RESUMO
OBJECTIVEGlioblastoma (GBM) and primary central nervous system lymphoma (PCNSL) are common intracranial pathologies encountered by neurosurgeons. They often may have similar radiological findings, making diagnosis difficult without surgical biopsy; however, management is quite different between these two entities. Recently, predictive analytics, including machine learning (ML), have garnered attention for their potential to aid in the diagnostic assessment of a variety of pathologies. Several ML algorithms have recently been designed to differentiate GBM from PCNSL radiologically with a high sensitivity and specificity. The objective of this systematic review and meta-analysis was to evaluate the implementation of ML algorithms in differentiating GBM and PCNSL.METHODSThe authors performed a systematic review of the literature using PubMed in accordance with PRISMA guidelines to select and evaluate studies that included themes of ML and brain tumors. These studies were further narrowed down to focus on works published between January 2008 and May 2018 addressing the use of ML in training models to distinguish between GBM and PCNSL on radiological imaging. Outcomes assessed were test characteristics such as accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC).RESULTSEight studies were identified addressing use of ML in training classifiers to distinguish between GBM and PCNSL on radiological imaging. ML performed well with the lowest reported AUC being 0.878. In studies in which ML was directly compared with radiologists, ML performed better than or as well as the radiologists. However, when ML was applied to an external data set, it performed more poorly.CONCLUSIONSFew studies have applied ML to solve the problem of differentiating GBM from PCNSL using imaging alone. Of the currently published studies, ML algorithms have demonstrated promising results and certainly have the potential to aid radiologists with difficult cases, which could expedite the neurosurgical decision-making process. It is likely that ML algorithms will help to optimize neurosurgical patient outcomes as well as the cost-effectiveness of neurosurgical care if the problem of overfitting can be overcome.
Assuntos
Neoplasias Encefálicas/diagnóstico por imagem , Glioblastoma/diagnóstico por imagem , Aprendizado de Máquina , Neuroimagem/métodos , Neoplasias do Sistema Nervoso Central/diagnóstico por imagem , Diagnóstico Diferencial , Humanos , Linfoma/diagnóstico por imagem , Aprendizado de Máquina/normas , Neuroimagem/normasRESUMO
BACKGROUND: Whether magnetic resonance imaging (MRI) adds value to surgical planning for patients with acute traumatic cervical spinal cord injury (ATCSCI) remains controversial. In this study, we compared surgeons' operative planning decisions with and without preoperative MRI. We had two hypotheses: (1) the surgical plan for ATCSCI would not change substantially after the MRI and (2) intersurgeon agreement on the surgical plan would also not change substantially after the MRI. METHODS: We performed a vignette-based survey study that included a retrospective review of all adult trauma patients who presented to our American College of Surgeons-verified level 1 trauma center from 2010 to 2019 with signs of acute quadriplegia and underwent computed tomography (CT), MRI, and subsequent cervical spine surgery within 48 hours of admission. We abstracted patient demographics, admission physiology, and injury details. Patient clinical scenarios were presented to three spine surgeons, first with only the CT and then, a minimum of 2 weeks later, with both the CT and MRI. At each presentation, the surgeons identified their surgical plan, which included timing (none, <8, <24, >24 hours), approach (anterior, posterior, circumferential), and targeted vertebral levels. The outcomes were change in surgical plan and intersurgeon agreement. We used Fleiss' kappa (κ) to measure intersurgeon agreement. RESULTS: Twenty-nine patients met the criteria and were included. Ninety-three percent of the surgical plans were changed after the MRI. Intersurgeon agreement was "slight" to "fair" both before the MRI (timing, κ = 0.22; approach, κ = 0.35; levels, κ = 0.13) and after the MRI (timing, κ = 0.06; approach, κ = 0.27; levels, κ = 0.10). CONCLUSION: Surgical plans for ATCSCI changed substantially when the MRI was presented in addition to the CT; however, intersurgeon agreement regarding the surgical plan was low and not improved by the addition of the MRI. LEVEL OF EVIDENCE: Diagnostic, level II.
