early accurate diagnosis; individualized treatment; predicting the response to treatments, including surgery, medications, and radiotherapy; and predicting the outcomes of patients with PAs. In addition, facial imaging-based AI and ML, pathological picture-based AI and ML, and surgical microscopic video-based AI and ML have also been reported to be useful in assisting the management of patients with PAs. In conclusion, the current use of AI and ML models has the potential to assist doctors and patients in making crucial surgical decisions by providing an accurate diagnosis, response to treatment, and prognosis of PAs. These AI and ML models can improve the quality and safety of medical services for patients with PAs and reduce the complication rates of neurosurgery. Further work is needed to obtain more reliable algorithms with high accuracy, sensitivity, and specificity for the management of PA patients.

Deep Learning for Outcome Prediction in Neurosurgery: A Systematic Review of Design, Reporting, and Reproducibility

Deep learning (DL) is a powerful machine learning technique that has increasingly been used to predict surgical outcomes. However, the large quantity of data required and lack of model interpretability represent substantial barriers to the validity and reproducibility of DL models. The objective of this study was to systematically review the characteristics of DL studies involving neurosurgical outcome prediction and to assess their bias and reporting quality. Literature search using the PubMed, Scopus, and Embase databases identified 1949 records of which 35 studies were included. Of these, 32 (91%) developed and validated a DL model while 3 (9%) validated a pre-existing model. The most commonly represented subspecialty areas were oncology (16 of 35, 46%), spine (8 of 35, 23%), and vascular (6 of 35, 17%). Risk of bias was low in 18 studies (51%), unclear in 5 (14%), and high in 12 (34%), most commonly because of data quality deficiencies. Adherence to transparent reporting of a multivariable prediction model for individual prognosis or diagnosis reporting standards was low, with a median of 12 transparent reporting of a multivariable prediction model for individual prognosis or diagnosis items (39%) per study not reported. Model transparency was severely limited because code was provided in only 3 studies (9%) and final models in 2 (6%). With the exception of public databases, no study data sets were readily available. No studies described DL models as ready for clinical use. The use of DL for neurosurgical outcome prediction remains nascent. Lack of appropriate data sets poses a major concern for bias. Although studies have demonstrated promising results, greater transparency in model development and reporting is needed to facilitate reproducibility and validation.

Machine Learning in Pituitary Surgery

Machine learning applications in neurosurgery are increasingly reported for diverse tasks such as faster and more accurate preoperative diagnosis, enhanced lesion characterization, as well as surgical outcome, complications and healthcare cost prediction. Even though the pertinent literature in pituitary surgery is less extensive with respect to other neurosurgical diseases, past research attempted to answer clinically relevant questions to better assist surgeons and clinicians. In the present chapter we review reported ML applications in pituitary surgery including differential diagnosis, preoperative lesion characterization (immunohistochemistry, cavernous sinus invasion, tumor consistency), surgical outcome and complication predictions (gross total resection, tumor recurrence, and endocrinological remission, cerebrospinal fluid leak, postoperative hyponatremia). Moreover, we briefly discuss from a practical standpoint the current barriers to clinical translation of machine learning research. On the topic of pituitary surgery, published reports can be considered mostly preliminary, requiring larger training populations and strong external validation. Thoughtful selection of clinically relevant outcomes of interest and transversal application of model development pipeline-together with accurate methodological planning and multicenter collaborations-have the potential to overcome current limitations and ultimately provide additional tools for more informed patient management.

Machine Learning and Intracranial Aneurysms: From Detection to Outcome Prediction

Machine learning (ML) is a rapidly rising research tool in biomedical sciences whose applications include segmentation, classification, disease detection, and outcome prediction. With respect to traditional statistical methods, ML algorithms have the potential to learn and improve their predictive performance when fed with large data sets without the need of being specifically programmed. In recent years, this technology has been increasingly applied for tackling clinical issues in intracranial aneurysm (IA) research. Several studies attempted to provide reliable models for enhanced aneurysm detection. Convolutional neural networks trained with variable degrees of human interaction on data from diverse imaging modalities showed high sensitivity in aneurysm detection tasks, also outperforming expert image analysis. Algorithms were also shown to differentiate ruptured from unruptured IAs, with however limited clinical relevance. For prediction of rupture and stability assessment, ML was preliminarily shown to achieve better performance compared to conventional statistical methods and existing risk scores. ML-based complication and functional outcome prediction in the event of SAH have been more extensively reported, in contrast with periprocedural outcome investigation in unruptured IA patients. ML has the potential to be a game changer in IA patient management. Currently clinical translation of experimental results is limited.

Simultaneous brain structure segmentation in magnetic resonance images using deep convolutional neural networks

In brain magnetic resonance imaging (MRI) examinations, rapidly acquired two-dimensional (2D) T1-weighted sagittal slices are typically used to confirm brainstem atrophy and the presence of signals in the posterior pituitary gland. Image segmentation is essential for the automatic evaluation of chronological changes in the brainstem and pituitary gland. Thus, the purpose of our study was to use deep learning to automatically segment internal organs (brainstem, corpus callosum, pituitary, cerebrum, and cerebellum) in midsagittal slices of 2D T1-weighted images. Deep learning for the automatic segmentation of seven regions in the images was accomplished using two different methods: patch-based segmentation and semantic segmentation. The networks used for patch-based segmentation were AlexNet, GoogLeNet, and ResNet50, whereas semantic segmentation was accomplished using SegNet, VGG16-weighted SegNet, and U-Net. The precision and Jaccard index were calculated, and the extraction accuracy of the six convolutional network (DCNN) systems was evaluated. The highest precision (0.974) was obtained with the VGG16-weighted SegNet, and the lowest precision (0.506) was obtained with ResNet50. Based on the data, calculation times, and Jaccard indices obtained in this study, segmentation on a 2D image may be considered a viable and effective approach. We found that the optimal automatic segmentation of organs (brainstem, corpus callosum, pituitary, cerebrum, and cerebellum) on brain sagittal T1-weighted images could be achieved using SegNet with VGG16.

