1
|
Buyck F, Vandemeulebroucke J, Ceranka J, Van Gestel F, Cornelius JF, Duerinck J, Bruneau M. Computer-vision based analysis of the neurosurgical scene - A systematic review. BRAIN & SPINE 2023; 3:102706. [PMID: 38020988 PMCID: PMC10668095 DOI: 10.1016/j.bas.2023.102706] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 08/30/2023] [Revised: 10/23/2023] [Accepted: 10/29/2023] [Indexed: 12/01/2023]
Abstract
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.
Collapse
Affiliation(s)
- Félix Buyck
- Department of Neurosurgery, Universitair Ziekenhuis Brussel (UZ Brussel), 1090, Brussels, Belgium
- Vrije Universiteit Brussel (VUB), Research group Center For Neurosciences (C4N-NEUR), 1090, Brussels, Belgium
| | - Jef Vandemeulebroucke
- Vrije Universiteit Brussel (VUB), Department of Electronics and Informatics (ETRO), 1050, Brussels, Belgium
- Department of Radiology, Universitair Ziekenhuis Brussel (UZ Brussel), 1090, Brussels, Belgium
- imec, 3001, Leuven, Belgium
| | - Jakub Ceranka
- Vrije Universiteit Brussel (VUB), Department of Electronics and Informatics (ETRO), 1050, Brussels, Belgium
- imec, 3001, Leuven, Belgium
| | - Frederick Van Gestel
- Department of Neurosurgery, Universitair Ziekenhuis Brussel (UZ Brussel), 1090, Brussels, Belgium
- Vrije Universiteit Brussel (VUB), Research group Center For Neurosciences (C4N-NEUR), 1090, Brussels, Belgium
| | - Jan Frederick Cornelius
- Department of Neurosurgery, Medical Faculty, Heinrich-Heine-University, 40225, Düsseldorf, Germany
| | - Johnny Duerinck
- Department of Neurosurgery, Universitair Ziekenhuis Brussel (UZ Brussel), 1090, Brussels, Belgium
- Vrije Universiteit Brussel (VUB), Research group Center For Neurosciences (C4N-NEUR), 1090, Brussels, Belgium
| | - Michaël Bruneau
- Department of Neurosurgery, Universitair Ziekenhuis Brussel (UZ Brussel), 1090, Brussels, Belgium
- Vrije Universiteit Brussel (VUB), Research group Center For Neurosciences (C4N-NEUR), 1090, Brussels, Belgium
| |
Collapse
|
2
|
Greenberg JK, Landman JM, Kelly MP, Pennicooke BH, Molina CA, Foraker RE, Ray WZ. Leveraging Artificial Intelligence and Synthetic Data Derivatives for Spine Surgery Research. Global Spine J 2023; 13:2409-2421. [PMID: 35373623 PMCID: PMC10538345 DOI: 10.1177/21925682221085535] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 12/16/2022] Open
Abstract
STUDY DESIGN Retrospective cohort study. OBJECTIVES Leveraging electronic health records (EHRs) for spine surgery research is impeded by concerns regarding patient privacy and data ownership. Synthetic data derivatives may help overcome these limitations. This study's objective was to validate the use of synthetic data for spine surgery research. METHODS Data came from the EHR from 15 hospitals. Patients that underwent anterior cervical or posterior lumbar fusion (2010-2020) were included. Real data were obtained from the EHR. Synthetic data was generated to simulate the properties of the real data, without maintaining a one-to-one correspondence with real patients. Within each cohort, ability to predict 30-day readmissions and 30-day complications was evaluated using logistic regression and extreme gradient boosting machines (XGBoost). RESULTS We identified 9,072 real and 9,088 synthetic cervical fusion patients. Descriptive characteristics were nearly identical between the 2 datasets. When predicting readmission, models built using real and synthetic data both had c-statistics of .69-.71 using logistic regression and XGBoost. Among 12,111 real and 12,126 synthetic lumbar fusion patients, descriptive characteristics were nearly the same for most variables. Using logistic regression and XGBoost to predict readmission, discrimination was similar with models built using real and synthetic data (c-statistics .66-.69). When predicting complications, models derived using real and synthetic data showed similar discrimination in both cohorts. Despite some differences, the most influential predictors were similar in the real and synthetic datasets. CONCLUSION Synthetic data replicate most descriptive and predictive properties of real data, and therefore may expand EHR research in spine surgery.
