An introduction and overview of machine learning in neurosurgical care

Usefulness of Artificial Intelligence in Traumatic Brain Injury: A Bibliometric Analysis and Mini-review

BACKGROUND

Traumatic brain injury (TBI) has become a major source of disability worldwide, increasing the interest in algorithms that use artificial intelligence (AI) to optimize the interpretation of imaging studies, prognosis estimation, and critical care issues. In this study we present a bibliometric analysis and mini-review on the main uses that have been developed for TBI in AI.

METHODS

The results informing this review come from a Scopus database search as of April 15, 2023. The bibliometric analysis was carried out via the mapping bibliographic metrics method. Knowledge mapping was made in the VOSviewer software (V1.6.18), analyzing the "link strength" of networks based on co-occurrence of key words, countries co-authorship, and co-cited authors. In the mini-review section, we highlight the main findings and contributions of the studies.

RESULTS

A total of 495 scientific publications were identified from 2000 to 2023, with 9262 citations published since 2013. Among the 160 journals identified, The Journal of Neurotrauma, Frontiers in Neurology, and PLOS ONE were those with the greatest number of publications. The most frequently co-occurring key words were: "machine learning", "deep learning", "magnetic resonance imaging", and "intracranial pressure". The United States accounted for more collaborations than any other country, followed by United Kingdom and China. Four co-citation author clusters were found, and the top 20 papers were divided into reviews and original articles.

CONCLUSIONS

AI has become a relevant research field in TBI during the last 20 years, demonstrating great potential in imaging, but a more modest performance for prognostic estimation and neuromonitoring.

A Neurosurgical Readmissions Reduction Program in an Academic Hospital Leveraging Machine Learning, Workflow Analysis, and Simulation

BACKGROUND

 Predicting 30-day hospital readmissions is crucial for improving patient outcomes, optimizing resource allocation, and achieving financial savings. Existing studies reporting the development of machine learning (ML) models predictive of neurosurgical readmissions do not report factors related to clinical implementation.

OBJECTIVES

 Train individual predictive models with good performance (area under the receiver operating characteristic curve or AUROC > 0.8), identify potential interventions through semi-structured interviews, and demonstrate estimated clinical and financial impact of these models.

METHODS

 Electronic health records were utilized with five ML methodologies: gradient boosting, decision tree, random forest, ridge logistic regression, and linear support vector machine. Variables of interest were determined by domain experts and literature. The dataset was split divided 80% for training and validation and 20% for testing randomly. Clinical workflow analysis was conducted using semi-structured interviews to identify possible intervention points. Calibrated agent-based models (ABMs), based on a previous study with interventions, were applied to simulate reductions of the 30-day readmission rate and financial costs.

RESULTS

 The dataset covered 12,334 neurosurgical intensive care unit (NSICU) admissions (11,029 patients); 1,903 spine surgery admissions (1,641 patients), and 2,208 traumatic brain injury (TBI) admissions (2,185 patients), with readmission rate of 13.13, 13.93, and 23.73%, respectively. The random forest model for NSICU achieved best performance with an AUROC score of 0.89, capturing potential patients effectively. Six interventions were identified through 12 semi-structured interviews targeting preoperative, inpatient stay, discharge phases, and follow-up phases. Calibrated ABMs simulated median readmission reduction rates and resulted in 13.13 to 10.12% (NSICU), 13.90 to 10.98% (spine surgery), and 23.64 to 21.20% (TBI). Approximately $1,300,614.28 in saving resulted from potential interventions.

CONCLUSION

 This study reports the successful development and simulation of an ML-based approach for predicting and reducing 30-day hospital readmissions in neurosurgery. The intervention shows feasibility in improving patient outcomes and reducing financial losses.

Predicting non-muscle invasive bladder cancer outcomes using artificial intelligence: a systematic review using APPRAISE-AI

Accurate prediction of recurrence and progression in non-muscle invasive bladder cancer (NMIBC) is essential to inform management and eligibility for clinical trials. Despite substantial interest in developing artificial intelligence (AI) applications in NMIBC, their clinical readiness remains unclear. This systematic review aimed to critically appraise AI studies predicting NMIBC outcomes, and to identify common methodological and reporting pitfalls. MEDLINE, EMBASE, Web of Science, and Scopus were searched from inception to February 5th, 2024 for AI studies predicting NMIBC recurrence or progression. APPRAISE-AI was used to assess methodological and reporting quality of these studies. Performance between AI and non-AI approaches included within these studies were compared. A total of 15 studies (five on recurrence, four on progression, and six on both) were included. All studies were retrospective, with a median follow-up of 71 months (IQR 32-93) and median cohort size of 125 (IQR 93-309). Most studies were low quality, with only one classified as high quality. While AI models generally outperformed non-AI approaches with respect to accuracy, c-index, sensitivity, and specificity, this margin of benefit varied with study quality (median absolute performance difference was 10 for low, 22 for moderate, and 4 for high quality studies). Common pitfalls included dataset limitations, heterogeneous outcome definitions, methodological flaws, suboptimal model evaluation, and reproducibility issues. Recommendations to address these challenges are proposed. These findings emphasise the need for collaborative efforts between urological and AI communities paired with rigorous methodologies to develop higher quality models, enabling AI to reach its potential in enhancing NMIBC care.

Exploring the Landscape of Machine Learning Applications in Neurosurgery: A Bibliometric Analysis and Narrative Review of Trends and Future Directions

BACKGROUND

The field of neurosurgery has consistently represented an area of innovation and integration of technology since its inception. As such, machine learning (ML) has found its way into applications within neurosurgery relatively rapidly. Through this bibliometric review and cluster analysis, we seek to identify trends and emerging applications of ML within neurosurgery.

METHODS

A bibliometric analysis was carried out in the Web of Science database on publications from January 2000 to March 2023. The full data set of the 200 most cited publications including title, author information, journal, citation count, keywords, and abstracts for each publication was evaluated in CiteSpace. CiteSpace was used to elucidate publication characteristics, trends, and topic clusters via collaborate network analysis using the Kamada-Kawai algorithm.

RESULTS

The 25 most cited titles were included in our analysis. Harvard University and its affiliates represented the top institution, contributing nearly 25% of publications in the literature. WORLD NEUROSURGERY was the journal with the highest net citation count of 747 (29%). Collaborative network analysis generated 12 unique clusters, the largest of which was machine learning, followed by feature importance and deep brain stimulation.

CONCLUSION

This review highlights the most impactful articles pertaining to ML in the field of neurosurgery. ML has been applied into several sub-specialties within neurosurgery to optimize patient care, with special attention to outcome predictors, patient selection, and surgical decision making.

Computer-vision based analysis of the neurosurgical scene - A systematic review

Introduction

With increasing use of robotic surgical adjuncts, artificial intelligence and augmented reality in neurosurgery, the automated analysis of digital images and videos acquired over various procedures becomes a subject of increased interest. While several computer vision (CV) methods have been developed and implemented for analyzing surgical scenes, few studies have been dedicated to neurosurgery.

Research question

In this work, we present a systematic literature review focusing on CV methodologies specifically applied to the analysis of neurosurgical procedures based on intra-operative images and videos. Additionally, we provide recommendations for the future developments of CV models in neurosurgery.

Material and methods

We conducted a systematic literature search in multiple databases until January 17, 2023, including Web of Science, PubMed, IEEE Xplore, Embase, and SpringerLink.

Results

We identified 17 studies employing CV algorithms on neurosurgical videos/images. The most common applications of CV were tool and neuroanatomical structure detection or characterization, and to a lesser extent, surgical workflow analysis. Convolutional neural networks (CNN) were the most frequently utilized architecture for CV models (65%), demonstrating superior performances in tool detection and segmentation. In particular, mask recurrent-CNN manifested most robust performance outcomes across different modalities.

Discussion and conclusion

Our systematic review demonstrates that CV models have been reported that can effectively detect and differentiate tools, surgical phases, neuroanatomical structures, as well as critical events in complex neurosurgical scenes with accuracies above 95%. Automated tool recognition contributes to objective characterization and assessment of surgical performance, with potential applications in neurosurgical training and intra-operative safety management.