Assuntos
Medula Cervical/lesões , Vértebras Cervicais , Traumatismos da Medula Espinal/diagnóstico por imagem , Adolescente , Adulto , Idoso , Idoso de 80 Anos ou mais , Medula Cervical/diagnóstico por imagem , Medula Cervical/cirurgia , Vértebras Cervicais/lesões , Vértebras Cervicais/cirurgia , Feminino , Humanos , Imageamento por Ressonância Magnética , Masculino , Pessoa de Meia-Idade , Período Pré-Operatório , Estudos Retrospectivos , Traumatismos da Medula Espinal/cirurgia , Fatores de Tempo , Tomografia Computadorizada por Raios X , Adulto JovemRESUMO
Surgical resection is the current standard of therapy for the treatment of arteriovenous malformation (AVM). Endovascular embolization is commonly used as an adjunct prior to surgical resection because it is believed to reduce the risk of intraoperative bleeding. Embolization has been associated with other complications including visual deficits, vessel perforation, and catheter adhesion. Catheter adhesion in which retained segments are contained within the embolization cast are not necessarily cause for concern, but retention of larger portions may confer an increased risk of thrombus formation. Such cases warrant the removal of the retained segments or the patient may suffer serious complications including death related to cerebrovascular events. In this case report, we describe a unique case of catheter adhesion in which the extension of the feeding catheter and the length of the introducer was left in its entirety down to the entry portion at the level of the groin and later retrieved in its entirety by craniotomy.
RESUMO
INTRODUCTION: The use of intraoperative CT-guidance during the percutaneous treatment of trigeminal neuralgia has become increasingly popular due to the greater ease of foramen ovale cannulation and decreased procedure times. Concerns regarding radiation dose to the patient, however, remain unaddressed. We sought to compare the emitted radiation dose from fluoroscopy with intraoperative CT for these procedures. METHODS: A retrospective review of percutaneous lesioning procedures for trigeminal neuralgia performed between 2010 until 2012 at our institution was conducted and radiation doses to the patient were recorded. We subsequently simulated four separate percutaneous trigeminal rhizotomies using the O-arm intraoperative CT (Medtronics, Minneapolis, MN, USA) to cannulate the foramen ovale bilaterally in two formalin-fixed cadaver heads. RESULTS: Seventeen successful percutaneous treatments for trigeminal neuralgia were performed during the study period. Eleven procedures containing complete records were included in the final analysis. For procedures using fluoroscopy, the mean dosage was 15.2 mGys (range: 1.15 - 47.95, 95% CI 7.34 - 22.99). Radiation dosage from the O-arm imaging system was 16.55 mGy for all four cases. An unequal variance t-test did not reach statistical significance (p=0.42). CONCLUSIONS: We did not observe a significant difference in radiation dose delivered to subjects when comparing CT-guided foramen ovale cannulation relative to fluoroscopy for percutaneous lesioning of the Gasserian ganglion. Additional study is required under operational settings.
RESUMO
OBJECTIVE: To determine if a computer-based simulation with haptic technology can help surgical trainees improve tactile discrimination using surgical instruments. MATERIAL AND METHODS: Twenty junior medical students participated in the study and were randomized into two groups. Subjects in Group A participated in virtual simulation training using the ImmersiveTouch simulator (ImmersiveTouch, Inc., Chicago, IL, USA) that required differentiating the firmness of virtual spheres using tactile and kinesthetic sensation via haptic technology. Subjects in Group B did not undergo any training. With their visual fields obscured, subjects in both groups were then evaluated on their ability to use the suction and bipolar instruments to find six elastothane objects with areas ranging from 1.5 to 3.5 cm2 embedded in a urethane foam brain cavity model while relying on tactile and kinesthetic sensation only. RESULTS: A total of 73.3% of the subjects in Group A (simulation training) were able to find the brain cavity objects in comparison to 53.3% of the subjects in Group B (no training) (P â=â 0.0183). There was a statistically significant difference in the total number of Group A subjects able to find smaller brain cavity objects (size ≤ 2.5 cm2) compared to that in Group B (72.5 vs. 40%, P â=â 0.0032). On the other hand, no significant difference in the number of subjects able to detect larger objects (size ⧠3 cm2) was found between Groups A and B (75 vs. 80%, P â=â 0.7747). CONCLUSION: Virtual computer-based simulators with integrated haptic technology may improve tactile discrimination required for microsurgical technique.