Neural network modeling for prediction of recurrence, progression, and hormonal non-remission in patients following resection of functional pituitary adenomas

PURPOSE

Functional pituitary adenomas (FPAs) cause severe neuro-endocrinopathies including Cushing's disease (CD) and acromegaly. While many are effectively cured following FPA resection, some encounter disease recurrence/progression or hormonal non-remission requiring adjuvant treatment. Identification of risk factors for suboptimal postoperative outcomes may guide initiation of adjuvant multimodal therapies.

METHODS

Patients undergoing endonasal transsphenoidal resection for CD, acromegaly, and mammosomatotroph adenomas between 1992 and 2019 were identified. Good outcomes were defined as hormonal remission without imaging/biochemical evidence of disease recurrence/progression, while suboptimal outcomes were defined as hormonal non-remission or MRI evidence of recurrence/progression despite adjuvant treatment. Multivariate regression modeling and multilayered neural networks (NN) were implemented. The training sets randomly sampled 60% of all FPA patients, and validation/testing sets were 20% samples each.

RESULTS

348 patients with mean age of 41.7 years were identified. Eighty-one patients (23.3%) reported suboptimal outcomes. Variables predictive of suboptimal outcomes included: Requirement for additional surgery in patients who previously had surgery and continue to have functionally active tumor (p = 0.0069; OR = 1.51, 95%CI 1.12-2.04), Preoperative visual deficit not improved after surgery (p = 0.0033; OR = 1.12, 95%CI 1.04-1.20), Transient diabetes insipidus (p = 0.013; OR = 1.27, 95%CI 1.05-1.52), Higher MIB-1/Ki-67 labeling index (p = 0.038; OR = 1.08, 95%CI 1.01-1.15), and preoperative low cortisol axis (p = 0.040; OR = 2.72, 95%CI 1.06-7.01). The NN had overall accuracy of 87.1%, sensitivity of 89.5%, specificity of 76.9%, positive predictive value of 94.4%, and negative predictive value of 62.5%. NNs for all FPAs were more robust than for CD or acromegaly/mammosomatotroph alone.

CONCLUSION

We demonstrate capability of predicting suboptimal postoperative outcomes with high accuracy. NNs may aid in stratifying patients for risk of suboptimal outcomes, thereby guiding implementation of adjuvant treatment in high-risk patients.

Machine learning-based prediction of outcomes of the endoscopic endonasal approach in Cushing disease: is the future coming?

OBJECTIVE

Machine learning (ML) is an innovative method to analyze large and complex data sets. The aim of this study was to evaluate the use of ML to identify predictors of early postsurgical and long-term outcomes in patients treated for Cushing disease (CD).

METHODS

All consecutive patients in our center who underwent surgery for CD through the endoscopic endonasal approach were retrospectively reviewed. Study endpoints were gross-tumor removal (GTR), postsurgical remission, and long-term control of disease. Several demographic, radiological, and histological factors were assessed as potential predictors. For ML-based modeling, data were randomly divided into 2 sets with an 80% to 20% ratio for bootstrapped training and testing, respectively. Several algorithms were tested and tuned for the area under the curve (AUC).

RESULTS

The study included 151 patients. GTR was achieved in 137 patients (91%), and postsurgical hypersecretion remission was achieved in 133 patients (88%). At last follow-up, 116 patients (77%) were still in remission after surgery and in 21 patients (14%), CD was controlled with complementary treatment (overall, of 131 cases, 87% were under control at follow-up). At internal validation, the endpoints were predicted with AUCs of 0.81-1.00, accuracy of 81%-100%, and Brier scores of 0.035-0.151. Tumor size and invasiveness and histological confirmation of adrenocorticotropic hormone (ACTH)-secreting cells were the main predictors for the 3 endpoints of interest.

CONCLUSIONS

ML algorithms were used to train and internally validate robust models for all the endpoints, giving accurate outcome predictions in CD cases. This analytical method seems promising for potentially improving future patient care and counseling; however, careful clinical interpretation of the results remains necessary before any clinical adoption of ML. Moreover, further studies and increased sample sizes are definitely required before the widespread adoption of ML to the study of CD.

Machine learning in predicting early remission in patients after surgical treatment of acromegaly: a multicenter study

PURPOSE

Accurate prediction of postoperative remission is beneficial for effective patient-physician communication in acromegalic patients. This study aims to train and validate machine learning prediction models for early endocrine remission of acromegalic patients.

METHODS

The training cohort included 833 patients with growth hormone (GH) secreting pituitary adenoma from 2010 to 2018. We trained a partial model (only using pre-operative variables) and a full model (using all variables) to predict off-medication endocrine remission at six-month follow-up after surgery using multiple algorithms. The models were validated in 99 prospectively collected patients from a second campus and 52 patients from a third institution.

RESULTS

C-statistic and the accuracy of the best partial model was 0.803 (95% CI 0.757-0.849) and 72.5% (95% CI 67.6-77.5%), respectively. C-statistic and the accuracy of the best full model was 0.888 (95% CI 0.861-0.914) and 80.3% (95% CI 77.5-83.1%), respectively. The c-statistics (and accuracy) of using only Knosp grade, total resection, or postoperative day 1 GH level as the single predictor were lower than our partial model or full model (p < 0.001). C-statistics remained similar in the prospective cohort (partial model 0.798, and full model 0.903) and in the external cohort (partial model 0.771, and full model 0.871). A web-based application integrated with the trained models was published at  https://deepvep.shinyapps.io/Acropred/ .

