ABSTRACT
OBJECTIVES: Non-contrast computed tomography of the brain (NCCTB) is commonly used to detect intracranial pathology but is subject to interpretation errors. Machine learning can augment clinical decision-making and improve NCCTB scan interpretation. This retrospective detection accuracy study assessed the performance of radiologists assisted by a deep learning model and compared the standalone performance of the model with that of unassisted radiologists. METHODS: A deep learning model was trained on 212,484 NCCTB scans drawn from a private radiology group in Australia. Scans from inpatient, outpatient, and emergency settings were included. Scan inclusion criteria were age ≥ 18 years and series slice thickness ≤ 1.5 mm. Thirty-two radiologists reviewed 2848 scans with and without the assistance of the deep learning system and rated their confidence in the presence of each finding using a 7-point scale. Differences in AUC and Matthews correlation coefficient (MCC) were calculated using a ground-truth gold standard. RESULTS: The model demonstrated an average area under the receiver operating characteristic curve (AUC) of 0.93 across 144 NCCTB findings and significantly improved radiologist interpretation performance. Assisted and unassisted radiologists demonstrated an average AUC of 0.79 and 0.73 across 22 grouped parent findings and 0.72 and 0.68 across 189 child findings, respectively. When assisted by the model, radiologist AUC was significantly improved for 91 findings (158 findings were non-inferior), and reading time was significantly reduced. CONCLUSIONS: The assistance of a comprehensive deep learning model significantly improved radiologist detection accuracy across a wide range of clinical findings and demonstrated the potential to improve NCCTB interpretation. CLINICAL RELEVANCE STATEMENT: This study evaluated a comprehensive CT brain deep learning model, which performed strongly, improved the performance of radiologists, and reduced interpretation time. The model may reduce errors, improve efficiency, facilitate triage, and better enable the delivery of timely patient care. KEY POINTS: ⢠This study demonstrated that the use of a comprehensive deep learning system assisted radiologists in the detection of a wide range of abnormalities on non-contrast brain computed tomography scans. ⢠The deep learning model demonstrated an average area under the receiver operating characteristic curve of 0.93 across 144 findings and significantly improved radiologist interpretation performance. ⢠The assistance of the comprehensive deep learning model significantly reduced the time required for radiologists to interpret computed tomography scans of the brain.
Subject(s)
Deep Learning , Adolescent , Humans , Radiography , Radiologists , Retrospective Studies , Tomography, X-Ray Computed/methods , AdultABSTRACT
Natural language processing (NLP), a domain of artificial intelligence (AI) that models human language, has been used in medicine to automate diagnostics, detect adverse events, support decision making and predict clinical outcomes. However, applications to the clinical neurosciences appear to be limited. NLP has matured with the implementation of deep transformer models (e.g., XLNet, BERT, T5, and RoBERTa) and transfer learning. The objectives of this study were to (1) systematically review NLP applications in the clinical neurosciences, and (2) explore NLP analysis to facilitate literature synthesis, providing clear examples to demonstrate the potential capabilities of these technologies for a clinical audience. Our NLP analysis consisted of keyword identification, text summarization and document classification. A total of 48 articles met inclusion criteria. NLP has been applied in the clinical neurosciences to facilitate literature synthesis, data extraction, patient identification, automated clinical reporting and outcome prediction. The number of publications applying NLP has increased rapidly over the past five years. Document classifiers trained to differentiate included and excluded articles demonstrated moderate performance (XLNet AUC = 0.66, BERT AUC = 0.59, RoBERTa AUC = 0.62). The T5 transformer model generated acceptable abstract summaries. The application of NLP has the potential to enhance research and practice in the clinical neurosciences.
