Investigation of probability maps in deep-learning-based brain ventricle parcellation.

Normal Pressure Hydrocephalus (NPH) is a brain disorder associated with ventriculomegaly. Accurate segmentation of the ventricle system into its sub-compartments from magnetic resonance images (MRIs) could help evaluate NPH patients for surgical intervention. In this paper, we modify a 3D U-net utilizing probability maps to perform accurate ventricle parcellation, even with grossly enlarged ventricles and post-surgery shunt artifacts, from MRIs. Our method achieves a mean dice similarity coefficient (DSC) on whole ventricles for healthy controls of 0.864 ± 0.047 and 0.961 ± 0.024 for NPH patients. Furthermore, with the benefit of probability maps, the proposed method provides superior performance on MRI with grossly enlarged ventricles (mean DSC value of 0.965 ± 0.027) or post-surgery shunt artifacts (mean DSC value of 0.964 ± 0.031). Results indicate that our method provides a high robust parcellation tool on the ventricular systems which is comparable to other state-of-the-art methods.

AUTOMATED VENTRICLE PARCELLATION AND EVAN'S RATIO COMPUTATION IN PRE- AND POST-SURGICAL VENTRICULOMEGALY.

Normal pressure hydrocephalus (NPH) is a brain disorder associated with enlarged ventricles and multiple cognitive and motor symptoms. The degree of ventricular enlargement can be measured using magnetic resonance images (MRIs) and characterized quantitatively using the Evan's ratio (ER). Automatic computation of ER is desired to avoid the extra time and variations associated with manual measurements on MRI. Because shunt surgery is often used to treat NPH, it is necessary that this process be robust to image artifacts caused by the shunt and related implants. In this paper, we propose a 3D regions-of-interest aware (ROI-aware) network for segmenting the ventricles. The method achieves state-of-the-art performance on both pre-surgery MRIs and post-surgery MRIs with artifacts. Based on our segmentation results, we also describe an automated approach to compute ER from these results. Experimental results on multiple datasets demonstrate the potential of the proposed method to assist clinicians in the diagnosis and management of NPH.

A joint ventricle and WMH segmentation from MRI for evaluation of healthy and pathological changes in the aging brain.

Age-related changes in brain structure include atrophy of the brain parenchyma and white matter changes of presumed vascular origin. Enlargement of the ventricles may occur due to atrophy or impaired cerebrospinal fluid (CSF) circulation. The co-occurrence of these changes in neurodegenerative diseases and in aging brains often requires investigators to take both into account when studying the brain, however, automated segmentation of enlarged ventricles and white matter hyperintensities (WMHs) can be a challenging task. Here, we present a hybrid multi-atlas segmentation and convolutional autoencoder approach for joint ventricle parcellation and WMH segmentation from magnetic resonance images (MRIs). Our fully automated approach uses a convolutional autoencoder to generate a standardized image of grey matter, white matter, CSF, and WMHs, which, in conjunction with labels generated by a multi-atlas segmentation approach, is then fed into a convolutional neural network to parcellate the ventricular system. Hence, our approach does not depend on manually delineated training data for new data sets. The segmentation pipeline was validated on both healthy elderly subjects and subjects with normal pressure hydrocephalus using ground truth manual labels and compared with state-of-the-art segmentation methods. We then applied the method to a cohort of 2401 elderly brains to investigate associations of ventricle volume and WMH load with various demographics and clinical biomarkers, using a multiple regression model. Our results indicate that the ventricle volume and WMH load are both highly variable in a cohort of elderly subjects and there is an independent association between the two, which highlights the importance of taking both the possibility of enlarged ventricles and WMHs into account when studying the aging brain.

The Role of Preoperative Magnetic Resonance Imaging in Assessing Neurovascular Compression Before Microvascular Decompression in Trigeminal Neuralgia.