CONCLUSION

We developed and validated interpretable and applicable machine learning models to predict early endocrine remission after surgical resection of a GH-secreting pituitary adenoma. Predication accuracy of the trained models were better than those using single variables.

Prediction of the Recurrence of Non-Functioning Pituitary Adenomas Using Preoperative Supra-Intra Sellar Volume and Tumor-Carotid Distance

BACKGROUND

Currently, it is difficult to estimate the possibility of recurrence of nonfunctioning pituitary adenomas (NFPAs). Markers such as Ki-67 or transcription factors rely on postoperative pathology, while few indices can be used for preoperative prediction. Therefore, we aimed to investigate the predictive effectiveness of supra-intrasellar volume and tumor-carotid distance based on measurements derived from preoperative magnetic resonance imaging (MRI) data.

METHOD

Ninety-eight cases of NFPAs were evaluated, along with their clinical characteristics and MRI features. Four radiologic indices were analyzed, including intrasellar tumor volume, suprasellar tumor volume, maximum horizontal tumor diameter, and intercarotid distance. The ratio of supra-intrasellar volume and ratio of tumor-carotid distance were measured using 3D Slicer software, and the sum of two ratios was defined as the V-D value. The correlation between recurrence and multiple factors was analyzed using univariate and multivariate logistic regression and Kaplan-Meier analysis, and ROC curves were used to estimate the prognostic performance of radiologic measurements in NFPAs.

RESULT

The supra-intrasellar volume ratio, tumor-carotid distance ratio and V-D value were significantly correlated with the recurrence of NFPAs. The predictive importance of the V-D value reached 84.5%, with a sensitivity of 83.7% and specificity of 67.3%. The cutoff limit of the V-D value was 1.53, and patients with V-D values higher than 1.53 tended to relapse much earlier.

CONCLUSION

The V-D value has predictive importance for the recurrence of NFPAs preoperatively. Patients with higher V-D values will undergo recurrence earlier and should be given greater consideration in terms of surgery and follow-up time.

Machine learning in neurosurgery: a global survey

Background

Recent technological advances have led to the development and implementation of machine learning (ML) in various disciplines, including neurosurgery. Our goal was to conduct a comprehensive survey of neurosurgeons to assess the acceptance of and attitudes toward ML in neurosurgical practice and to identify factors associated with its use.

Methods

The online survey consisted of nine or ten mandatory questions and was distributed in February and March 2019 through the European Association of Neurosurgical Societies (EANS) and the Congress of Neurosurgeons (CNS).

Results

Out of 7280 neurosurgeons who received the survey, we received 362 responses, with a response rate of 5%, mainly in Europe and North America. In total, 103 neurosurgeons (28.5%) reported using ML in their clinical practice, and 31.1% in research. Adoption rates of ML were relatively evenly distributed, with 25.6% for North America, 30.9% for Europe, 33.3% for Latin America and the Middle East, 44.4% for Asia and Pacific and 100% for Africa with only two responses. No predictors of clinical ML use were identified, although academic settings and subspecialties neuro-oncology, functional, trauma and epilepsy predicted use of ML in research. The most common applications were for predicting outcomes and complications, as well as interpretation of imaging.

Conclusions

This report provides a global overview of the neurosurgical applications of ML. A relevant proportion of the surveyed neurosurgeons reported clinical experience with ML algorithms. Future studies should aim to clarify the role and potential benefits of ML in neurosurgery and to reconcile these potential advantages with bioethical considerations.

Development of machine learning models to prognosticate chronic shunt-dependent hydrocephalus after aneurysmal subarachnoid hemorrhage

BACKGROUND

Shunt-dependent hydrocephalus significantly complicates subarachnoid hemorrhage (SAH), and reliable prognosis methods have been sought in recent years to reduce morbidity and costs associated with delayed treatment or neglected onset. Machine learning (ML) defines modern data analysis techniques allowing accurate subject-based risk stratifications. We aimed at developing and testing different ML models to predict shunt-dependent hydrocephalus after aneurysmal SAH.

METHODS

We consulted electronic records of patients with aneurysmal SAH treated at our institution between January 2013 and March 2019. We selected variables for the models according to the results of the previous works on this topic. We trained and tested four ML algorithms on three datasets: one containing binary variables, one considering variables associated with shunt-dependency after an explorative analysis, and one including all variables. For each model, we calculated AUROC, specificity, sensitivity, accuracy, PPV, and also, on the validation set, the NPV and the Matthews correlation coefficient (ϕ).

RESULTS

Three hundred eighty-six patients were included. Fifty patients (12.9%) developed shunt-dependency after a mean follow-up of 19.7 (± 12.6) months. Complete information was retrieved for 32 variables, used to train the models. The best models were selected based on the performances on the validation set and were achieved with a distributed random forest model considering 21 variables, with a ϕ = 0.59, AUC = 0.88; sensitivity and specificity of 0.73 (C.I.: 0.39-0.94) and 0.92 (C.I.: 0.84-0.97), respectively; PPV = 0.59 (0.38-0.77); and NPV = 0.96 (0.90-0.98). Accuracy was 0.90 (0.82-0.95).

CONCLUSIONS

Machine learning prognostic models allow accurate predictions with a large number of variables and a more subject-oriented prognosis. We identified a single best distributed random forest model, with an excellent prognostic capacity (ϕ = 0.58), which could be especially helpful in identifying low-risk patients for shunt-dependency.