Subject(s)
Natural Language Processing , Neurosciences , Artificial Intelligence , Humans , Machine LearningABSTRACT
Machine learning (ML) involves algorithms learning patterns in large, complex datasets to predict and classify. Algorithms include neural networks (NN), logistic regression (LR), and support vector machines (SVM). ML may generate substantial improvements in neurosurgery. This systematic review assessed the current state of neurosurgical ML applications and the performance of algorithms applied. Our systematic search strategy yielded 6866 results, 70 of which met inclusion criteria. Performance statistics analyzed included area under the receiver operating characteristics curve (AUC), accuracy, sensitivity, and specificity. Natural language processing (NLP) was used to model topics across the corpus and to identify keywords within surgical subspecialties. ML applications were heterogeneous. The densest cluster of studies focused on preoperative evaluation, planning, and outcome prediction in spine surgery. The main algorithms applied were NN, LR, and SVM. Input and output features varied widely and were listed to facilitate future research. The accuracy (F(2,19) = 6.56, p < 0.01) and specificity (F(2,16) = 5.57, p < 0.01) of NN, LR, and SVM differed significantly. NN algorithms demonstrated significantly higher accuracy than LR. SVM demonstrated significantly higher specificity than LR. We found no significant difference between NN, LR, and SVM AUC and sensitivity. NLP topic modeling reached maximum coherence at seven topics, which were defined by modeling approach, surgery type, and pathology themes. Keywords captured research foci within surgical domains. ML technology accurately predicts outcomes and facilitates clinical decision-making in neurosurgery. NNs frequently outperformed other algorithms on supervised learning tasks. This study identified gaps in the literature and opportunities for future neurosurgical ML research.
Subject(s)
Artificial Intelligence , Decision Support Systems, Clinical , Machine Learning , Neurosurgery/methods , Deep Learning , Humans , Neurosurgical Procedures/methods , Support Vector MachineABSTRACT
OBJECTIVES: The aim of the study was to train and test supervised machine-learning classifiers to predict acoustic hearing preservation after CI using preoperative clinical data. STUDY DESIGN: Retrospective predictive modeling study of prospectively collected single-institution CI dataset. METHODS: One hundred and seventy-five patients from a REDCap database including 761 patients >18 years who underwent CI and had audiometric testing preoperatively and one month after surgery were included. The primary outcome variable was the lowest quartile change in acoustic hearing at one month after CI using various formulae (standard pure tone average, SPTA; low-frequency PTA, LFPTA). Analysis involved applying multivariate logistic regression to detect statistical associations and training and testing supervised learning classifiers. Classifier performance was assessed with numerous metrics including area under the receiver operating characteristic curve (AUC) and Matthews correlation coefficient (MCC). RESULTS: Lowest quartile change (indicating hearing preservation) in SPTA was positively associated with a history of meningitis, preoperative LFPTA, and preoperative SPTA. Lowest quartile change in SPTA was negatively associated with sudden hearing loss, noise exposure, aural fullness, and abnormal anatomy. Lowest quartile change in LFPTA was positively associated with preoperative LFPTA. Lowest quartile change in LFPTA was negatively associated with tobacco use. Random forest demonstrated the highest mean classification performance on the validation dataset when predicting each of the outcome variables. CONCLUSIONS: Machine learning demonstrated utility for predicting preservation of residual acoustic hearing in patients undergoing CI surgery, and the detected associations facilitated the interpretation of our machine-learning models. The models and statistical associations together may be used to facilitate improvements in shared clinical decision-making and patient outcomes. LEVEL OF EVIDENCE: 3 Laryngoscope, 134:926-936, 2024.
Subject(s)
Cochlear Implantation , Cochlear Implants , Humans , Retrospective Studies , Treatment Outcome , Hearing , Machine Learning , Acoustics , Audiometry, Pure-ToneABSTRACT
Background: Central Nervous System Tuberculosis (CNS-TB) is a serious public health concern causing significant morbidity and mortality, especially in high TB burden countries. Despite the expanding research landscape of CNS-TB, there is no comprehensive map of this field. This work aims to (1) obtain a current and comprehensive overview of the CNS-TB research landscape, (2) investigate the intellectual and social structure of CNS-TB publications, and (3) detect geographical discrepancies in scientific production, highlighting regions requiring increased research focus. Methods: We conducted a bibliometric analysis on CNS-TB literature indexed in Web of Science from 2000 to 2022, evaluating 2130 articles. The dataset was analyzed in R for descriptive statistics. We used R-bibliometrix and VOSViewer for data visualization. Findings: Publication output grew annually at an average rate of 6·88%, driven primarily by India and China. International collaborations comprised 16·44% of total publications but contributed to 11 of the 15 top-cited papers. Additionally, we identified discrepancies of CNS-TB research in many low- and middleincome countries relative to their TB incidence. Interpretation: Our findings reveal a growing interest in CNS-TB research from China and India, countries with rapidly developing economies, high TB burdens, and a recent increase in research funding. Furthermore, we found that international collaborations are correlated with high impact and accessibility of CNS-TB research. Finally, we identified disparities in CNS-TB research in specific countries, particularly in many low- and middle-income countries, emphasizing the need for increased research focus in these regions.
