RESUMO
OBJECTIVE: In clinical spine surgery research, manually reviewing surgical forms to categorize patients by their surgical characteristics is a crucial yet time-consuming task. Natural language processing (NLP) is a machine learning tool used to adaptively parse and categorize important features from text. These systems function by training on a large, labeled data set in which feature importance is learned prior to encountering a previously unseen data set. The authors aimed to design an NLP classifier for surgical information that can review consent forms and automatically classify patients by the surgical procedure performed. METHODS: Thirteen thousand two hundred sixty-eight patients who underwent 15,227 surgeries from January 1, 2012, to December 31, 2022, at a single institution were initially considered for inclusion. From these surgeries, 12,239 consent forms were classified based on the Current Procedural Terminology (CPT) code, categorizing them into 7 of the most frequently performed spine surgeries at this institution. This labeled data set was split 80%/20% into train and test subsets, respectively. The NLP classifier was then trained and the results demonstrated its performance on the test data set using CPT codes to determine accuracy. RESULTS: This NLP surgical classifier had an overall weighted accuracy rate of 91% for sorting consents into correct surgical categories. Anterior cervical discectomy and fusion had the highest positive predictive value (PPV; 96.8%), whereas lumbar microdiscectomy had the lowest PPV in the testing data (85.0%). Sensitivity was highest for lumbar laminectomy and fusion (96.7%) and lowest for the least common operation, cervical posterior foraminotomy (58.3%). Negative predictive value and specificity were > 95% for all surgical categories. CONCLUSIONS: Utilizing NLP for text classification drastically improves the efficiency of classifying surgical procedures for research purposes. The ability to quickly classify surgical data can be significantly beneficial to institutions without a large database or substantial data review capabilities, as well as for trainees to track surgical experience, or practicing surgeons to evaluate and analyze their surgical volume. Additionally, the capability to quickly and accurately recognize the type of surgery will facilitate the extraction of new insights from the correlations between surgical interventions and patient outcomes. As the database of surgical information grows from this institution and others in spine surgery, the accuracy, usability, and applications of this model will continue to increase.
Assuntos
Termos de Consentimento , Processamento de Linguagem Natural , Humanos , Aprendizado de Máquina , Laminectomia , DiscotomiaRESUMO
OBJECTIVE: The US-based Spinal Laminectomy versus Instrumented Pedicle Screw (SLIP) trial reported improvement in disability following laminectomy with fusion versus laminectomy alone for patients with lumbar spondylolisthesis. Despite using similar methods, a concurrent Swedish trial investigating the same question did not reach the same conclusion. The authors performed a simulation-based analysis to elucidate potential causes of these divergent results. METHODS: The mean and standard deviation of the preoperative and 2-year postoperative Oswestry Disability Index (ODI) scores for each study group (laminectomy with fusion and laminectomy alone) were collected from the spondylolisthesis stratum of the Swedish trial and used to create a MATLAB simulator using linear transformations to predict postoperative ODI distributions. Applying this simulator to both varied and published preoperative ODI distributions from the SLIP trial, the authors simulated the results of the US-based trial using treatment effects from the Swedish study and compared simulated US results to those published in the SLIP trial. RESULTS: Simulated US results showed that as preoperative disability increased, the difference in postoperative ODI scores grew between treatment groups and increasingly favored laminectomy alone (p < 0.0001). In 100 simulations of a similarly sized US trial, the average mean change in ODI scores postoperatively was significantly higher than was published for laminectomy alone in the SLIP trial (-21.3 vs -17.9), whereas it was significantly lower than published for fusion (-16.9 vs -26.3). CONCLUSIONS: The expected benefit of surgical treatments for spondylolisthesis varied according to preoperative disability. Adapting Swedish-estimated treatment effects to the US context mildly overapproximated the improvement in postoperative disability scores for laminectomy, but more severely underapproximated the improvement reported for laminectomy and fusion in the SLIP trial. The observed heterogeneity between these studies is influenced more by patient response to fusion than response to laminectomy. This analysis paves the way for future studies on the impact of preoperative treatment group heterogeneity, differences in surgical methods, and empirical design on reported clinical benefits. Although bayesian reanalysis of published randomized controlled trial data is susceptible to biases that typically limit post hoc analyses, the authors' method offers a simple and cost-effective approach to improve the understanding of published clinical trial results and their implications for future studies.