Topographic brain tumor anatomy drives seizure risk and enables machine learning based prediction

OBJECTIVE

The aim of this study was to identify relevant risk factors for epileptic seizures upon initial diagnosis of a brain tumor and to develop and validate a machine learning based prediction to allow for a tailored risk-based antiepileptic therapy.

METHODS

Clinical, electrophysiological and high-resolution imaging data was obtained from a consecutive cohort of 1051 patients with newly diagnosed brain tumors. Factor-associated seizure risk difference allowed to determine the relevance of specific topographic, demographic and histopathologic variables available at the time of diagnosis for seizure risk. The data was divided in a 70/30 ratio into a training and test set. Different machine learning based predictive models were evaluated before a generalized additive model (GAM) was selected considering its traceability while maintaining high performance. Based on a clinical stratification of the risk factors, three different GAM were trained and internally validated.

RESULTS

A total of 923 patients had full data and were included. Specific topographic anatomical patterns that drive seizure risk could be identified. The involvement of allopallial, mesopallial or primary motor/somatosensory neopallial structures by brain tumors results in a significant and clinically relevant increase in seizure risk. While topographic input was most relevant for the GAM, the best prediction was achieved by a combination of topographic, demographic and histopathologic information (Validation: AUC: 0.79, Accuracy: 0.72, Sensitivity: 0.81, Specificity: 0.66).

CONCLUSIONS

This study identifies specific phylogenetic anatomical patterns as epileptic drivers. A GAM allowed the prediction of seizure risk using topographic, demographic and histopathologic data achieving fair performance while maintaining transparency.

Development of machine learning-based preoperative predictive analytics for unruptured intracranial aneurysm surgery: a pilot study

BACKGROUND

The decision to treat unruptured intracranial aneurysms (UIAs) or not is complex and requires balancing of risk factors and scores. Machine learning (ML) algorithms have previously been effective at generating highly accurate and comprehensive individualized preoperative predictive analytics in transsphenoidal pituitary and open tumor surgery. In this pilot study, we evaluate whether ML-based prediction of clinical endpoints is feasible for microsurgical management of UIAs.

METHODS

Based on data from a prospective registry, we developed and internally validated ML models to predict neurological outcome at discharge, as well as presence of new neurological deficits and any complication at discharge. Favorable neurological outcome was defined as modified Rankin scale (mRS) 0 to 2. According to the Clavien-Dindo grading (CDG), every adverse event during the post-operative course (surgery and not surgery related) is recorded as a complication. Input variables included age; gender; aneurysm complexity, diameter, location, number, and prior treatment; prior subarachnoid hemorrhage (SAH); presence of anticoagulation, antiplatelet therapy, and hypertension; microsurgical technique and approach; and various unruptured aneurysm scoring systems (PHASES, ELAPSS, UIATS).

RESULTS

We included 156 patients (26.3% male; mean [SD] age, 51.7 [11.0] years) with UIAs: 37 (24%) of them were treated for multiple aneurysm and 39 (25%) were treated for a complex aneurysm. Poor neurological outcome (mRS ≥ 3) was seen in 12 patients (7.7%) at discharge. New neurological deficits were seen in 10 (6.4%), and any kind of complication occurred in 20 (12.8%) patients. In the internal validation cohort, area under the curve (AUC) and accuracy values of 0.63-0.77 and 0.78-0.91 were observed, respectively.

CONCLUSIONS

Application of ML enables prediction of early clinical endpoints after microsurgery for UIAs. Our pilot study lays the groundwork for development of an externally validated multicenter clinical prediction model.

Big data, machine learning, and artificial intelligence: a field guide for neurosurgeons

Big data has transformed into a trend phrase in healthcare and neurosurgery, becoming a pervasive and inescapable phrase in everyday life. The upsurge in big data applications is a direct consequence of the drastic boom in information technology as well as the growing number of internet-connected devices called the Internet of Things in healthcare. Compared with business, marketing, and other sectors, healthcare applications are lagging due to a lack of technical knowledge among healthcare workers, technological limitations in acquiring and analyzing the data, and improper governance of healthcare big data. Despite these limitations, the medical literature is flooded with big data-related articles, and most of these are filled with abstruse terminologies such as machine learning, artificial intelligence, artificial neural network, and algorithm. Many of the recent articles are restricted to neurosurgical registries, creating a false impression that big data is synonymous with registries. Others advocate that the utilization of big data will be the panacea to all healthcare problems and research in the future. Without a proper understanding of these principles, it becomes easy to get lost without the ability to differentiate hype from reality. To that end, the authors give a brief narrative of big data analysis in neurosurgery and review its applications, limitations, and the challenges it presents for neurosurgeons and healthcare professionals naive to this field. Awareness of these basic concepts will allow neurosurgeons to understand the literature regarding big data, enabling them to make better decisions and deliver personalized care.

Feasibility of machine learning based predictive modelling of postoperative hyponatremia after pituitary surgery

PURPOSE

Hyponatremia after pituitary surgery is a frequent finding with potential severe complications and the most common cause for readmission. Several studies have found parameters associated with postoperative hyponatremia, but no reliable specific predictor was described yet. This pilot study evaluates the feasibility of machine learning (ML) algorithms to predict postoperative hyponatremia after resection of pituitary lesions.

METHODS

Retrospective screening of a prospective registry of patients who underwent transsphenoidal surgery for pituitary lesions. Hyponatremia within 30 days after surgery was the primary outcome. Several pre- and intraoperative clinical, procedural and laboratory features were selected to train different ML algorithms. Trained models were compared using common performance metrics. Final model was internally validated on the testing dataset.