ABSTRACT
BACKGROUND: tinnitus is a common and often debilitating condition with limited evidence-based treatment options. Deep brain stimulation (DBS) is an approved treatment modality for certain neurological conditions; its experimental use as a treatment modality for severe tinnitus is novel and beginning to show promise. This systematic review focuses on the current evidence for the safety and efficacy of DBS for treatment of refractory tinnitus. METHODS: a systematic search in PubMed and EMBASE was performed to identify peer-reviewed studies on DBS of non-cortical structures for the primary indication of tinnitus treatment. Three studies were identified as meeting these criteria, one of which had two related sub-studies. RESULTS: seven patients with available data who underwent DBS for tinnitus were identified. DBS targets included nucleus accumbens (NAc), ventral anterior limb of the internal capsule (vALIC), caudate nucleus, and the medial geniculate body (MGB) of the thalamus. All studies used the Tinnitus Functional Index (TFI) as a primary outcome measure. DBS of the caudate was most commonly reported (n = 5), with a mean TFI improvement of 23.3 points. Only one subject underwent DBS targeting the NAc/vALIC (extrapolated TFI improvement 46.8) and one subject underwent DBS targeting the MGB (TFI improvement 59 points). CONCLUSIONS: DBS is a promising treatment option for refractory subjective tinnitus, with early data, from small patient cohorts in multiple studies, suggesting its safety and efficacy. Further studies with a larger patient population are needed to support this safety and efficacy before implementing this treatment to daily practice.
ABSTRACT
OBJECTIVE: Understanding ergonomic impact is foundational to critically evaluating value and safety of enabling technologies in minimally invasive spine surgeries. This study assessed the impact of a tubular-mounted digital camera (TMDC) versus an optical surgical microscope (OSM) in single-level minimally invasive spine surgeries on operative times, durotomy rate, surgeon ergonomics, safety, and operating room workflow. METHODS: This retrospective study compared consecutive single-level minimally invasive lumbar decompression surgeries in a TMDC cohort (September 2021-June 2022) with an historical OSM cohort (January 2020-July 2021). Data included patient demographics, operative times, durotomy incidence, surgeon ergonomics (Rapid Entire Body Assessment scores), and equipment impact via staff surveys. Operative times were assessed by t test, while Pearson χ2 test compared sex. Age, body mass index, and Charlson Comorbidity Index comparisons were made by Wilcoxon rank sum tests, and survey results were analyzed with Wilcoxon signed rank tests. RESULTS: TMDC and OSM groups included 74 and 82 patients, respectively. Age, sex, and Charlson Comorbidity Index did not significantly differ between groups. The TMDC group had a higher body mass index (29.6 ± 5.1) than the OSM group (29.0 ± 7.5) (P = 0.04). The TMDC group had significantly shorter operative times (57.3 ± 16.6 minutes) than the OSM group) (66.7 ± 22.5 minutes) (P = 0.004), with no difference in durotomy rates (P = 0.42). TMDC use yielded lower Rapid Entire Body Assessment scores compared with OSM (4.1 ± 0.77) (P < 0.001). Surveys indicated improved safety, setup time, and workflow with TMDC (P < 0.001). CONCLUSIONS: TMDC in single-level minimally invasive lumbar decompression surgery improved surgeon ergonomics, reduced operative times, and maintained durotomy rates, enhancing operating room efficiency. Evaluating ergonomic impact of technology is vital for safety and value assessment.