Collapse
Affiliation(s)
- Jacob K. Greenberg
- Departments of Neurological Surgery, Medicine and Orthopaedic Surgery, Washington University School of Medicine in St Louis, St Louis, MO, USA
| | - Joshua M. Landman
- Departments of Neurological Surgery, Medicine and Orthopaedic Surgery, Washington University School of Medicine in St Louis, St Louis, MO, USA
| | | | - Brenton H. Pennicooke
- Departments of Neurological Surgery, Medicine and Orthopaedic Surgery, Washington University School of Medicine in St Louis, St Louis, MO, USA
| | - Camilo A. Molina
- Departments of Neurological Surgery, Medicine and Orthopaedic Surgery, Washington University School of Medicine in St Louis, St Louis, MO, USA
| | | | - Wilson Z. Ray
- Departments of Neurological Surgery, Medicine and Orthopaedic Surgery, Washington University School of Medicine in St Louis, St Louis, MO, USA
| |
Collapse
|
3
|
Cheung ATM, Nasir-Moin M, Fred Kwon YJ, Guan J, Liu C, Jiang L, Raimondo C, Chotai S, Chambless L, Ahmad HS, Chauhan D, Yoon JW, Hollon T, Buch V, Kondziolka D, Chen D, Al-Aswad LA, Aphinyanaphongs Y, Oermann EK. Methods and Impact for Using Federated Learning to Collaborate on Clinical Research. Neurosurgery 2023; 92:431-438. [PMID: 36399428 DOI: 10.1227/neu.0000000000002198] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/05/2022] [Accepted: 08/20/2022] [Indexed: 11/21/2022] Open
Abstract
BACKGROUND The development of accurate machine learning algorithms requires sufficient quantities of diverse data. This poses a challenge in health care because of the sensitive and siloed nature of biomedical information. Decentralized algorithms through federated learning (FL) avoid data aggregation by instead distributing algorithms to the data before centrally updating one global model. OBJECTIVE To establish a multicenter collaboration and assess the feasibility of using FL to train machine learning models for intracranial hemorrhage (ICH) detection without sharing data between sites. METHODS Five neurosurgery departments across the United States collaborated to establish a federated network and train a convolutional neural network to detect ICH on computed tomography scans. The global FL model was benchmarked against a standard, centrally trained model using a held-out data set and was compared against locally trained models using site data. RESULTS A federated network of practicing neurosurgeon scientists was successfully initiated to train a model for predicting ICH. The FL model achieved an area under the ROC curve of 0.9487 (95% CI 0.9471-0.9503) when predicting all subtypes of ICH compared with a benchmark (non-FL) area under the ROC curve of 0.9753 (95% CI 0.9742-0.9764), although performance varied by subtype. The FL model consistently achieved top three performance when validated on any site's data, suggesting improved generalizability. A qualitative survey described the experience of participants in the federated network. CONCLUSION This study demonstrates the feasibility of implementing a federated network for multi-institutional collaboration among clinicians and using FL to conduct machine learning research, thereby opening a new paradigm for neurosurgical collaboration.
Collapse
Affiliation(s)
| | | | | | | | - Chris Liu
- Department of Neurosurgery, NYU Langone Health, New York, New York, USA
| | - Lavender Jiang
- Department of Neurosurgery, NYU Langone Health, New York, New York, USA.,Center for Data Science, New York University, New York, New York, USA
| | | | - Silky Chotai
- Department of Neurosurgery, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Lola Chambless
- Department of Neurosurgery, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Hasan S Ahmad
- Department of Neurosurgery, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
| | - Daksh Chauhan
- Department of Neurosurgery, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
| | - Jang W Yoon
- Department of Neurosurgery, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
| | - Todd Hollon
- Department of Neurosurgery, University of Michigan School of Medicine, Ann Arbor, Michigan, USA
| | - Vivek Buch
- Department of Neurosurgery, Stanford University School of Medicine, Stanford, California, USA
| | | | - Dinah Chen
- Department of Ophthalmology, NYU Langone Health, New York, New York, USA
| | - Lama A Al-Aswad
- Department of Ophthalmology, NYU Langone Health, New York, New York, USA
| | | | - Eric Karl Oermann
- Department of Neurosurgery, NYU Langone Health, New York, New York, USA.,Center for Data Science, New York University, New York, New York, USA.,Department of Radiology, NYU Langone Health, New York, New York, USA
| |
Collapse
|
4
|
Kernbach JM, Staartjes VE. Foundations of Machine Learning-Based Clinical Prediction Modeling: Part I-Introduction and General Principles. ACTA NEUROCHIRURGICA. SUPPLEMENT 2021; 134:7-13. [PMID: 34862522 DOI: 10.1007/978-3-030-85292-4_2] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Subscribe] [Scholar Register] [Indexed: 11/28/2022]
Abstract
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.