RESULTS

From 207 patients included in the study, 44 (22%) showed a hyponatremia within 30 days postoperatively. Hyponatremic measurements peaked directly postoperatively (day 0-1) and around day 7. Bootstrapped performance metrics of different trained ML-models showed largest area under the receiver operating characteristic curve (AUROC) for the boosted generalized linear model (67.1%), followed by the Naïve Bayes classifier (64.6%). The discriminative capability of the final model was assessed by predicting on unseen dataset. Large AUROC (84.3%; 67.0-96.4), sensitivity (81.8%) and specificity (77.5%) with an overall accuracy of 78.4% (66.7-88.2) was reached.

CONCLUSION

Our trained ML-model was able to learn the complex risk factor interactions and showed a high discriminative capability on unseen patient data. In conclusion, ML-methods can predict postoperative hyponatremia and thus potentially reduce morbidity and improve patient safety.

[Artificial intelligence in neurosurgery]

BACKGROUND

Artificial intelligence (AI) in neurosurgery is becoming increasingly more important as the technology advances. This development can be measured by the increase of publications on AI in neurosurgery over the last years.

OBJECTIVE

This article provides insights into the current possibilities of using AI in neurosurgery.

MATERIAL AND METHODS

A review of the literature was carried out with a focus on exemplary work on the use of AI in neurosurgery.

RESULTS

The current neurosurgical publications on the use of AI show the diversity of the topic in this field. The main areas of application are diagnostics, outcome and treatment models.

CONCLUSION

The various areas of application of AI in the field of neurosurgery with a refined preoperative diagnostics and outcome predictions will significantly influence the future of neurosurgery. Neurosurgeons will continue to make the decisions on the indications for surgery but an optimized statement on diagnosis, treatment options and on the risk of surgery will be made by neurosurgeons with the help of AI in the future.

Machine learning-based preoperative predictive analytics for lumbar spinal stenosis

OBJECTIVEPatient-reported outcome measures (PROMs) following decompression surgery for lumbar spinal stenosis (LSS) demonstrate considerable heterogeneity. Individualized prediction tools can provide valuable insights for shared decision-making. The authors aim to evaluate the feasibility of predicting short- and long-term PROMs, reoperations, and perioperative parameters by machine learning (ML) methods.METHODSData were derived from a prospective registry. All patients had undergone single- or multilevel mini-open facet-sparing decompression for LSS. The prediction models were trained using various ML-based algorithms to predict the endpoints of interest. Models were selected by area under the receiver operating characteristic curve (AUC). The endpoints were dichotomized by minimum clinically important difference (MCID) and included 6-week and 12-month numeric rating scales for back pain (NRS-BP) and leg pain (NRS-LP) severity and the Oswestry Disability Index (ODI), as well as prolonged surgery (> 45 minutes), extended length of hospital stay (> 28 hours), and reoperations.RESULTSA total of 635 patients were included. The average age was 62 ± 10 years, and 333 patients (52%) were male. At 6 weeks, MCID was seen in 63%, 76%, and 61% of patients for ODI, NRS-LP, and NRS-BP, respectively. At internal validation, the models predicted MCID in these variables with accuracies of 69%, 76%, and 85%, and with AUCs of 0.75, 0.79, and 0.92. At 12 months, 66%, 63%, and 51% of patients reported MCID; the observed accuracies were 62%, 74%, and 66%, with AUCs of 0.68, 0.72, and 0.79. Reoperations occurred in 60 patients (9.5%), of which 27 (4.3%) occurred at the index level. Overall and index-level reoperations were predicted with 69% and 63% accuracy, respectively, and with AUCs of 0.66 and 0.61. In 15%, a length of surgery greater than 45 minutes was observed and predicted with 78% accuracy and AUC of 0.54. Only 15% of patients were admitted to the hospital for longer than 28 hours. The developed ML-based model enabled prediction of extended hospital stay with an accuracy of 77% and AUC of 0.58.CONCLUSIONSPreoperative prediction of a range of clinically relevant endpoints in decompression surgery for LSS using ML is feasible, and may enable enhanced informed patient consent and personalized shared decision-making. Access to individualized preoperative predictive analytics for outcome and treatment risks may represent a further step in the evolution of surgical care for patients with LSS.

Artificial intelligence and pituitary adenomas: A review

The aim of this review was to provide an overview of the main concepts in machine learning (ML) and to analyze the ML applications in the imaging of pituitary adenomas. After describing the clinical, pathological and imaging features of pituitary tumors, we defined the difference between ML and classical rule-based algorithms, we illustrated the fundamental ML techniques: supervised, unsupervised and reinforcement learning and explained the characteristic of deep learning, a ML approach employing networks inspired by brain’s structure. Pre-treatment assessment and neurosurgical outcome prediction were the potential ML applications using magnetic resonance imaging. Regarding pre-treatment assessment, ML methods were used to have information about tumor consistency, predict cavernous sinus invasion and high proliferative index, discriminate null cell adenomas, which respond to neo-adjuvant radiotherapy from other subtypes, predict somatostatin analogues response and visual pathway injury. Regarding neurosurgical outcome prediction, the following applications were discussed: Gross total resection prediction, evaluation of Cushing disease recurrence after transsphenoidal surgery and prediction of cerebrospinal fluid fistula’s formation after surgery. Although clinical applicability requires more replicability, generalizability and validation, results are promising, and ML software can be a potential power to facilitate better clinical decision making in pituitary tumor patients.