Subject(s)
Lumbar Vertebrae , Spinal Fusion , Humans , Retrospective Studies , Lumbar Vertebrae/surgery , Operative Time , Workflow , Spinal Fusion/methods , Decompression, Surgical , Minimally Invasive Surgical Procedures/methods , Treatment OutcomeABSTRACT
BACKGROUND: Posterior cervical decompression is a common spine procedure that can be performed with the patient in prone or sitting position. The sitting position provides the potential benefits of more facile retraction of surrounding soft tissues, increased operative field and fluoroscopic visualization, and decreased epidural bleeding. However, the surgeon's ergonomics of this positioning can be quite challenging when using the standard operative microscope to perform the procedure and may cause musculoskeletal harm to the surgeon. METHODS: A sterile digital camera was brought into the field to perform a sitting foraminotomy completed through the tube retractor at both C6-7 and C7-T1 levels. For half of the procedure, a typical neurosurgical operative microscope was brought into the field to evaluate surgeon ergonomics using baseline Rapid Entire Body Assessment (REBA) scores for 2 surgeons of differing stature. The digital camera was inserted onto the tubular retractor, and REBA scores were calculated. RESULTS: With a microscope, the surgeon with taller stature scored a 5 on the initial REBA scale, and the surgeon with shorter stature scored a 6, placing both in the medium-risk category. Once the tubular-based camera was placed, repeated REBA score of both surgeons was 3, placing them in the low-risk category. CONCLUSIONS: Using a tubular-based digital camera system, the ergonomics of the surgery are substantially improved. The surgeon can stand closer to the operative field and look directly at a front-facing screen, allowing increased relaxation of the upper extremity and cervical musculature; improving overall ergonomic function.
Subject(s)
Foraminotomy , Surgeons , Humans , Sitting Position , Neck , ErgonomicsABSTRACT
Chronic cluster headache (CCH) is a debilitating primary headache that causes excruciating pain without remission. Various medical and surgical treatments have been implemented over the years, yet many provide only short-term relief. Deep brain stimulation (DBS) is an emerging treatment alternative that has been shown to dramatically reduce the intensity and frequency of headache attacks. However, reports of greater than 10-year outcomes after DBS for CCH are scant. Here, we report the durability of DBS in the posterior inferior hypothalamus after 10 years on a patient with CCH. Our patient experienced an 82% decrease in the frequency of headaches after DBS, which was maintained for over 10 years. The side effects observed included depression, irritability, anxiety, and dizziness, which were alleviated by changing programming settings. In the context of current literature, DBS shows promise for long-term relief of cluster headaches when other treatments fail.
ABSTRACT
OBJECTIVE: To compare information online regarding lumbar disc herniation (LDH) on commonly searched websites and compare those findings with the evidence-based recommendations listed in the North American Spine Society (NASS) clinical practice guidelines. METHODS: NASS Clinical Practice Guidelines, Internet searches were performed utilizing three common search engines (Google, Bing, Yahoo) and keywords associated with LDH. The top 20 websites from each search were selected. The content regarding diagnosis and treatment of LDH was compared to the NASS clinical practice guidelines. RESULTS: On average, websites mentioned only 59% of recommendations supported by Level I evidence. Websites included an average of 3 recommendations not discussed in the NASS guidelines out of an average of 12 total recommendations. Muscle and sensory testing and physical therapy were the most frequent recommendations, appearing on over 80% of websites. Websites were equally likely to contain recommendations backed by high-quality evidence as recommendations not included in NASS guidelines. CONCLUSIONS: This study demonstrates that websites regarding LDH contain a mix of information, with only a fraction of recommendations aligning with NASS clinical guidelines. Patients who use these websites are presented with unsubstantiated information, conceivably impacting their understanding, expectations and decision-making in physician offices.
ABSTRACT
Key Clinical Message: Balloon kyphoplasty is a promising treatment option for osteoporotic vertebral compression fractures with posterior cortical defect, offering pain relief, vertebral height restoration, and low risk of cement leakage. Abstract: Millions of people worldwide suffer from osteoporotic vertebral compression fractures (OVCFs) annually, which cause pain and functional limitations, particularly in the elderly. Conservative treatments such as pain management, rest, and medication are frequently used, while surgical options such as vertebroplasty and kyphoplasty are considered. We present a case of 68-year-old female with vertebral compression fracture of L1 vertebra with posterior cortical defect and posterior wall retropulsion. She was treated successfully with balloon kyphoplasty. Kyphoplasty appears to be a better option than vertebroplasty in cases with posterior cortical defect due to lower chance of cement leakage.