BACKGROUND: Preoperative magnetic resonance imaging (MRI) is a standard component of the preoperative clinical workup for patients before microvascular decompression (MVD). However, its ability to accurately exclude neurovascular compression of the trigeminal nerve is not well understood. METHODS: We retrospectively reviewed 1020 patients with available preoperative MRI data before microvascular decompression. General patient demographics and clinical characteristics were collected for each case. We recorded both evidence of neurovascular conflict on preoperative MRI radiology notes and intraoperative compression from operative notes. Sensitivity, specificity, positive predictive value, and negative predictive value were determined for general MRI, high-resolution MRI, and non-high resolution. RESULTS: Overall, preoperative MRI before MVD showed a sensitivity of 75.8%, specificity of 65.8%, positive predictive value of 92.4%, and negative predictive value of 33.3% in predicting neurovascular compression of the trigeminal nerve. In particular, MRI was unable to identify 21.0% cases of sole arterial compression, 42.5% cases of sole venous compression, and combined arterial and venous compression in 18.5% of cases. A total of 958 patients (93.9%) underwent high-resolution preoperative MRI with skull base sequences. This imaging showed a sensitivity of 75.6%, specificity of 66.9%, positive predictive value of 92.5% and a negative predictive value of 33.4% in predicting trigeminal nerve neurovascular compression. Non-high-resolution MRI showed a sensitivity of 78.8%, specificity of 50.0%, positive predictive value of 89.1%, and negative predictive value of 31.3%. The negative predictive values of general, high-resolution, and non-high-resolution MRIs were all <50%. CONCLUSIONS: Preoperative MRI may offer a high predictive value for neurovascular conflict and should be part of the standard preoperative care workup for patients with trigeminal neuralgia. However, lack of neurovascular conflict on preoperative imaging is not sufficient to exclude patients from undergoing MVD.

High resolution 3D magnetic resonance imaging of Gruber's ligament: a pilot study.

INTRODUCTION: Gruber's ligament (GL), a surgical landmark, extends from the lateral upper clivus to the petrous apex (PA), forming the superior boundary of Dorello's canal (DC). It overlies the interdural segment of the abducens nerve (CN VI). High-resolution 3D skull base MRI (SB-MRI) demonstrates anatomic details visible to the surgeon, but not well seen on traditional cross-sectional imaging. The aim of this study was to demonstrate visualization of the GL and its relationship to CN VI utilizing contrast enhanced high-resolution SB-MRI. METHODS: Two neuroradiologists retrospectively reviewed in consensus the SB-MRIs of 27 skull base sides, among 14 patients. GL detection rate, confidence of detection, and GL length were recorded. When GL was successfully identified, the position of the interdural segment of CN VI within DC was recorded. RESULTS: GL was readily identified in 16 skull base sides (59%), identified with some difficulty in 2 skull base sides (7%), and failed to be identified in 9 skull base sides (33%). The mean GL length was 7.1 mm (4.5-9.3 mm). Among the 18 cases where GL was successfully identified, CN VI was readily identified in all cases (100%), coursing the lateral third of DC in 72% of sides, and middle third in the remaining 28% of sides. CONCLUSION: GL can be identified in approximately two-thirds of cases utilizing 3D high resolution SB-MRI. CN VI passes most commonly along the lateral third of DC. This is the first report demonstrating visualization of GL and its relation to CN VI, on imaging.

Systematic volumetric analysis predicts response to CSF drainage and outcome to shunt surgery in idiopathic normal pressure hydrocephalus.