Pituitary Tumors in the Computational Era, Exploring Novel Approaches to Diagnosis, and Outcome Prediction with Machine Learning

BACKGROUND

Machine learning has emerged as a viable asset in the setting of pituitary surgery. In the past decade, the number of machine learning models developed to aid in the diagnosis of pituitary lesions and predict intraoperative and postoperative complications following transsphenoidal surgery has increased exponentially. As computational processing power continues to increase, big data sets continue to expand, and learning algorithms continue to surpass gold standard predictive tools, machine learning will serve to become an important component in improving patient care and outcomes.

METHODS

Relevant studies were identified based on a literature search in PubMed and MEDLINE databases, as well as from other sources including reference lists of published articles.

RESULTS

Radiomics and artificial neural networks comprise the majority of machine learning-based applications in pituitary surgery. Radiomics serves to quantify specific imaging features, which can then be used to noninvasively identify tumor characteristics and make definitive diagnoses, circumventing presurgical biopsy altogether. Neural networks can be adapted to predict intraoperative changes in visual evoked potentials or cerebral spinal fluid leak. In addition, these algorithms may be combined with others to predict tumor aggressiveness, gross total resection, recurrence and remission, and even total cost burden.

CONCLUSIONS

The field of machine learning is broad, with radiomics and artificial neural networks comprising 2 commonly used supervised learning methods in pituitary surgery. Given the large heterogeneity of pituitary and sellar lesions, the promise of machine learning lies in its ability to identify relationships and patterns that are otherwise hidden from standard statistical methods. While machine learning has great potential as a clinical adjunct during the surgical preplanning process and in predicting complications and outcomes, challenges moving forward include standardization and validation of these paradigms.

Diabetes insipidus and syndrome of inappropriate antidiuresis (SIADH) after pituitary surgery: incidence and risk factors

Electrolyte disorders are relatively frequent and potentially serious complications after pituitary surgery. Both DI (diabetes insipidus) and SIADH (syndrome of inappropriate antidiuresis) can complicate and prolong hospital and intensive care unit stay, and the latter may even be preventable. We aim to assess the incidence of both electrolyte disorders and their risk factors. From a prospective registry of patients who underwent endoscopic transnasal transsphenoidal surgery (TSS) for pituitary adenoma, patients with postoperative DI and SIADH were identified. Univariable and multivariable statistics were carried out to identify factors independently associated with the occurrence of either DI or SIADH. A total of 174 patients were included, of which 73 (42%) were female. Mean age was 54 years (range 20-88). During postoperative hospital stay, 13 (7.5%) patients presenting with DI and 11 (6.3%) with SIADH were identified. Patients who developed DI after surgery had significantly longer hospital stays (p = 0.022), as did those who developed SIADH (p = 0.002). Four (2.3%) patients were discharged with a diagnosis of persistent DI, and 2 (1.1%) with the diagnosis of SIADH. At the last follow-up, 5 (2.9%) patients presented with persistent DI, while none of the patients suffered from SIADH. Younger age (odds ratio (OR) 0.97, 95% confidence interval (CI) 0.94-1.01, p = 0.166) and pituitary apoplexy (OR 2.69, 95% CI 0.53-10.65, p = 0.184) were weakly associated with the occurrence of DI. We identified younger age (OR 0.96, 95% CI 0.92-0.99, p = 0.045) and lower preoperative serum sodium (OR 0.83, 95% CI 0.71-0.95, p = 0.008) as independent risk factors for SIADH. Although we found a weak association among age, pituitary apoplexy, and the occurrence of DI, no independent predictor was identified for DI. For postoperative SIADH however, lower age and preoperative serum sodium were identified as significant predictors. None of these findings were sufficiently supported by preexisting literature. Both electrolyte disorders are exquisitely hard to predict preoperatively, and further research into their early detection and prevention is warranted.

Development and external validation of a clinical prediction model for functional impairment after intracranial tumor surgery

OBJECTIVE

Decision-making for intracranial tumor surgery requires balancing the oncological benefit against the risk for resection-related impairment. Risk estimates are commonly based on subjective experience and generalized numbers from the literature, but even experienced surgeons overestimate functional outcome after surgery. Today, there is no reliable and objective way to preoperatively predict an individual patient's risk of experiencing any functional impairment.

METHODS

The authors developed a prediction model for functional impairment at 3 to 6 months after microsurgical resection, defined as a decrease in Karnofsky Performance Status of ≥ 10 points. Two prospective registries in Switzerland and Italy were used for development. External validation was performed in 7 cohorts from Sweden, Norway, Germany, Austria, and the Netherlands. Age, sex, prior surgery, tumor histology and maximum diameter, expected major brain vessel or cranial nerve manipulation, resection in eloquent areas and the posterior fossa, and surgical approach were recorded. Discrimination and calibration metrics were evaluated.

RESULTS

In the development (2437 patients, 48.2% male; mean age ± SD: 55 ± 15 years) and external validation (2427 patients, 42.4% male; mean age ± SD: 58 ± 13 years) cohorts, functional impairment rates were 21.5% and 28.5%, respectively. In the development cohort, area under the curve (AUC) values of 0.72 (95% CI 0.69-0.74) were observed. In the pooled external validation cohort, the AUC was 0.72 (95% CI 0.69-0.74), confirming generalizability. Calibration plots indicated fair calibration in both cohorts. The tool has been incorporated into a web-based application available at https://neurosurgery.shinyapps.io/impairment/.

CONCLUSIONS

Functional impairment after intracranial tumor surgery remains extraordinarily difficult to predict, although machine learning can help quantify risk. This externally validated prediction tool can serve as the basis for case-by-case discussions and risk-to-benefit estimation of surgical treatment in the individual patient.

Multicenter external validation of the Zurich Pituitary Score

PURPOSE

Recently, the Zurich Pituitary Score (ZPS) has been proposed as a new quantitative preoperative classification scheme for predicting gross total resection (GTR), extent of resection (EOR), and residual tumor volume (RV) in endoscopic pituitary surgery. We evaluated the external validity of the ZPS.