ABSTRACT
Introduction: The establishment of local neurosurgery training programs in Nepal has proven critical for the expansion of the discipline across the country. This paper aims to describe the evolution, current status, challenges, and future directions of academic neurosurgery in Nepal. Research question: What is the current status and international standing of academic neurosurgery in Nepal? Material and methods: Information related to growth and development in Nepal was obtained from universities and regulatory bodies in Nepal. Variables described are the current number of neurosurgeons, the number of neurosurgical centers and centers with accreditation for training, the description of existing training models, the number of graduates, and the contribution of Nepalese neurosurgeons to world literature. Results: Formal neurosurgical training started in Nepal in 1999. Of 67 hospitals with neurosurgical facilities, 10 (14.9%) are accredited. Three training models (MCh, NBMS, and FCPS) currently exist. Of 116 neurosurgeons currently practicing in the country, 47 (40.5%) are homegrown. The contribution of the Nepalese neurosurgical community to the world includes the training of the first two Maldivian neurosurgeons and an increasing presence in world neurosurgical literature. Conclusions: Although comparable to other countries with similar economies, Nepal still faces some challenges to the sustainability and further developments of Neurosurgery. Continued concerted efforts will help Nepalese neurosurgeons achieve the goal of securing self-reliance in neurosurgical education.
ABSTRACT
OBJECTIVE: Review a single institution's vestibular schwannoma (VS) microsurgery experience to determine (1) correlations between demographics, comorbidities, and/or surgical approach on hospital length of stay (LOS) and discharge disposition and (2) trends in surgical approach over time. METHODS: Retrospective case series from a multidisciplinary skull base program at a tertiary care, academic hospital. All adult (>18 years) patients undergoing primary microsurgery for VS between 2008 and 2018 were included. RESULTS: A total of 147 subjects were identified. Surgical approach was split between middle fossa (MF) (16%), retrosigmoid (RS) (35%), and translabyrinthine (TL) (49%) craniotomies. For the 8% of patients had other than routine (OTR) discharge. Mean LOS was significantly longer for patients undergoing RS than either MF or TL. Brainstem compression by the tumor was associated with longer LOS as were diagnoses of chronic obstructive pulmonary disease (COPD) and peripheral vascular disease (PVD). For all discharges, the 40 to 50- and 50 to 60-year-old subgroups had significantly shorter LOS than the 70-years-and-older patients. For the 92% of patients routinely discharged, there was a significantly shorter LOS in the 40 to 50-year-olds compared to the 70-years-and-older patients. There was a significant shift in surgical approach from RS to TL over the study period. CONCLUSION: Over 90% of VS microsurgery patients were routinely discharged with a median hospital LOS of 3.2 days, both of which are consistent with published data. There is an inverse relationship between age and LOS with patients older than 70 years having significantly longer LOS. Brainstem compression, COPD, PVD, and the RS approach negatively affect LOS. LEVEL OF EVIDENCE: 4.
Subject(s)
Neuroma, Acoustic , Pulmonary Disease, Chronic Obstructive , Adult , Humans , Length of Stay , Microsurgery , Neuroma, Acoustic/surgery , Postoperative Complications/epidemiology , Postoperative Complications/surgery , Retrospective StudiesABSTRACT
Surgical process improvement strategies are increasingly being applied to specific procedures to improve value. A critical step in any process improvement strategy is the identification of performance benchmarks. Procedure length is a performance benchmark for anterior cervical discectomy and fusion (ACDF) procedures; therefore, we sought to establish reference procedure lengths for 1-level, 2-level, and 3-level ACDFs at both teaching and non-teaching institutions and to describe methods for using this information to advance surgical process improvement initiatives. We performed a retrospective analysis of consecutive ACDFs performed at a resident teaching institution (RT) and a non-teaching institution (NT) for all 1-level, 2-level, and 3-level ACDFs. Mean case lengths and patient outcomes were calculated for individual surgeons and institutions. After limiting cases to 1-level, 2-level, and 3-level ACDFs and applying all exclusion criteria, 991 cases at the RT institution and 131 cases at the NT institution (a total of 1122 cases) were available for analysis. The mean (SD) procedure length for 1-level, 2-level, and 3-level ACDFs at the RT versus NT institutions were 121.9 min (36.3 min) and 73.6 min (29.7 min) (p<0.001), 172.7 min (44.8 min) and 112.0 min (43.0 min) (p<0.001), and 218.3 min (54.9 min) and 167.6 min (54.2 min) (p<0.001), respectively. Thirty-day outcomes were the same between institutions, except that the RT institution had a shorter mean hospital length of stay for 2-level ACDFs (1.6 days versus 2.9 days, p=0.001). This study is the first to attempt to establish a standard reference procedure length for 1-level, 2-level, and 3-level ACDFs. These data can guide efforts in surgical process improvement.