RESUMO
BACKGROUND: Despite the expanding role of machine learning (ML) in health care and patient expectations for clinicians to understand ML-based tools, few for-credit curricula exist specifically for neurosurgical trainees to learn basic principles and implications of ML for medical research and clinical practice. We implemented a novel, remotely delivered curriculum designed to develop literacy in ML for neurosurgical trainees. METHODS: A 4-week pilot medical elective was designed specifically for trainees to build literacy in basic ML concepts. Qualitative feedback from interested and enrolled students was collected to assess students' and trainees' reactions, learning, and future application of course content. RESULTS: Despite 15 interested learners, only 3 medical students and 1 neurosurgical resident completed the course. Enrollment included students and trainees from 3 different institutions. All learners who completed the course found the lectures relevant to their future practice as clinicians and researchers and reported improved confidence in applying and understanding published literature applying ML techniques in health care. Barriers to ample enrollment and retention (e.g., balancing clinical responsibilities) were identified. CONCLUSIONS: This pilot elective demonstrated the interest, value, and feasibility of a remote elective to establish ML literacy; however, feedback to increase accessibility and flexibility of the course encouraged our team to implement changes. Future elective iterations will have a semiannual, 2-week format, splitting lectures more clearly between theory (the method and its value) and application (coding instructions) and will make lectures open-source prerequisites to allow tailoring of student learning to their planned application of these methods in their practice and research.
Assuntos
Educação de Graduação em Medicina , Estudantes de Medicina , Humanos , Currículo , Atenção à Saúde , Educação de Graduação em Medicina/métodos , RetroalimentaçãoRESUMO
INTRODUCTION: In a multilevel cervical laminoplasty operation for patients with cervical spondylotic myelopathy (CSM), a partial or complete C3 laminectomy may be performed at the upper level instead of a C3 plated laminoplasty. It is unknown whether C3 technique above the laminoplasty affects loss of cervical lordosis or range of motion. METHODS: Patients undergoing multilevel laminoplasty of the cervical spine (C3-C6/C7) at a single institution were retrospectively reviewed. Patients were divided into two cohorts based on surgical technique at C3: C3-C6/C7 plated laminoplasty ("C3 laminoplasty only", N = 61), C3 partial or complete laminectomy, plus C4-C6/C7 plated laminoplasty (N = 39). All patients had at least 1-year postoperative X-ray treatment. RESULTS: Of 100 total patients, C3 laminoplasty and C3 laminectomy were equivalent in all demographic data, except for age (66.4 vs. 59.4 years, p = 0.012). None of the preoperative radiographic parameters differed between the C3 laminoplasty and C3 laminectomy cohorts: cervical lordosis (13.1° vs. 11.1°, p = 0.259), T1 slope (32.9° vs. 29.2°, p = 0.072), T1 slope-cervical lordosis (19.8° vs. 18.6°, p = 0.485), or cervical sagittal vertical axis (3.1 cm vs. 2.7 cm, p = 0.193). None of the postoperative radiographic parameters differed between the C3 laminoplasty and C3 laminectomy cohorts: cervical lordosis (9.4° vs. 11.2°, p = 0.369), T1 slope-cervical lordosis (21.7° vs. 18.1°, p = 0.126), to cervical sagittal vertical axis (3.3 cm vs. 3.6 cm, p = 0.479). In the total cohort, 31% had loss of cervical lordosis >5°. Loss of lordosis reached 5-10° (mild change) in 13% of patients and >10° (moderate change) in 18% of patients. C3 laminoplasty and C3 laminectomy cohorts did not differ with respect to no change (<5°: 65.6% vs. 74.3%, respectively), mild change (5-10°: 14.8% vs. 10.3%), and moderate change (>10°: 19.7% vs. 15.4%) in cervical lordosis, p = 0.644. When controlling for age, ordinal regression showed that surgical technique at C3 did not increase the odds of postoperative loss of cervical lordosis. C3 laminectomy versus C3 laminoplasty did not differ in the postoperative range of motion on cervical flexion-extension X-rays (23.9° vs. 21.7°, p = 0.451, N = 91). CONCLUSION: There was no difference in postoperative loss of cervical lordosis or postoperative range of motion in patients who underwent either C3-C6/C7 plated laminoplasty or C3 laminectomy plus C4-C6/C7 plated laminoplasty.