Assuntos
Simulação por Computador , Neurocirurgia/educação , Humanos , Tato , Interface Usuário-ComputadorRESUMO
OBJECTIVE: This study explores the usefulness of virtual simulation training for learning to place pedicle screws in the lumbar spine. METHODS: Twenty-six senior medical students anonymously participated and were randomized into two groups (Aâ=âno simulation; Bâ=âsimulation). Both groups were given 15 minutes to place two pedicle screws in a sawbones model. Students in Group A underwent traditional visual/verbal instruction whereas students in Group B underwent training on pedicle screw placement in the ImmersiveTouch simulator. The students in both groups then placed two pedicle screws each in a lumbar sawbones models that underwent triplanar thin slice computerized tomography and subsequent analysis based on coronal entry point, axial and sagittal deviations, length error, and pedicle breach. The average number of errors per screw was calculated for each group. Semi-parametric regression analysis for clustered data was used with generalized estimating equations accommodating a negative binomial distribution to determine any statistical difference of significance. RESULTS: A total of 52 pedicle screws were analyzed. The reduction in the average number of errors per screw after a single session of simulation training was 53.7% (P â=â 0.0067). The average number of errors per screw in the simulation group was 0.96 versus 2.08 in the non-simulation group. The simulation group outperformed the non-simulation group in all variables measured. The three most benefited measured variables were length error (86.7%), coronal error (71.4%), and pedicle breach (66.7%). CONCLUSIONS: Computer-based simulation appears to be a valuable teaching tool for non-experts in a highly technical procedural task such as pedicle screw placement that involves sequential learning, depth perception, and understanding triplanar anatomy.
Assuntos
Simulação por Computador , Neurocirurgia/educação , Medula Espinal/cirurgia , Interface Usuário-Computador , Humanos , Análise e Desempenho de TarefasRESUMO
OBJECTIVE: To understand the perceived utility of a novel simulator to improve operative skill, eye-hand coordination, and depth perception. METHODS: We used the ImmersiveTouch simulation platform (ImmersiveTouch, Inc., Chicago, Illinois, USA) in two U.S. Accreditation Council for Graduate Medical Education-accredited neurosurgical training programs: the University of Chicago and the University of Texas Medical Branch. A total of 54 trainees participated in the study, which consisted of 14 residents (group A), 20 senior medical students who were neurosurgery candidates (group B), and 20 junior medical students (group C). The participants performed a simulation task that established bipolar hemostasis in a virtual brain cavity and provided qualitative feedback regarding perceived benefits in eye-hand coordination, depth perception, and potential to assist in improving operating skills. RESULTS: The perceived ability of the simulator to positively influence skills judged by the three groups: group A, residents; group B, senior medical students; and group C, junior medical students was, respectively, 86%, 100%, and 100% for eye-hand coordination; 86%, 100%, and 95% for depth perception; and 79%, 100%, and 100% for surgical skills in the operating room. From all groups, 96.2% found the simulation somewhat or very useful to improve eye-hand coordination, and 94% considered it beneficial to improve depth perception and operating room skills. CONCLUSION: This simulation module may be suitable for resident training, as well as for the development of career interest and skill acquisition; however, validation for this type of simulation needs to be further developed.
Assuntos
Encéfalo/fisiologia , Encéfalo/cirurgia , Competência Clínica , Técnicas Hemostáticas/instrumentação , Destreza Motora/fisiologia , Neurocirurgia/educação , Desempenho Psicomotor/fisiologia , Interface Usuário-Computador , Simulação por Computador , Eletrocoagulação , Humanos , Procedimentos Neurocirúrgicos/instrumentação , Procedimentos Neurocirúrgicos/métodosRESUMO
BACKGROUND: The effort required to introduce simulation in neurosurgery academic programs and the benefits perceived by residents have not been systematically assessed. OBJECTIVE: To create a neurosurgery simulation curriculum encompassing basic and advanced skills, cadaveric dissection, cranial and spine surgery simulation, and endovascular and computerized haptic training. METHODS: A curriculum with 68 core exercises per academic year was distributed in individualized sets of 30 simulations to 6 neurosurgery residents. The total number of procedures completed during the academic year was set to 180. The curriculum includes 79 simulations with physical models, 57 cadaver dissections, and 44 haptic/computerized sessions. Likert-type evaluations regarding self-perceived performance were completed after each exercise. Subject identification was blinded to junior (postgraduate years 1-3) or senior resident (postgraduate years 4-6). Wilcoxon rank testing was used to detect differences within and between groups. RESULTS: One hundred eighty procedures and surveys were analyzed. Junior residents reported proficiency improvements in 82% of simulations performed (P < .001). Senior residents reported improvement in 42.5% of simulations (P < .001). Cadaver simulations accrued the highest reported benefit (71.5%; P < .001), followed by physical simulators (63.8%; P < .001) and haptic/computerized (59.1; P < .001). Initial cost is $341,978.00, with $27,876.36 for annual operational expenses. CONCLUSION: The systematic implementation of a simulation curriculum in a neurosurgery training program is feasible, is favorably regarded, and has a positive impact on trainees of all levels, particularly in junior years. All simulation forms, cadaver, physical, and haptic/computerized, have a role in different stages of learning and should be considered in the development of an educational simulation program.