Neurosurgical Management of Cerebrospinal Tumors in the Era of Artificial Intelligence : A Scoping Review

Central nervous system tumors are identified as tumors of the brain and spinal cord. The associated morbidity and mortality of cerebrospinal tumors are disproportionately high compared to other malignancies. While minimally invasive techniques have initiated a revolution in neurosurgery, artificial intelligence (AI) is expediting it. Our study aims to analyze AI's role in the neurosurgical management of cerebrospinal tumors. We conducted a scoping review using the Arksey and O'Malley framework. Upon screening, data extraction and analysis were focused on exploring all potential implications of AI, classification of these implications in the management of cerebrospinal tumors. AI has enhanced the precision of diagnosis of these tumors, enables surgeons to excise the tumor margins completely, thereby reducing the risk of recurrence, and helps to make a more accurate prediction of the patient's prognosis than the conventional methods. AI also offers real-time training to neurosurgeons using virtual and 3D simulation, thereby increasing their confidence and skills during procedures. In addition, robotics is integrated into neurosurgery and identified to increase patient outcomes by making surgery less invasive. AI, including machine learning, is rigorously considered for its applications in the neurosurgical management of cerebrospinal tumors. This field requires further research focused on areas clinically essential in improving the outcome that is also economically feasible for clinical use. The authors suggest that data analysts and neurosurgeons collaborate to explore the full potential of AI.

Revolutionizing Neurosurgery with GPT-4: A Leap Forward or Ethical Conundrum?

Neurosurgery, a highly specialized and sophisticated branch of medicine, is devoted to the surgical intervention of maladies impacting both the central and peripheral nervous systems. The intricate nature and meticulous precision demanded by neurosurgery has piqued the interest of artificial intelligence experts. In our comprehensive analysis, we encapsulate the prospective applications of the revolutionary GPT-4 technology within the sphere of neurosurgery, encompassing areas such as preoperative evaluation and preparation, tailored surgical simulations, postoperative care and rehabilitation, enriched patient communication, fostering collaboration and knowledge dissemination, as well as training and education. Furthermore, we plunge into the complex and intellectually stimulating conundrums that arise when integrating the cutting-edge GPT-4 technology into neurosurgery, taking into account the moral considerations and substantial hurdles intrinsic to its adoption. Our stance is that GPT-4 will not supplant neurosurgeons; on the contrary, it possesses the potential to serve as an invaluable instrument in augmenting the precision and effectiveness of neurosurgical procedures, ultimately enhancing patient outcomes and propelling the field forward.

Application of artificial intelligence in gametes and embryos selection

Gamete and embryo quality are critical to the success rate of Assisted Reproductive Technology (ART) cycles, but there remains a lack of methods to accurately measure the quality of sperm, oocytes and embryos. The ability of Artificial Intelligence (AI) technology to analyze large amounts of data, especially video and images, is particularly useful in gamete and embryo assessment and selection. The well-trained model has fast calculation speed and high accuracy, which can help embryologists to perform more objective gamete and embryo selection. Various artificial intelligence models have been developed for gamete and embryo assessment, some of which exhibit good performance. In this review, we summarize the latest applications of AI technology in semen analysis, as well as selection for sperm, oocyte and embryo, and discuss the existing problems and development directions of artificial intelligence in this field.

Daily Pain Prediction Using Smartphone Speech Recordings of Patients With Spine Disease

BACKGROUND

Pain evaluation remains largely subjective in neurosurgical practice, but machine learning provides the potential for objective pain assessment tools.

OBJECTIVE

To predict daily pain levels using speech recordings from personal smartphones of a cohort of patients with diagnosed neurological spine disease.

METHODS

Patients with spine disease were enrolled through a general neurosurgical clinic with approval from the institutional ethics committee. At-home pain surveys and speech recordings were administered at regular intervals through the Beiwe smartphone application. Praat audio features were extracted from the speech recordings to be used as input to a K-nearest neighbors (KNN) machine learning model. The pain scores were transformed from a 0 to 10 scale to low and high pain for better discriminative capacity.

RESULTS

A total of 60 patients were enrolled, and 384 observations were used to train and test the prediction model. Using the KNN prediction model, an accuracy of 71% with a positive predictive value of 0.71 was achieved in classifying pain intensity into high and low. The model showed 0.71 precision for high pain and 0.70 precision for low pain. Recall of high pain was 0.74, and recall of low pain was 0.67. The overall F1 score was 0.73.

CONCLUSION

Our study uses a KNN to model the relationship between speech features and pain levels collected from personal smartphones of patients with spine disease. The proposed model is a stepping stone for the development of objective pain assessment in neurosurgery clinical practice.

A new computational approach, based on images trajectories, to identify the subjacent heterogeneity of sperm to the effects of ketanserin

The identification of kinematic subpopulations is of paramount importance to understanding the biological nature of the sperm heterogeneity. Nowadays, the data of motility parameters obtained by a computer-assisted sperm analysis (CASA) system has been used as input to distinct algorithms to identify kinematic subpopulations. In contrast, the images of the trajectories were depicted only as examples of the patterns of motility in each subpopulation. Here, python code was written to reconstruct the images of trajectories, from their coordinates, then the images of trajectories were used as input to a machine learning clustering algorithm of classification, and the subpopulations were described statistically by the motility parameters. Finally, the images of trajectories in each subpopulation were displayed in a way we called Pollock plots. Semen samples of boar sperm were treated with distinct concentrations of ketanserin (an antagonist of the 5-HT2 receptor of serotonin) and untreated samples were used as a control. The motility of sperm in each sample was analyzed at 0 and 30 min of incubation. Six subpopulations were found. The subpopulation 2 presented the highest values of velocities at 0 or 30 min. After 30 min of incubation, the ketanserin increased the values of the curvilinear velocity at high concentrations, whereas the linearity and the straight velocity decreased. Our computational model permits better identification of the kinematic subpopulations than the traditional approach and provides insights onto the heterogeneity of the response to ketanserin; thus, it could significantly impact the research on the relationship between sperm heterogeneity-fertility.

Machine learning algorithms for predicting outcomes of traumatic brain injury: A systematic review and meta-analysis

BACKGROUND

Traumatic brain injury (TBI) is a leading cause of death and disability worldwide. The use of machine learning (ML) has emerged as a key advancement in TBI management. This study aimed to identify ML models with demonstrated effectiveness in predicting TBI outcomes.

METHODS

We conducted a systematic review in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis statement. In total, 15 articles were identified using the search strategy. Patient demographics, clinical status, ML outcome variables, and predictive characteristics were extracted. A small meta-analysis of mortality prediction was performed, and a meta-analysis of diagnostic accuracy was conducted for ML algorithms used across multiple studies.

RESULTS

ML algorithms including support vector machine (SVM), artificial neural networks (ANN), random forest, and Naïve Bayes were compared to logistic regression (LR). Thirteen studies found significant improvement in prognostic capability using ML versus LR. The accuracy of the above algorithms was consistently over 80% when predicting mortality and unfavorable outcome measured by Glasgow Outcome Scale. Receiver operating characteristic curves analyzing the sensitivity of ANN, SVM, decision tree, and LR demonstrated consistent findings across studies. Lower admission Glasgow Coma Scale (GCS), older age, elevated serum acid, and abnormal glucose were associated with increased adverse outcomes and had the most significant impact on ML algorithms.

CONCLUSION

ML algorithms were stronger than traditional regression models in predicting adverse outcomes. Admission GCS, age, and serum metabolites all have strong predictive power when used with ML and should be considered important components of TBI risk stratification.