RESUMO
(1) Background: Intracortical microelectrodes (IMEs) are essential to basic brain research and clinical brain-machine interfacing applications. However, the foreign body response to IMEs results in chronic inflammation and an increase in levels of reactive oxygen and nitrogen species (ROS/RNS). The current study builds on our previous work, by testing a new delivery method of a promising antioxidant as a means of extending intracortical microelectrodes performance. While resveratrol has shown efficacy in improving tissue response, chronic delivery has proven difficult because of its low solubility in water and low bioavailability due to extensive first pass metabolism. (2) Methods: Investigation of an intraventricular delivery of resveratrol in rats was performed herein to circumvent bioavailability hurdles of resveratrol delivery to the brain. (3) Results: Intraventricular delivery of resveratrol in rats delivered resveratrol to the electrode interface. However, intraventricular delivery did not have a significant impact on electrophysiological recordings over the six-week study. Histological findings indicated that rats receiving intraventricular delivery of resveratrol had a decrease of oxidative stress, yet other biomarkers of inflammation were found to be not significantly different from control groups. However, investigation of the bioavailability of resveratrol indicated a decrease in resveratrol accumulation in the brain with time coupled with inconsistent drug elution from the cannulas. Further inspection showed that there may be tissue or cellular debris clogging the cannulas, resulting in variable elution, which may have impacted the results of the study. (4) Conclusions: These results indicate that the intraventricular delivery approach described herein needs further optimization, or may not be well suited for this application.
RESUMO
Long-term reliability of intracortical microelectrodes remains a challenge for increased acceptance and deployment. There are conflicting reports comparing measurements associated with recording quality with postmortem histology, in attempts to better understand failure of intracortical microelectrodes (IMEs). Our group has recently introduced the assessment of motor behavior tasks as another metric to evaluate the effects of IME implantation. We hypothesized that adding the third dimension to our analysis, functional behavior testing, could provide substantial insight on the health of the tissue, success of surgery/implantation, and the long-term performance of the implanted device. Here we present our novel analysis scheme including: (1) the use of numerical formal concept analysis (nFCA) and (2) a regression analysis utilizing modern model/variable selection. The analyses found complimentary relationships between the variables. The histological variables for glial cell activation had associations between each other, as well as the neuronal density around the electrode interface. The neuronal density had associations to the electrophysiological recordings and some of the motor behavior metrics analyzed. The novel analyses presented herein describe a valuable tool that can be utilized to assess and understand relationships between diverse variables being investigated. These models can be applied to a wide range of ongoing investigations utilizing various devices and therapeutics.
RESUMO
Progress has been made in the field of neural interfacing using both mouse and rat models, yet standardization of these models' interchangeability has yet to be established. The mouse model allows for transgenic, optogenetic, and advanced imaging modalities which can be used to examine the biological impact and failure mechanisms associated with the neural implant itself. The ability to directly compare electrophysiological data between mouse and rat models is crucial for the development and assessment of neural interfaces. The most obvious difference in the two rodent models is size, which raises concern for the role of device-induced tissue strain. Strain exerted on brain tissue by implanted microelectrode arrays is hypothesized to affect long-term recording performance. Therefore, understanding any potential differences in tissue strain caused by differences in the implant to tissue size ratio is crucial for validating the interchangeability of rat and mouse models. Hence, this study is aimed at investigating the electrophysiological variances and predictive device-induced tissue strain. Rat and mouse electrophysiological recordings were collected from implanted animals for eight weeks. A finite element model was utilized to assess the tissue strain from implanted intracortical microelectrodes, taking into account the differences in the depth within the cortex, implantation depth, and electrode geometry between the two models. The rat model demonstrated a larger percentage of channels recording single unit activity and number of units recorded per channel at acute but not chronic time points, relative to the mouse model Additionally, the finite element models also revealed no predictive differences in tissue strain between the two rodent models. Collectively our results show that these two models are comparable after taking into consideration some recommendations to maintain uniform conditions for future studies where direct comparisons of electrophysiological and tissue strain data between the two animal models will be required.