OBJECTIVES: Idiopathic normal pressure hydrocephalus (INPH) is a neurodegenerative disorder characterized by excess cerebrospinal fluid (CSF) in the ventricles, which can be diagnosed by invasive CSF drainage test and treated by shunt placement. Here, we aim to investigate the diagnostic and prognostic power of systematic volumetric analysis based on brain structural MRI for INPH. METHODS: We performed a retrospective study with a cohort of 104 probable INPH patients who underwent CSF drainage tests and another cohort of 41 INPH patients who had shunt placement. High-resolution T1-weighted images of the patients were segmented using an automated pipeline into 283 structures that are grouped into different granularity levels for volumetric analysis. Volumes at multi-granularity levels were used in a recursive feature elimination model to classify CSF drainage responders and non-responders. We then used pre-surgical brain volumes to predict Tinetti and MMSE scores after shunting, based on the least absolute shrinkage and selection operator. RESULTS: The classification accuracy of differentiating the CSF drainage responders and non-responders increased as the granularity increased. The highest diagnostic accuracy was achieved at the finest segmentation with a sensitivity/specificity/precision/accuracy of 0.89/0.91/0.84/0.90 and an area under the curve of 0.94. The predicted post-surgical neurological scores showed high correlations with the ground truth, with r = 0.80 for Tinetti and r = 0.88 for MMSE. The anatomical features that played important roles in the diagnostic and prognostic tasks were also illustrated. CONCLUSIONS: We demonstrated that volumetric analysis with fine segmentation could reliably differentiate CSF drainage responders from other INPH-like patients, and it could accurately predict the neurological outcomes after shunting. KEY POINTS: • We performed a fully automated segmentation of brain MRI at multiple granularity levels for systematic volumetric analysis of idiopathic normal pressure hydrocephalus (INPH) patients. • We were able to differentiate patients that responded to CSF drainage test with an accuracy of 0.90 and area under the curve of 0.94 in a cohort of 104 probable INPH patients, as well as to predict the post-shunt gait and cognitive scores with a coefficient of 0.80 for Tinetti and 0.88 for MMSE. • Feature analysis showed the inferior lateral ventricle, bilateral hippocampus, and orbital cortex are positive indicators of CSF drainage responders, whereas the posterior deep white matter and parietal subcortical white matter were negative predictors.

Segmental Imaging of the Trochlear Nerve: Anatomic and Pathologic Considerations.

BACKGROUND: The trochlear nerve (the fourth cranial nerve) is the only cranial nerve that arises from the dorsal aspect of the midbrain. The nerve has a lengthy course making it highly susceptible to injury. It is also the smallest cranial nerve and is often difficult to identify on neuroimaging. EVIDENCE ACQUISITION: High-resolution 3-dimensional skull base MRI allows for submillimeter isotropic acquisition and is optimal for cranial nerve evaluation. In this text, the detailed anatomy of the fourth cranial nerve applicable to imaging will be reviewed. RESULTS: Detailed anatomic knowledge of each segment of the trochlear nerve is necessary in patients with trochlear nerve palsy. A systematic approach to identification and assessment of each trochlear nerve segment is essential. Pathologic cases are provided for each segment. CONCLUSIONS: A segmental approach to high-resolution 3-dimensional MRI for the study of the trochlear nerve is suggested.

Applications of a deep learning method for anti-aliasing and super-resolution in MRI.

Magnetic resonance (MR) images with both high resolutions and high signal-to-noise ratios (SNRs) are desired in many clinical and research applications. However, acquiring such images takes a long time, which is both costly and susceptible to motion artifacts. Acquiring MR images with good in-plane resolution and poor through-plane resolution is a common strategy that saves imaging time, preserves SNR, and provides one viewpoint with good resolution in two directions. Unfortunately, this strategy also creates orthogonal viewpoints that have poor resolution in one direction and, for 2D MR acquisition protocols, also creates aliasing artifacts. A deep learning approach called SMORE that carries out both anti-aliasing and super-resolution on these types of acquisitions using no external atlas or exemplars has been previously reported but not extensively validated. This paper reviews the SMORE algorithm and then demonstrates its performance in four applications with the goal to demonstrate its potential for use in both research and clinical scenarios. It is first shown to improve the visualization of brain white matter lesions in FLAIR images acquired from multiple sclerosis patients. Then it is shown to improve the visualization of scarring in cardiac left ventricular remodeling after myocardial infarction. Third, its performance on multi-view images of the tongue is demonstrated and finally it is shown to improve performance in parcellation of the brain ventricular system. Both visual and selected quantitative metrics of resolution enhancement are demonstrated.