Artificial Intelligence in Neurosurgery: A State-of-the-Art Review from Past to Future

In recent years, there has been a significant surge in discussions surrounding artificial intelligence (AI), along with a corresponding increase in its practical applications in various facets of everyday life, including the medical industry. Notably, even in the highly specialized realm of neurosurgery, AI has been utilized for differential diagnosis, pre-operative evaluation, and improving surgical precision. Many of these applications have begun to mitigate risks of intraoperative and postoperative complications and post-operative care. This article aims to present an overview of the principal published papers on the significant themes of tumor, spine, epilepsy, and vascular issues, wherein AI has been applied to assess its potential applications within neurosurgery. The method involved identifying high-cited seminal papers using PubMed and Google Scholar, conducting a comprehensive review of various study types, and summarizing machine learning applications to enhance understanding among clinicians for future utilization. Recent studies demonstrate that machine learning (ML) holds significant potential in neuro-oncological care, spine surgery, epilepsy management, and other neurosurgical applications. ML techniques have proven effective in tumor identification, surgical outcomes prediction, seizure outcome prediction, aneurysm prediction, and more, highlighting its broad impact and potential in improving patient management and outcomes in neurosurgery. This review will encompass the current state of research, as well as predictions for the future of AI within neurosurgery.

Tool-tissue force segmentation and pattern recognition for evaluating neurosurgical performance

Surgical data quantification and comprehension expose subtle patterns in tasks and performance. Enabling surgical devices with artificial intelligence provides surgeons with personalized and objective performance evaluation: a virtual surgical assist. Here we present machine learning models developed for analyzing surgical finesse using tool-tissue interaction force data in surgical dissection obtained from a sensorized bipolar forceps. Data modeling was performed using 50 neurosurgery procedures that involved elective surgical treatment for various intracranial pathologies. The data collection was conducted by 13 surgeons of varying experience levels using sensorized bipolar forceps, SmartForceps System. The machine learning algorithm constituted design and implementation for three primary purposes, i.e., force profile segmentation for obtaining active periods of tool utilization using T-U-Net, surgical skill classification into Expert and Novice, and surgical task recognition into two primary categories of Coagulation versus non-Coagulation using FTFIT deep learning architectures. The final report to surgeon was a dashboard containing recognized segments of force application categorized into skill and task classes along with performance metrics charts compared to expert level surgeons. Operating room data recording of > 161 h containing approximately 3.6 K periods of tool operation was utilized. The modeling resulted in Weighted F1-score = 0.95 and AUC = 0.99 for force profile segmentation using T-U-Net, Weighted F1-score = 0.71 and AUC = 0.81 for surgical skill classification, and Weighted F1-score = 0.82 and AUC = 0.89 for surgical task recognition using a subset of hand-crafted features augmented to FTFIT neural network. This study delivers a novel machine learning module in a cloud, enabling an end-to-end platform for intraoperative surgical performance monitoring and evaluation. Accessed through a secure application for professional connectivity, a paradigm for data-driven learning is established.

Artificial intelligence in neuroradiology: a scoping review of some ethical challenges

Artificial intelligence (AI) has great potential to increase accuracy and efficiency in many aspects of neuroradiology. It provides substantial opportunities for insights into brain pathophysiology, developing models to determine treatment decisions, and improving current prognostication as well as diagnostic algorithms. Concurrently, the autonomous use of AI models introduces ethical challenges regarding the scope of informed consent, risks associated with data privacy and protection, potential database biases, as well as responsibility and liability that might potentially arise. In this manuscript, we will first provide a brief overview of AI methods used in neuroradiology and segue into key methodological and ethical challenges. Specifically, we discuss the ethical principles affected by AI approaches to human neuroscience and provisions that might be imposed in this domain to ensure that the benefits of AI frameworks remain in alignment with ethics in research and healthcare in the future.

Computational Intelligence in Cancer Diagnostics: A Contemporary Review of Smart Phone Apps, Current Problems, and Future Research Potentials

Cancer is a dangerous and sometimes life-threatening disease that can have several negative consequences for the body, is a leading cause of mortality, and is becoming increasingly difficult to detect. Each form of cancer has its own set of traits, symptoms, and therapies, and early identification and management are important for a positive prognosis. Doctors utilize a variety of approaches to detect cancer, depending on the kind and location of the tumor. Imaging tests such as X-rays, Computed Tomography scans, Magnetic Resonance Imaging scans, and Positron Emission Tomography (PET) scans, which may provide precise pictures of the body's interior structures to spot any abnormalities, are some of the tools that doctors use to diagnose cancer. This article evaluates computational-intelligence approaches and provides a means to impact future work by focusing on the relevance of machine learning and deep learning models such as K Nearest Neighbour (KNN), Support Vector Machine (SVM), Naïve Bayes, Decision Tree, Deep Neural Network, Deep Boltzmann machine, and so on. It evaluates information from 114 studies using Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR). This article explores the advantages and disadvantages of each model and provides an outline of how they are used in cancer diagnosis. In conclusion, artificial intelligence shows significant potential to enhance cancer imaging and diagnosis, despite the fact that there are a number of clinical issues that need to be addressed.

Ensemble machine learning algorithm for predicting acute kidney injury in patients admitted to the neurointensive care unit following brain surgery

Acute kidney injury (AKI) is a common postoperative complication among patients in the neurological intensive care unit (NICU), often resulting in poor prognosis and high mortality. In this retrospective cohort study, we established a model for predicting AKI following brain surgery based on an ensemble machine learning algorithm using data from 582 postoperative patients admitted to the NICU at the Dongyang People's Hospital from March 1, 2017, to January 31, 2020. Demographic, clinical, and intraoperative data were collected. Four machine learning algorithms (C5.0, support vector machine, Bayes, and XGBoost) were used to develop the ensemble algorithm. The AKI incidence in critically ill patients after brain surgery was 20.8%. Intraoperative blood pressure; postoperative oxygenation index; oxygen saturation; and creatinine, albumin, urea, and calcium levels were associated with the postoperative AKI occurrence. The area under the curve value for the ensembled model was 0.85. The accuracy, precision, specificity, recall, and balanced accuracy values were 0.81, 0.86, 0.44, 0.91, and 0.68, respectively, indicating good predictive ability. Ultimately, the models using perioperative variables exhibited good discriminatory ability for early prediction of postoperative AKI risk in patients admitted to the NICU. Thus, the ensemble machine learning algorithm may be a valuable tool for forecasting AKI.

Mortality prediction in severe traumatic brain injury using traditional and machine learning algorithms

Prognostic prediction of traumatic brain injury (TBI) in patients is crucial in clinical decision and health care policy making. This study aimed to develop and validate prediction models for in-hospital mortality after severe traumatic brain injury (sTBI). We developed and validated logistic regression (LR), LASSO regression, and machine learning (ML) algorithms including support vector machines (SVM) and XGBoost models. Fifty-four candidate predictors were included. Model performance was expressed in terms of discrimination (C-statistic) and calibration (intercept and slope). For model development, 2804 patients with sTBI in the Collaborative European NeuroTrauma Effectiveness Research in TBI (CENTER-TBI) China Registry study were included. External validation was performed in 1113 patients with sTBI in the CENTER-TBI European Registry study. XGBoost achieved high discrimination in mortality prediction, and it outperformed logistic and LASSO regression. The XGBoost model established in this study also outperformed prediction models currently available, including the International Mission for Prognosis and Analysis of Clinical Trials (IMPACT) core and International Mission for Prognosis and Analysis of Clinical Trials (CRASH) basic models. When including 54 variables, XGBoost and SVM reached C-statistics of 0.87 (95% confidence interval [CI]: 0.81-0.92) and 0.85 (95% CI: 0.79-0.90) at internal validation, and 0.88 (95% CI: 0.87-0.88) and 0.86 (95% CI: 0.85-0.87) at external validation, respectively. A simplified version of XGBoost and SVM using 26 variables selected by recursive feature elimination (RFE) reached C-statistics of 0.87 (95% CI: 0.82-0.92) and 0.86 (95% CI: 0.80-0.91) at internal validation, and 0.87 (95% CI: 0.87-0.88) and 0.87 (95% CI: 0.86-0.87) at external validation, respectively. However, when the number of variables included decreased, the difference between ML and LR diminished. All the prediction models can be accessed via a web-based calculator. Glasgow Coma Scale (GCS) score, age, pupillary light reflex, Injury Severity Score (ISS) for brain region, and the presence of acute subdural hematoma were the five strongest predictors for mortality prediction. The study showed that ML techniques such as XGBoost may capture information hidden in demographic and clinical predictors of patients with sTBI and yield more precise predictions compared with LR approaches.