RESUMO
Intracortical microelectrodes (IME) are neural devices that initially were designed to function as neuroscience tools to enable researchers to understand the nervous system. Over the years, technology that aids interfacing with the nervous system has allowed the ability to treat patients with a wide range of neurological injuries and diseases. Despite the substantial success that has been demonstrated using IME in neural interface applications, these implants eventually fail due to loss of quality recording signals. Recent strategies to improve interfacing with the nervous system have been inspired by methods that mimic the native tissue. This review focusses on one strategy in particular, nano-architecture, a term we introduce that encompasses the approach of roughening the surface of the implant. Various nano-architecture approaches have been hypothesized to improve the biocompatibility of IMEs, enhance the recording quality, and increase the longevity of the implant. This review will begin by introducing IME technology and discuss the challenges facing the clinical deployment of IME technology. The biological inspiration of nano-architecture approaches will be explained as well as leading fabrication methods used to create nano-architecture and their limitations. A review of the effects of nano-architecture surfaces on neural cells will be examined, depicting the various cellular responses to these modified surfaces in both in vitro and pre-clinical models. The proposed mechanism elucidating the ability of nano-architectures to influence cellular phenotype will be considered. Finally, the frontiers of next generation nano-architecture IMEs will be identified, with perspective given on the future impact of this interfacing approach.
RESUMO
Clinical implantation of intracortical microelectrodes has been hindered, at least in part, by the perpetual inflammatory response occurring after device implantation. The neuroinflammatory response observed after device implantation has been correlated to oxidative stress that occurs due to neurological injury and disease. However, there has yet to be a definitive link of oxidative stress to intracortical microelectrode implantation. Thus, the objective of this study is to give direct evidence of oxidative stress following intracortical microelectrode implantation. This study also aims to identify potential molecular targets to attenuate oxidative stress observed postimplantation. Here, we implanted adult rats with silicon non-functional microelectrode probes for 4 weeks and compared the oxidative stress response to no surgery controls through postmortem gene expression analysis and qualitative histological observation of oxidative stress markers. Gene expression analysis results at 4 weeks postimplantation indicated that EH domain-containing 2, prion protein gene (Prnp), and Stearoyl-Coenzyme A desaturase 1 (Scd1) were all significantly higher for animals implanted with intracortical microelectrode probes compared to no surgery control animals. To the contrary, NADPH oxidase activator 1 (Noxa1) relative gene expression was significantly lower for implanted animals compared to no surgery control animals. Histological observation of oxidative stress showed an increased expression of oxidized proteins, lipids, and nucleic acids concentrated around the implant site. Collectively, our results reveal there is a presence of oxidative stress following intracortical microelectrode implantation compared to no surgery controls. Further investigation targeting these specific oxidative stress linked genes could be beneficial to understanding potential mechanisms and downstream therapeutics that can be utilized to reduce oxidative stress-mediated damage following microelectrode implantation.
RESUMO
Published reports of status epilepticus due to intraperitoneal injection containing propylene glycol in rats are sparse. In fact, there are no reports specifying a maximum safe dose of propylene glycol through intraperitoneal administration. We report here a case of unexpected seizures in Sprague Dawley rats after receiving an intraperitoneal injection containing propylene glycol. Nine-week-old, 225-250 gram male rats were reported to experience tremor progressing to seizures within minutes after given injections of resveratrol (30 mg/kg) dissolved in a 40 : 60 propylene glycol/corn oil vehicle solution by direct intraperitoneal (IP) slow bolus injection or via a preplaced intraperitoneal catheter. The World Health Organization suggests a maximum dose of 25 mg/kg/day of propylene glycol taken orally and no more than 25 mg/dL in blood serum, whereas the animals used in our study got a calculated maximum 0.52 g/kg (25 times lower dose). Blood tests from the seizing rat support a diagnosis of hemolysis and lactic acidosis which may have led to the seizures, all of which appeared to be a consequence of the propylene glycol administration. These findings are consistent with oral and intravenous administration of propylene glycol toxicity as previously reported in other species, including humans. To our knowledge, this report represents the first published case of status epilepticus due to an IP injection containing propylene glycol.