METHODS

In three reference centers for pituitary surgery, the ZPS was applied and correlated to GTR, EOR, and RV. Furthermore, its inter-rater agreement was assessed.

RESULTS

A total of 485 patients (53% male; age, 53.8 ± 15.7) were included. ZPS grades I, II, III, and IV were observed in 110 (23%), 270 (56%), 64 (13%), and 41 (8%) patients, respectively. GTR was achieved in 358 (74%) cases, with mean EOR of 87.6% ± 20.3% and RV of 1.42 ± 2.80 cm³. With increasing ZPS grade, strongly significant decreasing trends for GTR (I, 92%; II, 77%; III, 67%; IV, 15%; p < 0.001) and EOR (I, 93.8%; II, 89.9%; III, 88.1%; IV, 75.4%; p < 0.001) were found. Similarly, RV increased steadily ([cm³] I, 0.16; II, 0.61; III, 2.01; IV, 3.84; p < 0.001). We observed intraclass correlation coefficients of 0.837 (95% CI, 0.804-0.865) for intercarotid distance and 0.964 (95% CI, 0.956-0.970) for adenoma diameter, and Cohen's kappa of 0.972 (95% CI, 0.952-0.992) for the ZPS grades.

CONCLUSIONS

Application of the ZPS in three external cohorts was successful. The ZPS generalized well in terms of GTR, EOR, and RV; demonstrated excellent inter-rater agreement; and can safely and effectively be applied as a quantitative classification of adenomas with relevance to surgical outcome.

Machine learning applications in imaging analysis for patients with pituitary tumors: a review of the current literature and future directions

PURPOSE

To provide an overview of fundamental concepts in machine learning (ML), review the literature on ML applications in imaging analysis of pituitary tumors for the last 10 years, and highlight the future directions on potential applications of ML for pituitary tumor patients.

METHOD

We presented an overview of the fundamental concepts in ML, its various stages used in healthcare, and highlighted the key components typically present in an imaging-based tumor analysis pipeline. A search was conducted across four databases (PubMed, Ovid, Embase, and Google Scholar) to gather research articles from the past 10 years (2009-2019) involving imaging related to pituitary tumor and ML. We grouped the studies by imaging modalities and analyzed the ML tasks in terms of the data inputs, reference standards, methodologies, and limitations.

RESULTS

Of the 16 studies included in our analysis, 10 appeared in 2018-2019. Most of the studies utilized retrospective data and followed a semi-automatic ML pipeline. The studies included use of magnetic resonance imaging (MRI), facial photographs, surgical microscopic video, spectrometry, and spectroscopy imaging. The objectives of the studies covered 14 distinct applications and majority of the studies addressed a binary classification problem. Only five of the 11 MRI-based studies had an external validation or a holdout set to test the performance of a final trained model.

CONCLUSION

Through our concise evaluation and comparison of the studies using the concepts presented, we highlight future directions so that potential ML applications using different imaging modalities can be developed to benefit the clinical care of pituitary tumor patients.

Machine Learning Algorithm Identifies Patients at High Risk for Early Complications After Intracranial Tumor Surgery: Registry-Based Cohort Study

INTRODUCTION

Reliable preoperative identification of patients at high risk for early postoperative complications occurring within 24 h (EPC) of intracranial tumor surgery can improve patient safety and postoperative management. Statistical analysis using machine learning algorithms may generate models that predict EPC better than conventional statistical methods.

OBJECTIVE

To train such a model and to assess its predictive ability.

METHODS

This cohort study included patients from an ongoing prospective patient registry at a single tertiary care center with an intracranial tumor that underwent elective neurosurgery between June 2015 and May 2017. EPC were categorized based on the Clavien-Dindo classification score. Conventional statistical methods and different machine learning algorithms were used to predict EPC using preoperatively available patient, clinical, and surgery-related variables. The performance of each model was derived from examining classification performance metrics on an out-of-sample test dataset.

RESULTS

EPC occurred in 174 (26%) of 668 patients included in the analysis. Gradient boosting machine learning algorithms provided the model best predicting the probability of an EPC. The model scored an accuracy of 0.70 (confidence interval [CI] 0.59-0.79) with an area under the curve (AUC) of 0.73 and a sensitivity and specificity of 0.80 (CI 0.58-0.91) and 0.67 (CI 0.53-0.77) on the test set. The conventional statistical model showed inferior predictive power (test set: accuracy: 0.59 (CI 0.47-0.71); AUC: 0.64; sensitivity: 0.76 (CI 0.64-0.85); specificity: 0.53 (CI 0.41-0.64)).

CONCLUSION

Using gradient boosting machine learning algorithms, it was possible to create a prediction model superior to conventional statistical methods. While conventional statistical methods favor patients' characteristics, we found the pathology and surgery-related (histology, anatomical localization, surgical access) variables to be better predictors of EPC.

Development and assessment of machine learning algorithms for predicting remission after transsphenoidal surgery among patients with acromegaly

PURPOSE

Preoperative prediction of transsphenoidal surgical (TSS) response is important for determining individual treatment strategies for acromegaly. There is currently no accurate predictive model for TSS response for acromegaly. The current study sought to develop and validate machine learning (ML)-based models for preoperative prediction of TSS response for acromegaly.

METHODS

Six hundred sixty-eight patients with acromegaly were enrolled and divided into training (n = 534) and text datasets (n = 134) in this retrospective, data mining and ML study. The forward search algorithm was used to select features, and six ML algorithms were applied to construct TSS response prediction models. The performance of these ML models was validated using receiver operating characteristics analysis. Model calibration, discrimination ability, and clinical usefulness were also assessed.