Ethics and governance of trustworthy medical artificial intelligence

BACKGROUND

The growing application of artificial intelligence (AI) in healthcare has brought technological breakthroughs to traditional diagnosis and treatment, but it is accompanied by many risks and challenges. These adverse effects are also seen as ethical issues and affect trustworthiness in medical AI and need to be managed through identification, prognosis and monitoring.

METHODS

We adopted a multidisciplinary approach and summarized five subjects that influence the trustworthiness of medical AI: data quality, algorithmic bias, opacity, safety and security, and responsibility attribution, and discussed these factors from the perspectives of technology, law, and healthcare stakeholders and institutions. The ethical framework of ethical values-ethical principles-ethical norms is used to propose corresponding ethical governance countermeasures for trustworthy medical AI from the ethical, legal, and regulatory aspects.

RESULTS

Medical data are primarily unstructured, lacking uniform and standardized annotation, and data quality will directly affect the quality of medical AI algorithm models. Algorithmic bias can affect AI clinical predictions and exacerbate health disparities. The opacity of algorithms affects patients' and doctors' trust in medical AI, and algorithmic errors or security vulnerabilities can pose significant risks and harm to patients. The involvement of medical AI in clinical practices may threaten doctors 'and patients' autonomy and dignity. When accidents occur with medical AI, the responsibility attribution is not clear. All these factors affect people's trust in medical AI.

CONCLUSIONS

In order to make medical AI trustworthy, at the ethical level, the ethical value orientation of promoting human health should first and foremost be considered as the top-level design. At the legal level, current medical AI does not have moral status and humans remain the duty bearers. At the regulatory level, strengthening data quality management, improving algorithm transparency and traceability to reduce algorithm bias, and regulating and reviewing the whole process of the AI industry to control risks are proposed. It is also necessary to encourage multiple parties to discuss and assess AI risks and social impacts, and to strengthen international cooperation and communication.

Identification of chronic mild traumatic brain injury using resting state functional MRI and machine learning techniques

Mild traumatic brain injury (mTBI) is a major public health concern that can result in a broad spectrum of short-term and long-term symptoms. Recently, machine learning (ML) algorithms have been used in neuroscience research for diagnostics and prognostic assessment of brain disorders. The present study aimed to develop an automatic classifier to distinguish patients suffering from chronic mTBI from healthy controls (HCs) utilizing multilevel metrics of resting-state functional magnetic resonance imaging (rs-fMRI). Sixty mTBI patients and forty HCs were enrolled and allocated to training and testing datasets with a ratio of 80:20. Several rs-fMRI metrics including fractional amplitude of low-frequency fluctuation (fALFF), regional homogeneity (ReHo), degree centrality (DC), voxel-mirrored homotopic connectivity (VMHC), functional connectivity strength (FCS), and seed-based FC were generated from two main analytical categories: local measures and network measures. Statistical two-sample t-test was employed comparing between mTBI and HCs groups. Then, for each rs-fMRI metric the features were selected extracting the mean values from the clusters showing significant differences. Finally, the support vector machine (SVM) models based on separate and multilevel metrics were built and the performance of the classifiers were assessed using five-fold cross-validation and via the area under the receiver operating characteristic curve (AUC). Feature importance was estimated using Shapley additive explanation (SHAP) values. Among local measures, the range of AUC was 86.67-100% and the optimal SVM model was obtained based on combined multilevel rs-fMRI metrics and DC as a separate model with AUC of 100%. Among network measures, the range of AUC was 80.42-93.33% and the optimal SVM model was obtained based on the combined multilevel seed-based FC metrics. The SHAP analysis revealed the DC value in the left postcentral and seed-based FC value between the motor ventral network and right superior temporal as the most important local and network features with the greatest contribution to the classification models. Our findings demonstrated that different rs-fMRI metrics can provide complementary information for classifying patients suffering from chronic mTBI. Moreover, we showed that ML approach is a promising tool for detecting patients with mTBI and might serve as potential imaging biomarker to identify patients at individual level.

Clinical trial registration

[clinicaltrials.gov], identifier [NCT03241732].

Statistical and machine learning approaches to predict the necessity for computed tomography in children with mild traumatic brain injury

BACKGROUND

Minor head trauma in children is a common reason for emergency department visits, but the risk of traumatic brain injury (TBI) in those children is very low. Therefore, physicians should consider the indication for computed tomography (CT) to avoid unnecessary radiation exposure to children. The purpose of this study was to statistically assess the differences between control and mild TBI (mTBI). In addition, we also investigate the feasibility of machine learning (ML) to predict the necessity of CT scans in children with mTBI.

METHODS AND FINDINGS

The study enrolled 1100 children under the age of 2 years to assess pre-verbal children. Other inclusion and exclusion criteria were per the PECARN study. Data such as demographics, injury details, medical history, and neurological assessment were used for statistical evaluation and creation of the ML algorithm. The number of children with clinically important TBI (ciTBI), mTBI on CT, and controls was 28, 30, and 1042, respectively. Statistical significance between the control group and clinically significant TBI requiring hospitalization (csTBI: ciTBI+mTBI on CT) was demonstrated for all nonparametric predictors except severity of the injury mechanism. The comparison between the three groups also showed significance for all predictors (p<0.05). This study showed that supervised ML for predicting the need for CT scan can be generated with 95% accuracy. It also revealed the significance of each predictor in the decision tree, especially the "days of life."

CONCLUSIONS

These results confirm the role and importance of each of the predictors mentioned in the PECARN study and show that ML could discriminate between children with csTBI and the control group.

Utilizing artificial intelligence and electroencephalography to assess expertise on a simulated neurosurgical task

Virtual reality surgical simulators have facilitated surgical education by providing a safe training environment. Electroencephalography (EEG) has been employed to assess neuroelectric activity during surgical performance. Machine learning (ML) has been applied to analyze EEG data split into frequency bands. Although EEG is widely used in fields requiring expert performance, it has yet been used to classify surgical expertise. Thus, the goals of this study were to (a) develop an ML model to accurately differentiate skilled and less-skilled performance using EEG data recorded during a simulated surgery, (b) explore the relative importance of each EEG bandwidth to expertise, and (c) analyze differences in EEG band powers between skilled and less-skilled individuals. We hypothesized that EEG recordings during a virtual reality surgery task would accurately predict the expertise level of the participant. Twenty-one participants performed three simulated brain tumor resection procedures on the NeuroVR™ platform (CAE Healthcare, Montreal, Canada) while EEG data was recorded. Participants were divided into 2 groups. The skilled group was composed of five neurosurgeons and five senior neurosurgical residents (PGY4-6), and the less-skilled group was composed of six junior residents (PGY1-3) and five medical students. A total of 13 metrics from EEG frequency bands and ratios (e.g., alpha, theta/beta ratio) were generated. Seven ML model types were trained using EEG activity to differentiate between skilled and less-skilled groups. The artificial neural network achieved the highest testing accuracy of 100% (AUROC = 1.0). Model interpretation via Shapley analysis identified low alpha (8-10 Hz) as the most important metric for classifying expertise. Skilled surgeons displayed higher (p = 0.044) low-alpha than the less-skilled group. Furthermore, skilled surgeons displayed significantly lower TBR (p = 0.048) and significantly higher beta (13-30 Hz, p = 0.049), beta 1 (15-18 Hz, p = 0.014), and beta 2 (19-22 Hz, p = 0.015), thus establishing these metrics as important markers of expertise. ACGME CORE COMPETENCIES: Practice-Based Learning and Improvement.

Machine learning-based algorithm as an innovative approach for the differentiation between diabetes insipidus and primary polydipsia in clinical practice

OBJECTIVE

Differentiation between central diabetes insipidus (cDI) and primary polydipsia (PP) remains challenging in clinical practice. Although the hypertonic saline infusion test led to high diagnostic accuracy, it is a laborious test requiring close monitoring of plasma sodium levels. As such, we leverage machine learning (ML) to facilitate differential diagnosis of cDI.