Collapse
Affiliation(s)
- Julius M Kernbach
- Neurosurgical Artificial Intelligence Laboratory Aachen (NAILA), Department of Neurosurgery, RWTH Aachen University Hospital, Aachen, Germany
| | - Victor E Staartjes
- Machine Intelligence in Clinical Neuroscience (MICN) Laboratory, Department of Neurosurgery, Clinical Neuroscience Center, University Hospital Zurich, University of Zurich, Zurich, Switzerland.
| |
Collapse
|
5
|
Hale AT, Riva-Cambrin J, Wellons JC, Jackson EM, Kestle JRW, Naftel RP, Hankinson TC, Shannon CN. Machine learning predicts risk of cerebrospinal fluid shunt failure in children: a study from the hydrocephalus clinical research network. Childs Nerv Syst 2021; 37:1485-1494. [PMID: 33515058 DOI: 10.1007/s00381-021-05061-7] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 01/05/2021] [Accepted: 01/22/2021] [Indexed: 10/22/2022]
Abstract
PURPOSE While conventional statistical approaches have been used to identify risk factors for cerebrospinal fluid (CSF) shunt failure, these methods may not fully capture the complex contribution of clinical, radiologic, surgical, and shunt-specific variables influencing this outcome. Using prospectively collected data from the Hydrocephalus Clinical Research Network (HCRN) patient registry, we applied machine learning (ML) approaches to create a predictive model of CSF shunt failure. METHODS Pediatric patients (age < 19 years) undergoing first-time CSF shunt placement at six HCRN centers were included. CSF shunt failure was defined as a composite outcome including requirement for shunt revision, endoscopic third ventriculostomy, or shunt infection within 5 years of initial surgery. Performance of conventional statistical and 4 ML models were compared. RESULTS Our cohort consisted of 1036 children undergoing CSF shunt placement, of whom 344 (33.2%) experienced shunt failure. Thirty-eight clinical, radiologic, surgical, and shunt-design variables were included in the ML analyses. Of all ML algorithms tested, the artificial neural network (ANN) had the strongest performance with an area under the receiver operator curve (AUC) of 0.71. The ANN had a specificity of 90% and a sensitivity of 68%, meaning that the ANN can effectively rule-in patients most likely to experience CSF shunt failure (i.e., high specificity) and moderately effective as a tool to rule-out patients at high risk of CSF shunt failure (i.e., moderately sensitive). The ANN was independently validated in 155 patients (prospectively collected, retrospectively analyzed). CONCLUSION These data suggest that the ANN, or future iterations thereof, can provide an evidence-based tool to assist in prognostication and patient-counseling immediately after CSF shunt placement.
Collapse
Affiliation(s)
- Andrew T Hale
- Medical Scientist Training Program, Vanderbilt University School of Medicine, 2200 Pierce Ave., Light Hall 514, Nashville, TN, 37232, USA. .,Surgical Outcomes Center for Kids, Monroe Carell Jr. Children's Hospital of Vanderbilt University, Nashville, TN, USA.
| | - Jay Riva-Cambrin
- Department of Clinical Neurosciences, Alberta Children's Hospital, University of Calgary, Calgary, Alberta, Canada
| | - John C Wellons
- Surgical Outcomes Center for Kids, Monroe Carell Jr. Children's Hospital of Vanderbilt University, Nashville, TN, USA.,Division of Pediatric Neurosurgery, Monroe Carell Jr. Children's Hospital of Vanderbilt University, Nashville, TN, USA
| | - Eric M Jackson
- Department of Neurosurgery, Johns Hopkins Children's Center, Johns Hopkins University School of Medicine, Baltimore, MD, USA
| | - John R W Kestle
- Department of Neurosurgery, University of Utah, Salt Lake City, UT, USA
| | - Robert P Naftel
- Surgical Outcomes Center for Kids, Monroe Carell Jr. Children's Hospital of Vanderbilt University, Nashville, TN, USA.,Division of Pediatric Neurosurgery, Monroe Carell Jr. Children's Hospital of Vanderbilt University, Nashville, TN, USA
| | - Todd C Hankinson
- Division of Pediatric Neurosurgery, Children's Hospital Colorado, Aurora, CO, USA
| | - Chevis N Shannon
- Surgical Outcomes Center for Kids, Monroe Carell Jr. Children's Hospital of Vanderbilt University, Nashville, TN, USA.,Division of Pediatric Neurosurgery, Monroe Carell Jr. Children's Hospital of Vanderbilt University, Nashville, TN, USA
| | | |
Collapse
|
6
|