RESULTS

Three hundred forty-nine (52.2%) patients achieved postoperative remission criteria and exhibited good TSS response. A univariate analysis was conducted and eight features, including age, hypertension, ophthalmic disorders, GH, IGF-1, nadir GH, maximal tumor diameter, and Knosp grade, were significantly associated with the TSS response in patients with acromegaly. After feature selection, the gradient boosting decision tree (GBDT), which was constructed with the eight significant features showed the best favorable discriminatory ability both the training (AUC = 0.8555) and validation (AUC = 0.8178) cohorts. The GBDT model showed good discrimination ability and calibration, with the highest levels of accuracy and specificity, and provided better estimates of TTS responses of patients with acromegaly compared with using only the Knosp grade. Decision curve analysis confirmed that the model was clinically useful.

CONCLUSIONS

ML-based models could aid neurosurgeons in the preoperative prediction of TTS response for patients with acromegaly, and could contribute to determining individual treatment strategies.

Prediction of high proliferative index in pituitary macroadenomas using MRI-based radiomics and machine learning

PURPOSE

Pituitary adenomas are among the most frequent intracranial tumors. They may exhibit clinically aggressive behavior, with recurrent disease and resistance to multimodal therapy. The ki-67 labeling index represents a proliferative marker which correlates with pituitary adenoma aggressiveness. Aim of our study was to assess the accuracy of machine learning analysis of texture-derived parameters from pituitary adenomas preoperative MRI for the prediction of ki-67 proliferation index class.

METHODS

A total of 89 patients who underwent an endoscopic endonasal procedure for pituitary adenoma removal with available ki-67 labeling index were included. From T2w MR images, 1128 quantitative imaging features were extracted. To select the most informative features, different supervised feature selection methods were employed. Subsequently, a k-nearest neighbors (k-NN) classifier was employed to predict macroadenoma high or low proliferation index. Algorithm validation was performed with a train-test approach.

RESULTS

Of the 12 subsets derived from feature selection, the best performing one was constituted by the 4 highest correlating parameters at Pearson's test. These all showed very good (ICC ≥ 0.85) inter-observer reproducibility. The overall accuracy of the k-NN in the test group was of 91.67% (33/36) of correctly classified patients.

CONCLUSIONS

Machine learning analysis of texture-derived parameters from preoperative T2 MRI has proven to be effective for the prediction of pituitary macroadenomas ki-67 proliferation index class. This might aid the surgical strategy making a more accurate preoperative lesion classification and allow for a more focused and cost-effective follow-up and long-term management.

A systematic review on machine learning in sellar region diseases: quality and reporting items

INTRODUCTION

Machine learning methods in sellar region diseases present a particular challenge because of the complexity and the necessity for reproducibility. This systematic review aims to compile the current literature on sellar region diseases that utilized machine learning methods and to propose a quality assessment tool and reporting checklist for future studies.

METHODS

PubMed and Web of Science were searched to identify relevant studies. The quality assessment included five categories: unmet needs, reproducibility, robustness, generalizability and clinical significance.

RESULTS

Seventeen studies were included with the diagnosis of general pituitary neoplasms, acromegaly, Cushing's disease, craniopharyngioma and growth hormone deficiency. 87.5% of the studies arbitrarily chose one or two machine learning models. One study chose ensemble models, and one study compared several models. 43.8% of studies did not provide the platform for model training, and roughly half did not offer parameters or hyperparameters. 62.5% of the studies provided a valid method to avoid over-fitting, but only five reported variations in the validation statistics. Only one study validated the algorithm in a different external database. Four studies reported how to interpret the predictors, and most studies (68.8%) suggested possible clinical applications of the developed algorithm. The workflow of a machine-learning study and the recommended reporting items were also provided based on the results.

CONCLUSIONS

Machine learning methods were used to predict diagnosis and posttreatment outcomes in sellar region diseases. Though most studies had substantial unmet need and proposed possible clinical application, replicability, robustness and generalizability were major limits in current studies.

Neural network-based identification of patients at high risk for intraoperative cerebrospinal fluid leaks in endoscopic pituitary surgery

OBJECTIVE

Although rates of postoperative morbidity and mortality have become relatively low in patients undergoing transnasal transsphenoidal surgery (TSS) for pituitary adenoma, cerebrospinal fluid (CSF) fistulas remain a major driver of postoperative morbidity. Persistent CSF fistulas harbor the potential for headache and meningitis. The aim of this study was to investigate whether neural network-based models can reliably identify patients at high risk for intraoperative CSF leakage.

METHODS

From a prospective registry, patients who underwent endoscopic TSS for pituitary adenoma were identified. Risk factors for intraoperative CSF leaks were identified using conventional statistical methods. Subsequently, the authors built a prediction model for intraoperative CSF leaks based on deep learning.

RESULTS

Intraoperative CSF leaks occurred in 45 (29%) of 154 patients. No risk factors for CSF leaks were identified using conventional statistical methods. The deep neural network-based prediction model classified 88% of patients in the test set correctly, with an area under the curve of 0.84. Sensitivity (83%) and specificity (89%) were high. The positive predictive value was 71%, negative predictive value was 94%, and F1 score was 0.77. High suprasellar Hardy grade, prior surgery, and older age contributed most to the predictions.

CONCLUSIONS

The authors trained and internally validated a robust deep neural network-based prediction model that identifies patients at high risk for intraoperative CSF. Machine learning algorithms may predict outcomes and adverse events that were previously nearly unpredictable, thus enabling safer and improved patient care and better patient counseling.