DESIGN

We analyzed data of 59 patients with cDI and 81 patients with PP from a prospective multicenter study evaluating the hypertonic saline test as new test approach to diagnose cDI. Our primary outcome was the diagnostic accuracy of the ML-based algorithm in differentiating cDI from PP patients.

METHODS

The data set used included 56 clinical, biochemical, and radiological covariates. We identified a set of five covariates which were crucial for differentiating cDI from PP patients utilizing standard ML methods. We developed ML-based algorithms on the data and validated them with an unseen test data set.

RESULTS

Urine osmolality, plasma sodium and glucose, known transsphenoidal surgery, or anterior pituitary deficiencies were selected as input parameters for the basic ML-based algorithm. Testing it on an unseen test data set resulted in a high area under the curve (AUC) score of 0.87. A further improvement of the ML-based algorithm was reached with the addition of MRI characteristics and the results of the hypertonic saline infusion test (AUC: 0.93 and 0.98, respectively).

CONCLUSION

The developed ML-based algorithm facilitated differentiation between cDI and PP patients with high accuracy even if only clinical information and laboratory data were available, thereby possibly avoiding cumbersome clinical tests in the future.

Machine learning for outcome prediction of neurosurgical aneurysm treatment: Current methods and future directions

INTRODUCTION

Machine learning algorithms have received increased attention in neurosurgical literature for improved accuracy over traditional predictive methods. In this review, the authors sought to assess current applications of machine learning for outcome prediction of neurosurgical treatment of intracranial aneurysms and identify areas for future research.

METHODS

A PRISMA-compliant systematic review of the PubMed, MEDLINE, and EMBASE databases was conducted for all studies utilizing machine learning for outcome prediction of intracranial aneurysm treatment. Patient characteristics, machine learning methods, outcomes of interest, and accuracy metrics were recorded from included studies.

RESULTS

16 studies were ultimately included in qualitative synthesis. Studies primarily analyzed angiographic outcomes, functional outcomes, or complication prediction using clinical, radiological, or composite variables. The majority of included studies utilized supervised learning algorithms for analysis of dichotomized outcomes.

CONCLUSIONS

Commonly included variables were demographics, presentation variables (including ruptured or unruptured status), and treatment used. Areas for future research include increased generalizability across institutions and for smaller datasets, as well as development of front-end tools for clinical applicability of published algorithms.

Deep neural networks allow expert-level brain meningioma segmentation and present potential for improvement of clinical practice

Accurate brain meningioma segmentation and volumetric assessment are critical for serial patient follow-up, surgical planning and monitoring response to treatment. Current gold standard of manual labeling is a time-consuming process, subject to inter-user variability. Fully-automated algorithms for meningioma segmentation have the potential to bring volumetric analysis into clinical and research workflows by increasing accuracy and efficiency, reducing inter-user variability and saving time. Previous research has focused solely on segmentation tasks without assessment of impact and usability of deep learning solutions in clinical practice. Herein, we demonstrate a three-dimensional convolutional neural network (3D-CNN) that performs expert-level, automated meningioma segmentation and volume estimation on MRI scans. A 3D-CNN was initially trained by segmenting entire brain volumes using a dataset of 10,099 healthy brain MRIs. Using transfer learning, the network was then specifically trained on meningioma segmentation using 806 expert-labeled MRIs. The final model achieved a median performance of 88.2% reaching the spectrum of current inter-expert variability (82.6–91.6%). We demonstrate in a simulated clinical scenario that a deep learning approach to meningioma segmentation is feasible, highly accurate and has the potential to improve current clinical practice.

Clinical application of the “sellar barrier’s concept” for predicting intraoperative CSF leak in endoscopic endonasal surgery for pituitary adenomas with a machine learning analysis

Background

Recently, it was defined that the sellar barrier entity could be identified as a predictor of cerebrospinal fluid (CSF) intraoperative leakage. The aim of this study is to validate the application of the sellar barrier concept for predicting intraoperative CSF leak in endoscopic endonasal surgery for pituitary adenomas with a machine learning approach.

Methods

We conducted a prospective cohort study, from June 2019 to September 2020: data from 155 patients with pituitary subdiaphragmatic adenoma operated through endoscopic approach at the Division of Neurosurgery, Università degli Studi di Napoli “Federico II,” were included. Preoperative magnetic resonance images (MRI) and intraoperative findings were analyzed. After processing patient data, the experiment was conducted as a novelty detection problem, splitting outliers (i.e., patients with intraoperative fistula, n = 11/155) and inliers into separate datasets, the latter further separated into training (n = 115/144) and inlier test (n = 29/144) datasets. The machine learning analysis was performed using different novelty detection algorithms [isolation forest, local outlier factor, one-class support vector machine (oSVM)], whose performance was assessed separately and as an ensemble on the inlier and outlier test sets.

Results

According to the type of sellar barrier, patients were classified into two groups, i.e., strong and weak barrier; a third category of mixed barrier was defined when a case was neither weak nor strong. Significant differences between the three datasets were found for Knosp classification score (p = 0.0015), MRI barrier: strong (p = 1.405 × 10⁻⁶), MRI barrier: weak (p = 4.487 × 10⁻⁸), intraoperative barrier: strong (p = 2.788 × 10⁻⁷), and intraoperative barrier: weak (p = 2.191 × 10⁻¹⁰). We recorded 11 cases of intraoperative leakage that occurred in the majority of patients presenting a weak sellar barrier (p = 4.487 × 10⁻⁸) at preoperative MRI. Accuracy, sensitivity, and specificity for outlier detection were 0.70, 0.64, and 0.72 for IF; 0.85, 0.45, and 1.00 for LOF; 0.83, 0.64, and 0.90 for oSVM; and 0.83, 0.55, and 0.93 for the ensemble, respectively.

Conclusions

There is a true correlation between the type of sellar barrier at MRI and its in vivo features as observed during endoscopic endonasal surgery. The novelty detection models highlighted differences between patients who developed an intraoperative CSF leak and those who did not.

Survival Prediction After Neurosurgical Resection of Brain Metastases: A Machine Learning Approach

BACKGROUND

Current prognostic models for brain metastases (BMs) have been constructed and validated almost entirely with data from patients receiving up-front radiotherapy, leaving uncertainty about surgical patients.

OBJECTIVE

To build and validate a model predicting 6-month survival after BM resection using different machine learning algorithms.

METHODS

An institutional database of 1062 patients who underwent resection for BM was split into an 80:20 training and testing set. Seven different machine learning algorithms were trained and assessed for performance; an established prognostic model for patients with BM undergoing radiotherapy, the diagnosis-specific graded prognostic assessment, was also evaluated. Model performance was assessed using area under the curve (AUC) and calibration.

RESULTS

The logistic regression showed the best performance with an AUC of 0.71 in the hold-out test set, a calibration slope of 0.76, and a calibration intercept of 0.03. The diagnosis-specific graded prognostic assessment had an AUC of 0.66. Patients were stratified into regular-risk, high-risk and very high-risk groups for death at 6 months; these strata strongly predicted both 6-month and longitudinal overall survival ( P < .0005). The model was implemented into a web application that can be accessed through http://brainmets.morethanml.com .

CONCLUSION

We developed and internally validated a prediction model that accurately predicts 6-month survival after neurosurgical resection for BM and allows for meaningful risk stratification. Future efforts should focus on external validation of our model.

Fair Transmission of Individual Signals and Formation of Mainstream Information: Evidence from Herd Behaviours in Emergencies

Risk society is full of emergencies, accompanied by uncertainties and losses. Under emergencies, controlling herd behaviour is challenging due to more interactions and changes among individuals. This research establishes Bayes conditional probability models to explain the fair transmission of individual signals and individual decision-making after receiving others' signals. The simulation shows the following conclusions: first, each individual has a fair chance to influence the mainstream information; second, the order in which individuals make decisions during an emergency affects the difficulties and likelihood of making a rational decision; third, the high authority of information can become mainstream and guide individual behaviour; and fourth, two individual characteristics, including risk appetite and personal experience, are important in the fair transmission of individual signals and formation of mainstream information. According to the findings, this research proposes two strategies, including interfering with information and controlling existing key opinion leaders to control the mainstream information within a group in emergencies. These two strategies are proved to be useful in detecting and preventing approaches to alleviate individual herd behaviour, which should be monitored and controlled in machine learning models for individual behaviour simulation and prediction. Compared to previous research that focuses on media and public opinion in emergencies, this research focuses on a specific type of information (i.e., individual decision-making and actions) on the individual level and its effects on herd behaviours within the group. This research complements the explanation of the micro-mechanism of how individuals receive information and make decisions and actions.

