Focused ultrasound (FUS) has been under investigation for neurosurgical applications since the 1940s. Early experiments demonstrated ultrasound as an effective tool for the creation of intracranial lesions; however, they were limited by the need for craniotomy to avoid trajectory damage and wave distortion by the skull, and they also lacked effective techniques for monitoring. Since then, the development and hemispheric distribution of phased arrays has resolved the issue of the skull and allowed for a completely transcranial procedure. Similarly, advances in MR technology have allowed for the real-time guidance of FUS procedures using MR thermometry. MR-guided FUS (MRgFUS) has primarily been investigated for its thermal lesioning capabilities and was recently approved for use in essential tremor. In this capacity, the use of MRgFUS is being investigated for other ablative indications in functional neurosurgery and neurooncology. Other applications of MRgFUS that are under active investigation include opening of the blood-brain barrier to facilitate delivery of therapeutic agents, neuromodulation, and thrombolysis. These recent advances suggest a promising future for MRgFUS as a viable and noninvasive neurosurgical tool, with strong potential for yet-unrealized applications.
Maya Harary, David J. Segar, Kevin T. Huang, Ian J. Tafel, Pablo A. Valdes and G. Rees Cosgrove
Timothy R. Smith, M. Maher Hulou, Kevin T. Huang, Breno Nery, Samuel Miranda de Moura, David J. Cote and Edward R. Laws
The purpose of this study was to describe complications associated with the endonasal, transsphenoidal approach for the treatment of adrenocorticotropic hormone (ACTH)–positive staining tumors (Cushing's disease [CD] and silent corticotroph adenomas [SCAs]) performed by 1 surgeon at a high-volume academic medical center.
Medical records from Brigham and Women's Hospital were retrospectively reviewed. Selected for study were 82 patients with CD who during April 2008–April 2014 had consecutively undergone transsphenoidal resection or who had subsequent pathological confirmation of ACTH-positive tumor staining. In addition to demographic, patient, tumor, and surgery characteristics, complications were evaluated. Complications of interest included syndrome of inappropriate antidiuretic hormone secretion, diabetes insipidus (DI), CSF leakage, carotid artery injury, epistaxis, meningitis, and vision changes.
Of the 82 patients, 68 (82.9%) had CD and 14 (17.1%) had SCAs; 55 patients were female and 27 were male. Most common (n = 62 patients, 82.7%) were microadenomas, followed by macroadenomas (n = 13, 14.7%). A total of 31 (37.8%) patients underwent reoperation. Median follow-up time was 12.0 months (range 3–69 months). The most common diagnosis was ACTH-secreting (n = 68, 82.9%), followed by silent tumors/adenomas (n = 14, 17.1%). ACTH hyperplasia was found in 8 patients (9.8%). Of the 74 patients who had verified tumors, 12 (16.2%) had tumors with atypical features.
The overall (CD and SCA) rate of minor complications was 35.4%; the rate of major complications was 8.5% (n = 7). All permanent morbidity was associated with DI (n = 5, 6.1%). In 16 CD patients (23.5%), transient DI developed. Transient DI was more likely to develop in CD patients who had undergone a second operation (37.9%) than in those who had undergone a first operation only (12.8%, p < 0.05). Permanent DI developed in 4 CD patients (5.9%) and 1 SCA patient (7.1%). For 1 CD patient, intraoperative carotid artery injury required endovascular sacrifice of the injured artery, but the patient remained neurologically intact. For another CD patient, aseptic meningitis developed and was treated effectively with corticosteroids. One CD patient experienced major postoperative epistaxis requiring another operative procedure to achieve hemostasis. For 2 CD patients, development of sinus mucoceles was managed conservatively. For 1 SCA patient, an abdominal wound dehisced at the fat graft site. No patients experienced postoperative CSF leakage, visual impairment, or deep vein thrombosis.
Transsphenoidal surgery is the treatment of choice for patients with CD and other ACTH-positive staining tumors. Recent advances in endoscopic technology and increasing surgeon comfort with this technology are making transsphenoidal procedures safer, faster, and more effective. Serious complications are uncommon and can be managed successfully.
Kevin T. Huang, Michael A. Silva, Alfred P. See, Kyle C. Wu, Troy Gallerani, Hasan A. Zaidi, Yi Lu, John H. Chi, Michael W. Groff and Omar M. Arnaout
Recent advances in computer vision have revolutionized many aspects of society but have yet to find significant penetrance in neurosurgery. One proposed use for this technology is to aid in the identification of implanted spinal hardware. In revision operations, knowing the manufacturer and model of previously implanted fusion systems upfront can facilitate a faster and safer procedure, but this information is frequently unavailable or incomplete. The authors present one approach for the automated, high-accuracy classification of anterior cervical hardware fusion systems using computer vision.
Patient records were searched for those who underwent anterior-posterior (AP) cervical radiography following anterior cervical discectomy and fusion (ACDF) at the authors’ institution over a 10-year period (2008–2018). These images were then cropped and windowed to include just the cervical plating system. Images were then labeled with the appropriate manufacturer and system according to the operative record. A computer vision classifier was then constructed using the bag-of-visual-words technique and KAZE feature detection. Accuracy and validity were tested using an 80%/20% training/testing pseudorandom split over 100 iterations.
A total of 321 total images were isolated containing 9 different ACDF systems from 5 different companies. The correct system was identified as the top choice in 91.5% ± 3.8% of the cases and one of the top 2 or 3 choices in 97.1% ± 2.0% and 98.4 ± 13% of the cases, respectively. Performance persisted despite the inclusion of variable sizes of hardware (i.e., 1-level, 2-level, and 3-level plates). Stratification by the size of hardware did not improve performance.
A computer vision algorithm was trained to classify at least 9 different types of anterior cervical fusion systems using relatively sparse data sets and was demonstrated to perform with high accuracy. This represents one of many potential clinical applications of machine learning and computer vision in neurosurgical practice.