ABSTRACT
Brain computed tomography (CTB) scans are widely used to evaluate intracranial pathology. The implementation and adoption of CTB has led to clinical improvements. However, interpretation errors occur and may have substantial morbidity and mortality implications for patients. Deep learning has shown promise for facilitating improved diagnostic accuracy and triage. This research charts the potential of deep learning applied to the analysis of CTB scans. It draws on the experience of practicing clinicians and technologists involved in development and implementation of deep learning-based clinical decision support systems. We consider the past, present and future of the CTB, along with limitations of existing systems as well as untapped beneficial use cases. Implementing deep learning CTB interpretation systems and effectively navigating development and implementation risks can deliver many benefits to clinicians and patients, ultimately improving efficiency and safety in healthcare.
Subject(s)
Decision Support Systems, Clinical , Deep Learning , Humans , Neuroimaging , Tomography, X-Ray Computed/methodsABSTRACT
Glioma is the most common primary intraparenchymal tumor of the brain and the 5-year survival rate of high-grade glioma is poor. Magnetic resonance imaging (MRI) is essential for detecting, characterizing and monitoring brain tumors but definitive diagnosis still relies on surgical pathology. Machine learning has been applied to the analysis of MRI data in glioma research and has the potential to change clinical practice and improve patient outcomes. This systematic review synthesizes and analyzes the current state of machine learning applications to glioma MRI data and explores the use of machine learning for systematic review automation. Various datapoints were extracted from the 153 studies that met inclusion criteria and analyzed. Natural language processing (NLP) analysis involved keyword extraction, topic modeling and document classification. Machine learning has been applied to tumor grading and diagnosis, tumor segmentation, non-invasive genomic biomarker identification, detection of progression and patient survival prediction. Model performance was generally strong (AUC = 0.87 ± 0.09; sensitivity = 0.87 ± 0.10; specificity = 0.0.86 ± 0.10; precision = 0.88 ± 0.11). Convolutional neural network, support vector machine and random forest algorithms were top performers. Deep learning document classifiers yielded acceptable performance (mean 5-fold cross-validation AUC = 0.71). Machine learning tools and data resources were synthesized and summarized to facilitate future research. Machine learning has been widely applied to the processing of MRI data in glioma research and has demonstrated substantial utility. NLP and transfer learning resources enabled the successful development of a replicable method for automating the systematic review article screening process, which has potential for shortening the time from discovery to clinical application in medicine.
Subject(s)
Artificial Intelligence/trends , Brain Neoplasms/diagnostic imaging , Glioma/diagnostic imaging , Machine Learning/trends , Neural Networks, Computer , Neuroimaging/trends , Algorithms , Brain Neoplasms/surgery , Glioma/surgery , Humans , Magnetic Resonance Imaging/methods , Magnetic Resonance Imaging/trends , Neuroimaging/methods , Neurosurgical Procedures/methods , Neurosurgical Procedures/trends , Support Vector MachineABSTRACT
BACKGROUND: Obtaining successful arthrodesis at the craniocervical junction and atlantoaxial joint can be more challenging than in other segments of the cervical spine. This challenge stems from the relatively hypermobile joints between the occipital condyles, the motion that occurs at C1 and C2, as well as the paucity of dorsal bony surfaces for posterior arthrodesis. While multiple different techniques for spinal fixation in this region have been well described, there has been little investigation into auxiliary methods to improve fusion rates. OBJECTIVE: To describe the use of an occipital bone graft to augment bony arthrodesis in the supraaxial cervical spine using a multidisciplinary approach. METHODS: We review the technique for harvesting and placing a vascularized occipital bone graft in 2 patients undergoing revision surgery at the craniocervical junction. RESULTS: The differentiation from nonvascularized bone graft, either allograft or autograft, to a bone graft using vascularized tissue is a key principle of this technique. It has been well established that vascularized bone heals and fuses in the spine better than structural autogenous grafts. However, the morbidity and added operative time of harvesting a vascularized flap, such as from the fibula or rib, precludes its utility in most degenerative spine surgeries. CONCLUSION: By adapting the standard neurosurgical procedure for a suboccipital craniectomy and utilizing the tenets of flap-based reconstructive surgery to maintain the periosteal and muscular blood supply, we describe the feasibility of using a vascularized and pedicled occipital bone graft to augment instrumented upper cervical spinal fusion. The use of this vascularized bone graft may increase fusion rates in complex spine surgeries.