Could Machine Learning Better Predict Postoperative C5 Palsy of Cervical Ossification of the Posterior Longitudinal Ligament?

STUDY DESIGN

This was a retrospective cohort study.

OBJECTIVE

The objective of this study was to investigate whether machine learning (ML) can perform better than a conventional logistic regression in predicting postoperative C5 palsy of cervical ossification of the posterior longitudinal ligament (OPLL) patients.

SUMMARY OF BACKGROUND DATA

C5 palsy is one of the most common postoperative complications after surgical treatment of OPLL, with an incidence rate of 1.4%-18.4%. ML has recently been used to predict the outcomes of neurosurgery. To our knowledge there has not been a study to predict postoperative C5 palsy of cervical OPLL patient with ML.

METHODS

Four sampling methods were used for data balancing. Six ML algorithms and conventional logistic regression were used for model development. A total of 35 ML prediction model and 5 conventional logistic prediction models were generated. The performances of each model were compared with the area under the curve (AUC). Patients who underwent surgery for cervical OPLL at our institute from January 1998 to January 2012 were reviewed. Twenty-five variables of each patient were used to make a prediction model.

RESULTS

In total, 901 patients were included [651 male and 250 female, median age: 55 (49-63), mean±SD: 55.9±9.802]. Twenty-six (2.8%) patients developed postoperative C5 palsy. Age (P=0.043), surgical method (P=0.0112), involvement of OPLL at C1-3 (P=0.0359), and postoperative shoulder pain (P≤0.001) were significantly associated with C5 palsy. Among all ML models, a model using an adaptive reinforcement learning algorithm and downsampling showed the largest AUC (0.88; 95% confidence interval: 0.79-0.96), better than that of logistic regression (0.69; 95% confidence interval: 0.43-0.94).

CONCLUSIONS

The ML algorithm seems to be superior to logistic regression for predicting postoperative C5 palsy of OPLL patient after surgery with respect to AUC. Age, surgical method, and involvement of OPLL at C1-C3 were significantly associated with C5 palsy. This study demonstrates that shoulder pain immediately after surgery is closely associated with postoperative C5 palsy of OPLL patient.

GENESIS: Gene-Specific Machine Learning Models for Variants of Uncertain Significance Found in Catecholaminergic Polymorphic Ventricular Tachycardia and Long QT Syndrome-Associated Genes

BACKGROUND

Cardiac channelopathies such as catecholaminergic polymorphic tachycardia and long QT syndrome predispose patients to fatal arrhythmias and sudden cardiac death. As genetic testing has become common in clinical practice, variants of uncertain significance (VUS) in genes associated with catecholaminergic polymorphic ventricular tachycardia and long QT syndrome are frequently found. The objective of this study was to predict pathogenicity of catecholaminergic polymorphic ventricular tachycardia-associated RYR2 VUS and long QT syndrome-associated VUS in KCNQ1, KCNH2, and SCN5A by developing gene-specific machine learning models and assessing them using cross-validation, cellular electrophysiological data, and clinical correlation.

METHODS

The GENe-specific EnSemble grId Search framework was developed to identify high-performing machine learning models for RYR2, KCNQ1, KCNH2, and SCN5A using variant- and protein-specific inputs. Final models were applied to datasets of VUS identified from ClinVar and exome sequencing. Whole cell patch clamp and clinical correlation of selected VUS was performed.

RESULTS

The GENe-specific EnSemble grId Search models outperformed alternative methods, with area under the receiver operating characteristics up to 0.87, average precisions up to 0.83, and calibration slopes as close to 1.0 (perfect) as 1.04. Blinded voltage-clamp analysis of HEK293T cells expressing 2 predicted pathogenic variants in KCNQ1 each revealed an ≈80% reduction of peak Kv7.1 current compared with WT. Normal Kv7.1 function was observed in KCNQ1-V241I HEK cells as predicted. Though predicted benign, loss of Kv7.1 function was observed for KCNQ1-V106D HEK cells. Clinical correlation of 9/10 variants supported model predictions.

CONCLUSIONS

Gene-specific machine learning models may have a role in post-genetic testing diagnostic analyses by providing high performance prediction of variant pathogenicity.

What works where and how for uptake and impact of artificial intelligence in pathology: A review of theories for a realist evaluation (Preprint)

BACKGROUND

There is increasing interest in the use of artificial intelligence (AI) in pathology to increase accuracy and efficiency. To date, studies of clinicians' perceptions of AI have found only moderate acceptability, suggesting the need for further research regarding how to integrate it into clinical practice.

OBJECTIVE

The aim of the study was to determine contextual factors that may support or constrain the uptake of AI in pathology.

METHODS

To go beyond a simple listing of barriers and facilitators, we drew on the approach of realist evaluation and undertook a review of the literature to elicit stakeholders' theories of how, for whom, and in what circumstances AI can provide benefit in pathology. Searches were designed by an information specialist and peer-reviewed by a second information specialist. Searches were run on the arXiv.org repository, MEDLINE, and the Health Management Information Consortium, with additional searches undertaken on a range of websites to identify gray literature. In line with a realist approach, we also made use of relevant theory. Included documents were indexed in NVivo 12, using codes to capture different contexts, mechanisms, and outcomes that could affect the introduction of AI in pathology. Coded data were used to produce narrative summaries of each of the identified contexts, mechanisms, and outcomes, which were then translated into theories in the form of context-mechanism-outcome configurations.

RESULTS

A total of 101 relevant documents were identified. Our analysis indicates that the benefits that can be achieved will vary according to the size and nature of the pathology department's workload and the extent to which pathologists work collaboratively; the major perceived benefit for specialist centers is in reducing workload. For uptake of AI, pathologists' trust is essential. Existing theories suggest that if pathologists are able to "make sense" of AI, engage in the adoption process, receive support in adapting their work processes, and can identify potential benefits to its introduction, it is more likely to be accepted.

CONCLUSIONS

For uptake of AI in pathology, for all but the most simple quantitative tasks, measures will be required that either increase confidence in the system or provide users with an understanding of the performance of the system. For specialist centers, efforts should focus on reducing workload rather than increasing accuracy. Designers also need to give careful thought to usability and how AI is integrated into pathologists' workflow.

Nerve optic segmentation in CT images using a deep learning model and a texture descriptor

The increased intracranial pressure (ICP) can be described as an increase in pressure around the brain and can lead to serious health problems. The assessment of ultrasound images is commonly conducted by skilled experts which is a time-consuming approach, but advanced computer-aided diagnosis (CAD) systems can assist the physician to decrease the time of ICP diagnosis. The accurate detection of the nerve optic regions, with drawing a precise slope line behind the eyeball and calculating the diameter of nerve optic, are the main aims of this research. First, the Fuzzy C-mean (FCM) clustering is employed for segmenting the input CT screening images into the different parts. Second, a histogram equalization approach is used for region-based image quality enhancement. Then, the Local Directional Number method (LDN) is used for representing some key information in a new image. Finally, a cascade Convolutional Neural Network (CNN) is employed for nerve optic segmentation by two distinct input images. Comprehensive experiments on the CT screening dataset [The Cancer Imaging Archive (TCIA)] consisting of 1600 images show the competitive results of inaccurate extraction of the brain features. Also, the indexes such as Dice, Specificity, and Precision for the proposed approach are reported 87.7%, 91.3%, and 90.1%, respectively. The final classification results show that the proposed approach effectively and accurately detects the nerve optic and its diameter in comparison with the other methods. Therefore, this method can be used for early diagnose of ICP and preventing the occurrence of serious health problems in patients.