Velagapudi L, Mouchtouris N, Baldassari MP, Nauheim D, Khanna O, Saiegh FA, Herial N, Gooch MR, Tjoumakaris S, Rosenwasser RH, Jabbour P. Discrepancies in Stroke Distribution and Dataset Origin in Machine Learning for Stroke. J Stroke Cerebrovasc Dis 2021; 30:105832. [PMID: 33940363 DOI: 10.1016/j.jstrokecerebrovasdis.2021.105832] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/28/2021] [Revised: 04/11/2021] [Accepted: 04/11/2021] [Indexed: 01/02/2023] Open
Abstract
BACKGROUND Machine learning algorithms depend on accurate and representative datasets for training in order to become valuable clinical tools that are widely generalizable to a varied population. We aim to conduct a review of machine learning uses in stroke literature to assess the geographic distribution of datasets and patient cohorts used to train these models and compare them to stroke distribution to evaluate for disparities. AIMS 582 studies were identified on initial searching of the PubMed database. Of these studies, 106 full texts were assessed after title and abstract screening which resulted in 489 papers excluded. Of these 106 studies, 79 were excluded due to using cohorts from outside the United States or being review articles or editorials. 27 studies were thus included in this analysis. SUMMARY OF REVIEW Of the 27 studies included, 7 (25.9%) used patient data from California, 6 (22.2%) were multicenter, 3 (11.1%) were in Massachusetts, 2 (7.4%) each in Illinois, Missouri, and New York, and 1 (3.7%) each from South Carolina, Washington, West Virginia, and Wisconsin. 1 (3.7%) study used data from Utah and Texas. These were qualitatively compared to a CDC study showing the highest distribution of stroke in Mississippi (4.3%) followed by Oklahoma (3.4%), Washington D.C. (3.4%), Louisiana (3.3%), and Alabama (3.2%) while the prevalence in California was 2.6%. CONCLUSIONS It is clear that a strong disconnect exists between the datasets and patient cohorts used in training machine learning algorithms in clinical research and the stroke distribution in which clinical tools using these algorithms will be implemented. In order to ensure a lack of bias and increase generalizability and accuracy in future machine learning studies, datasets using a varied patient population that reflects the unequal distribution of stroke risk factors would greatly benefit the usability of these tools and ensure accuracy on a nationwide scale.
Collapse
Affiliation(s)
- Lohit Velagapudi
- Department of Neurosurgery, Thomas Jefferson University, Philadelphia, PA
| | | | | | - David Nauheim
- Department of Neurosurgery, Thomas Jefferson University, Philadelphia, PA
| | - Omaditya Khanna
- Department of Neurosurgery, Thomas Jefferson University, Philadelphia, PA
| | - Fadi Al Saiegh
- Department of Neurosurgery, Thomas Jefferson University, Philadelphia, PA
| | - Nabeel Herial
- Department of Neurosurgery, Thomas Jefferson University, Philadelphia, PA
| | - M Reid Gooch
- Department of Neurosurgery, Thomas Jefferson University, Philadelphia, PA
| | | | | | - Pascal Jabbour
- Department of Neurosurgery, Thomas Jefferson University, Philadelphia, PA.
| |
Collapse
|
7
|
Raju B, Jumah F, Ashraf O, Narayan V, Gupta G, Sun H, Hilden P, Nanda A. Big data, machine learning, and artificial intelligence: a field guide for neurosurgeons. J Neurosurg 2020; 135:373-383. [PMID: 33007750 DOI: 10.3171/2020.5.jns201288] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/16/2020] [Accepted: 05/27/2020] [Indexed: 11/06/2022]
Abstract
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.
Collapse
Affiliation(s)
- Bharath Raju
- 1Department of Neurosurgery, Rutgers-Robert Wood Johnson Medical School and University Hospital; and
| | - Fareed Jumah
- 1Department of Neurosurgery, Rutgers-Robert Wood Johnson Medical School and University Hospital; and
| | - Omar Ashraf
- 1Department of Neurosurgery, Rutgers-Robert Wood Johnson Medical School and University Hospital; and
| | - Vinayak Narayan
- 1Department of Neurosurgery, Rutgers-Robert Wood Johnson Medical School and University Hospital; and
| | - Gaurav Gupta
- 1Department of Neurosurgery, Rutgers-Robert Wood Johnson Medical School and University Hospital; and
| | - Hai Sun
- 1Department of Neurosurgery, Rutgers-Robert Wood Johnson Medical School and University Hospital; and
| | - Patrick Hilden
- 2Rutgers Neurosurgery Health Outcomes, Policy, and Economics (HOPE) Center, New Brunswick, New Jersey
| | - Anil Nanda
- 1Department of Neurosurgery, Rutgers-Robert Wood Johnson Medical School and University Hospital; and
| |
Collapse
|