Subject(s)
Atlanto-Axial Joint , Spinal Fusion , Atlanto-Axial Joint/surgery , Bone Transplantation , Cervical Vertebrae/surgery , Humans , Occipital Bone/surgeryABSTRACT
Successful arthrodesis at the craniocervical junction and atlantoaxial joint can be more challenging than in other segments of the cervical spine. Different techniques for spinal fixation in this region have been well described, along with auxiliary methods to improve fusion rates. The occipital vascularized bone graft is a novel technique that can be used to augment bony arthrodesis in the supra-axial cervical spine. It provides the benefits of a vascularized autologous graft, such as accelerated healing, earlier fusion, and increased strength. This technique can be learned with relative ease and may be particularly helpful in cases with high risk of nonunion or pseudoarthrosis in the upper cervical spine.
ABSTRACT
BACKGROUND: Deep brain stimulation (DBS) surgery is an option for patients experiencing medically resistant neurologic symptoms. DBS complications are rare; finding significant predictors requires a large number of surgeries. Machine learning algorithms may be used to effectively predict these outcomes. The aims of this study were to 1) investigate preoperative clinical risk factors and 2) build machine learning models to predict adverse outcomes. METHODS: This multicenter registry collected clinical and demographic characteristics of patients undergoing DBS surgery (n = 501) and tabulated occurrence of complications. Logistic regression was used to evaluate risk factors. Supervised learning algorithms were trained and validated on 70% and 30%, respectively, of both oversampled and original registry data. Performance was evaluated using area under the receiver operating characteristics curve (AUC), sensitivity, specificity, and accuracy. RESULTS: Logistic regression showed that the risk of complication was related to the operating institution in which the surgery was performed (odds ratio [OR] = 0.44, confidence interval [CI] = 0.25-0.78), body mass index (OR = 0.94, CI = 0.89-0.99), and diabetes (OR = 2.33, CI = 1.18-4.60). Patients with diabetes were almost 3× more likely to return to the operating room (OR = 2.78, CI = 1.31-5.88). Patients with a history of smoking were 4× more likely to experience postoperative infection (OR = 4.20, CI = 1.21-14.61). Supervised learning algorithms demonstrated high discrimination performance when predicting any complication (AUC = 0.86), a complication within 12 months (AUC = 0.91), return to the operating room (AUC = 0.88), and infection (AUC = 0.97). Age, body mass index, procedure side, gender, and a diagnosis of Parkinson disease were influential features. CONCLUSIONS: Multiple significant complication risk factors were identified, and supervised learning algorithms effectively predicted adverse outcomes in DBS surgery.
Subject(s)
Algorithms , Deep Brain Stimulation/adverse effects , Machine Learning , Postoperative Complications/diagnosis , Postoperative Complications/etiology , Aged , Deep Brain Stimulation/trends , Female , Humans , Male , Middle Aged , Predictive Value of Tests , Registries , Retrospective Studies , Risk FactorsABSTRACT
Many institutions have developed shared decision-making conferences as a mechanism for reducing treatment costs and improving patient outcomes. Little is known about the process of shared decision-making that takes place in these conferences, and there is the possibility of bias among surgeons and nonsurgeons for treatment within their respective specialties. This study was conducted to determine who is contributing to the decision-making process in a multidisciplinary spine conference and to what extent treatment biases exist among the surgical and nonsurgical members of this conference. Voting data were collected during weekly multidisciplinary spine conferences. Descriptive statistics were calculated on the cases presented and the number and type of physicians voting for each case. The likelihood of a particular vote in the surgeon and nonsurgeon cohorts was evaluated using relative risk calculation and multinomial logistic regression. A total of 262 consecutive cases were analyzed. No significant differences in treatment recommendation were observed between surgery and nonsurgical management (relative risk, 1.1; 95% CI, 0.97-1.25) when comparing votes from the surgeon and nonsurgeon cohorts. Multinomial logistic regression showed the odds of nonsurgeons recommending nonsurgical management over surgery was 20% greater than receiving that recommendation from their surgeon colleagues. Individual surgeon and nonsurgeon voters were evenly distributed above and below the mean for treatment recommendation. Individual and group biases exist among surgeons and nonsurgeons treating degenerative spine diseases. Multidisciplinary conferences may or may not level these biases, depending on how they are conducted.