The intersection of prognostication and code status in patients with severe brain injury

Accurately estimating the prognosis of brain injury patients can be difficult, especially early in their course. Prognostication is important because it largely determines the care level we provide, from aggressive treatment for patients we predict could have a good outcome to withdrawal of treatment for those we expect will have a poor outcome. Accurate prognostication is required for ethical decision-making. However, several studies have shown that prognostication is frequently inaccurate and variable. Overly optimistic prognostication can lead to false hope and futile care. Overly pessimistic prognostication can lead to therapeutic nihilism. Overlapping is the powerful effect that cognitive biases, in particular code status, can play in shaping our perceptions and the care level we provide. The presence of Do Not Resuscitate orders has been shown to be associated with increased mortality. Based on a comprehensive search of peer-reviewed journals using a wide range of key terms, including prognostication, critical illness, brain injury, cognitive bias, and code status, the following is a review of prognostic accuracy and the effect of code status on outcome. Because withdrawal of treatment is the most common cause of death in the ICU, a clearer understanding of this intersection of prognostication and code status is needed.

Prediction of Shunt Responsiveness in Suspected Patients With Normal Pressure Hydrocephalus Using the Lumbar Infusion Test: A Machine Learning Approach

BACKGROUND

Machine learning (ML) approaches can significantly improve the classical Rout-based evaluation of the lumbar infusion test (LIT) and the clinical management of the normal pressure hydrocephalus.

OBJECTIVE

To develop a ML model that accurately identifies patients as candidates for permanent cerebral spinal fluid shunt implantation using only intracranial pressure and electrocardiogram signals recorded throughout LIT.

METHODS

This was a single-center cohort study of prospectively collected data of 96 patients who underwent LIT and 5-day external lumbar cerebral spinal fluid drainage (external lumbar drainage) as a reference diagnostic method. A set of selected 48 intracranial pressure/electrocardiogram complex signal waveform features describing nonlinear behavior, wavelet transform spectral signatures, or recurrent map patterns were calculated for each patient. After applying a leave-one-out cross-validation training-testing split of the data set, we trained and evaluated the performance of various state-of-the-art ML algorithms.

RESULTS

The highest performing ML algorithm was the eXtreme Gradient Boosting. This model showed a good calibration and discrimination on the testing data, with an area under the receiver operating characteristic curve of 0.891 (accuracy: 82.3%, sensitivity: 86.1%, and specificity: 73.9%) obtained for 8 selected features. Our ML model clearly outperforms the classical Rout-based manual classification commonly used in clinical practice with an accuracy of 62.5%.

CONCLUSION

This study successfully used the ML approach to predict the outcome of a 5-day external lumbar drainage and hence which patients are likely to benefit from permanent shunt implantation. Our automated ML model thus enhances the diagnostic utility of LIT in management.

[Artificial intelligence guided predicting the length of hospital-stay in a neurosurgical hospital based on the text data of electronic medical records]

BACKGROUND

Rational use of internal resources of hospitals including bed fund turnover is important objective in high-tech medicine. Machine learning technologies can improve neurosurgical care and contribute to patient-oriented approach.

OBJECTIVE

To evaluate the quality of AI-guided predicting the length of hospital-stay in a neurosurgical hospital based on the text data of electronic medical records in comparison with expectations of patients and physicians.

MATERIAL AND METHODS

AI-guided prediction was based on analysis of unstructured text records of the electronic medical history (preoperative examination and surgical protocol). Predictive models were learned on the data of the Burdenko Neurosurgery Center accumulated for the period 2000-2017 (90.688 cases). Model testing was performed on 111 completed neurosurgical cases in a prospective study. We compared the accuracy of prediction models compared to expectations of patients and physicians regarding hospital-stay.

RESULTS

The median absolute error of machine prediction in the final test was 2.00 days. This value was comparable with the doctor's prediction error.

CONCLUSION

This study demonstrated the possibility of using unstructured textual data to predict the length of hospital-stay in a neurosurgical hospital.

Machine learning to differentiate small round cell malignant tumors and non-small round cell malignant tumors of the nasal and paranasal sinuses using apparent diffusion coefficient values

OBJECTIVE

We used radiomics feature-based machine learning classifiers of apparent diffusion coefficient (ADC) maps to differentiate small round cell malignant tumors (SRCMTs) and non-SRCMTs of the nasal and paranasal sinuses.

MATERIALS

A total of 267 features were extracted from each region of interest (ROI). Datasets were randomized into two sets, a training set (∼70%) and a test set (∼30%). We performed dimensional reductions using the Pearson correlation coefficient and feature selection analyses (analysis of variance [ANOVA], relief, recursive feature elimination [RFE]) and classifications using 10 machine learning classifiers. Results were evaluated with a leave-one-out cross-validation analysis.

RESULTS

We compared the AUC for all the pipelines in the validation dataset using FeAture Explorer (FAE) software. The pipeline using RFE feature selection and Gaussian process classifier yielded the highest AUCs with ten features. When the "one-standard error" rule was used, FAE produced a simpler model with eight features, including Perc.01%, Perc.10%, Perc.90%, Perc.99%, S(1,0) SumAverg, S(5,5) AngScMom, S(5,5) Correlat, and WavEnLH_s-2. The AUCs of the training, validation, and test datasets achieved 0.995, 0.902, and 0.710, respectively. For ANOVA, the pipeline with the auto-encoder classifier yielded the highest AUC using only one feature, Perc.10% (training/validation/test datasets: 0.886/0.895/0.809, respectively). For the relief, the AUCs of the training, validation, and test datasets that used the LRLasso classifier using five features (Perc.01%, Perc.10%, S(4,4) Correlat, S(5,0) SumAverg, S(5,0) Contrast) were 0.892, 0.886, and 0.787, respectively. Compared with the RFE and relief, the results of all algorithms of ANOVA feature selection were more stable with the AUC values higher than 0.800.

CONCLUSIONS

We demonstrated the feasibility of combining artificial intelligence with the radiomics from ADC values in the differential diagnosis of SRCMTs and non-SRCMTs and the potential of this non-invasive approach for clinical applications.

KEY POINTS

• The parameter with the best diagnostic performance in differentiating SRCMTs from non-SRCMTs was the Perc.10% ADC value. • Results of all the algorithms of ANOVA feature selection were more stable and the AUCs were higher than 0.800, as compared with RFE and relief. • The pipeline using RFE feature selection and Gaussian process classifier yielded the highest AUC.

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.

Texture Analysis of Fat-Suppressed T2-Weighted Magnetic Resonance Imaging and Use of Machine Learning to Discriminate Nasal and Paranasal Sinus Small Round Malignant Cell Tumors

Objective

We used texture analysis and machine learning (ML) to classify small round cell malignant tumors (SRCMTs) and Non-SRCMTs of nasal and paranasal sinus on fat-suppressed T2 weighted imaging (Fs-T2WI).

Materials

Preoperative MRI scans of 164 patients from 1 January 2018 to 1 January 2021 diagnosed with SRCMTs and Non-SRCMTs were included in this study. A total of 271 features were extracted from each regions of interest. Datasets were randomly divided into two sets, including a training set (∼70%) and a test set (∼30%). The Pearson correlation coefficient (PCC) and principal component analysis (PCA) methods were performed to reduce dimensions, and the Analysis of Variance (ANOVA), Kruskal-Wallis (KW), and Recursive Feature Elimination (RFE) and Relief were performed for feature selections. Classifications were performed using 10 ML classifiers. Results were evaluated using a leave one out cross-validation analysis.

Results

We compared the AUC of all pipelines on the validation dataset with FeAture Explorer (FAE) software. The pipeline using a PCC dimension reduction, relief feature selection, and gaussian process (GP) classifier yielded the highest area under the curve (AUC) using 15 features. When the “one-standard error” rule was used, FAE also produced a simpler model with 13 features, including S(5,-5)SumAverg, S(3,0)InvDfMom, Skewness, WavEnHL_s-3, Horzl_GlevNonU, Horzl_RLNonUni, 135dr_GlevNonU, WavEnLL_s-3, Teta4, Teta2, S(5,5)DifVarnc, Perc.01%, and WavEnLH_s-2. The AUCs of the training/validation/test datasets were 1.000/0.965/0.979, and the accuracies, sensitivities, and specificities were 0.890, 0.880, and 0.920, respectively. The best algorithm was GP whose AUCs of the training/validation/test datasets by the two-dimensional reduction methods and four feature selection methods were greater than approximately 0.800. Especially, the AUCs of different datasets were greater than approximately 0.900 using the PCC, RFE/Relief, and GP algorithms.

Conclusions

We demonstrated the feasibility of combining artificial intelligence and the radiomics from Fs-T2WI to differentially diagnose SRCMTs and Non-SRCMTs. This non-invasive approach could be very promising in clinical oncology.

Foundations of Machine Learning-Based Clinical Prediction Modeling: Part V-A Practical Approach to Regression Problems

This chapter goes through the steps required to train and validate a simple, machine learning-based clinical prediction model for any continuous outcome. We supply fully structured code for the readers to download and execute in parallel to this section, as well as a simulated database of 10,000 glioblastoma patients who underwent microsurgery, and predict survival from diagnosis in months. We walk the reader through each step, including import, checking, splitting of data. In terms of pre-processing, we focus on how to practically implement imputation using a k-nearest neighbor algorithm. We also illustrate how to select features based on recursive feature elimination and how to use k-fold cross validation. We demonstrate a generalized linear model, a generalized additive model, a random forest, a ridge regressor, and a Least Absolute Shrinkage and Selection Operator (LASSO) regressor. Specifically for regression, we discuss how to evaluate root mean square error (RMSE), mean average error (MAE), and the R² statistic, as well as how a quantile-quantile plot can be used to assess the performance of the regressor along the spectrum of the outcome variable, similarly to calibration when dealing with binary outcomes. Finally, we explain how to arrive at a measure of variable importance using a universal, nonparametric method.

Foundations of Machine Learning-Based Clinical Prediction Modeling: Part IV-A Practical Approach to Binary Classification Problems

We illustrate the steps required to train and validate a simple, machine learning-based clinical prediction model for any binary outcome, such as, for example, the occurrence of a complication, in the statistical programming language R. To illustrate the methods applied, we supply a simulated database of 10,000 glioblastoma patients who underwent microsurgery, and predict the occurrence of 12-month survival. We walk the reader through each step, including import, checking, and splitting of datasets. In terms of pre-processing, we focus on how to practically implement imputation using a k-nearest neighbor algorithm, and how to perform feature selection using recursive feature elimination. When it comes to training models, we apply the theory discussed in Parts I-III. We show how to implement bootstrapping and to evaluate and select models based on out-of-sample error. Specifically for classification, we discuss how to counteract class imbalance by using upsampling techniques. We discuss how the reporting of a minimum of accuracy, area under the curve (AUC), sensitivity, and specificity for discrimination, as well as slope and intercept for calibration-if possible alongside a calibration plot-is paramount. Finally, we explain how to arrive at a measure of variable importance using a universal, AUC-based method. We provide the full, structured code, as well as the complete glioblastoma survival database for the readers to download and execute in parallel to this section.

A Brief History of Machine Learning in Neurosurgery

The history of machine learning in neurosurgery spans three decades and continues to develop at a rapid pace. The earliest applications of machine learning within neurosurgery were first published in the 1990s as researchers began developing artificial neural networks to analyze structured datasets and supervised tasks. By the turn of the millennium, machine learning had evolved beyond proof-of-concept; algorithms had success detecting tumors in unstructured clinical imaging, and unsupervised learning showed promise for tumor segmentation. Throughout the 2000s, the role of machine learning in neurosurgery was further refined. Well-trained models began to consistently best expert clinicians at brain tumor diagnosis. Additionally, the digitization of the healthcare industry provided ample data for analysis, both structured and unstructured. By the 2010s, the use of machine learning within neurosurgery had exploded. The rapid deployment of an exciting new toolset also led to the growing realization that it may offer marginal benefit at best over conventional logistical regression models for analyzing tabular datasets. Additionally, the widespread adoption of machine learning in neurosurgical clinical practice continues to lag until additional validation can ensure generalizability. Many exciting contemporary applications nonetheless continue to demonstrate the unprecedented potential of machine learning to revolutionize neurosurgery when applied to appropriate clinical challenges.

Foundations of Machine Learning-Based Clinical Prediction Modeling: Part I-Introduction and General Principles

We provide explanations on the general principles of machine learning, as well as analytical steps required for successful machine learning-based predictive modeling, which is the focus of this series. In particular, we define the terms machine learning, artificial intelligence, as well as supervised and unsupervised learning, continuing by introducing optimization, thus, the minimization of an objective error function as the central dogma of machine learning. In addition, we discuss why it is important to separate predictive and explanatory modeling, and most importantly state that a prediction model should not be used to make inferences. Lastly, we broadly describe a classical workflow for training a machine learning model, starting with data pre-processing and feature engineering and selection, continuing on with a training structure consisting of a resampling method, hyperparameter tuning, and model selection, and ending with evaluation of model discrimination and calibration as well as robust internal or external validation of the fully developed model. Methodological rigor and clarity as well as understanding of the underlying reasoning of the internal workings of a machine learning approach are required, otherwise predictive applications despite being strong analytical tools are not well accepted into the clinical routine.

Machine Intelligence in Clinical Neuroscience: Taming the Unchained Prometheus

The democratization of machine learning (ML) through availability of open-source learning libraries, the availability of datasets in the "big data" era, increasing computing power even on mobile devices, and online training resources have both led to an explosion in applications and publications of ML in the clinical neurosciences, but has also enabled a dangerous amount of flawed analyses and cardinal methodological errors committed by benevolent authors. While powerful ML methods are nowadays available to almost anyone and can be applied after just few minutes of familiarizing oneself with these methods, that does not imply that one has mastered these techniques. This textbook for clinicians aims to demystify ML by illustrating its methodological foundations, as well as some specific applications throughout clinical neuroscience, and its limitations. While our mind can recognize, abstract, and deal with the many uncertainties in clinical practice, algorithms cannot. Algorithms must remain tools of our own mind, tools that we should be able to master, control, and apply to our advantage in an adjunctive manner. Our hope is that this book inspires and instructs physician-scientists to continue to develop the seeds that have been planted for machine intelligence in clinical neuroscience, not forgetting their inherent limitations.

Machine learning in deep brain stimulation: A systematic review

Deep Brain Stimulation (DBS) is an increasingly common therapy for a large range of neurological disorders, such as abnormal movement disorders. The effectiveness of DBS in terms of controlling patient symptomatology has made this procedure increasingly used over the past few decades. Concurrently, the popularity of Machine Learning (ML), a subfield of artificial intelligence, has skyrocketed and its influence has more recently extended to medical domains such as neurosurgery. Despite its growing research interest, there has yet to be a literature review specifically on the use of ML in DBS. We have followed a fully systematic methodology to obtain a corpus of 73 papers. In each paper, we identified the clinical application, the type/amount of data used, the method employed, and the validation strategy, further decomposed into 12 different sub-categories. The papers overall illustrated some existing trends in how ML is used in the context of DBS, including the breath of the problem domain and evolving techniques, as well as common frameworks and limitations. This systematic review analyzes at a broad level how ML have been recently used to address clinical problems on DBS, giving insight into how these new computational methods are helping to push the state-of-the-art of functional neurosurgery. DBS clinical workflow is complex, involves many specialists, and raises several clinical issues which have partly been addressed with artificial intelligence. However, several areas remain and those that have been recently addressed with ML are by no means considered "solved" by the community nor are they closed to new and evolving methods.

Artificial Intelligence in Neurosurgery: a Systematic Review Using Topic Modeling. Part I: Major Research Areas

In recent years, the number of scientific publications on artificial intelligence (AI), primarily on machine learning, with respect to neurosurgery, has increased.

The aim of the study was to conduct a systematic literature review and identify the main areas of AI applications in neurosurgery.