Brain ventricle parcellation using a deep neural network: Application to patients with ventriculomegaly.

Numerous brain disorders are associated with ventriculomegaly, including both neuro-degenerative diseases and cerebrospinal fluid disorders. Detailed evaluation of the ventricular system is important for these conditions to help understand the pathogenesis of ventricular enlargement and elucidate novel patterns of ventriculomegaly that can be associated with different diseases. One such disease is normal pressure hydrocephalus (NPH), a chronic form of hydrocephalus in older adults that causes dementia. Automatic parcellation of the ventricular system into its sub-compartments in patients with ventriculomegaly is quite challenging due to the large variation of the ventricle shape and size. Conventional brain labeling methods are time-consuming and often fail to identify the boundaries of the enlarged ventricles. We propose a modified 3D U-Net method to perform accurate ventricular parcellation, even with grossly enlarged ventricles, from magnetic resonance images (MRIs). We validated our method on a data set of healthy controls as well as a cohort of 95 patients with NPH with mild to severe ventriculomegaly and compared with several state-of-the-art segmentation methods. On the healthy data set, the proposed network achieved mean Dice similarity coefficient (DSC) of 0.895 ±â€¯0.03 for the ventricular system. On the NPH data set, we achieved mean DSC of 0.973 ±â€¯0.02, which is significantly (p < 0.005) higher than four state-of-the-art segmentation methods we compared with. Furthermore, the typical processing time on CPU-base implementation of the proposed method is 2 min, which is much lower than the several hours required by the other methods. Results indicate that our method provides: 1) highly robust parcellation of the ventricular system that is comparable in accuracy to state-of-the-art methods on healthy controls; 2) greater robustness and significantly more accurate results on cases of ventricular enlargement; and 3) a tool that enables computation of novel imaging biomarkers for dilated ventricular spaces that characterize the ventricular system.

Aqueductal Cerebrospinal Fluid Stroke Volume Flow in a Rodent Model of Chronic Communicating Hydrocephalus: Establishing a Homogeneous Study Population for Cerebrospinal Fluid Dynamics Exploration.

BACKGROUND: Idiopathic normal pressure hydrocephalus (iNPH) is a cause of dementia that can be reversed when treated timely with cerebrospinal fluid (CSF) diversion. Understanding CSF dynamics throughout the development of hydrocephalus is crucial to identify prognostic markers to estimate benefit/risk to shunts. OBJECTIVE: To explore the cerebral aqueduct CSF flow dynamics with phase-contrast magnetic resonance imaging (MRI) in a novel rodent model of adult chronic communicating hydrocephalus. METHODS: Kaolin was injected into the subarachnoid space at the convexities in Sprague-Dawley adult rats. 11.7-T Bruker MRI was used to acquire T2-weighted images for anatomic identification and phase-contrast MRI at the cerebral aqueduct. Aqueductal stroke volume (ASV) results were compared with the ventricular volume (VV) at 15, 60, 90, and 120 days. RESULTS: Significant ventricular enlargement was found in kaolin-injected animals at all times (P < 0.001). ASV differed between cases and controls/shams at every time point (P = 0.004, 0.001, 0.001, and <0.001 at 15, 60, 90, and 120 days, respectively). After correlation between the ASV and the VV, there was a significant correlation at 15 (P = 0.015), 60 (P = 0.001), 90 (P < 0.001), and 120 days. Moreover, there was a significant positive correlation between the VV expansion and the aqueductal CSF stroke between 15 and 60 days. CONCLUSIONS: An initial active phase of rapid ventricular enlargement shows a strong correlation between the expansion of the VV and the increment in the ASV during the first 60 days, followed by a second phase with less ventricular enlargement and heterogeneous behavior in the ASV. Further correlation with complementary data from intracranial pressure and histologic/microstructural brain parenchyma assessments are needed to better understand the ASV variations after 60 days.

First spine surgery utilizing real-time image-guided robotic assistance.

Robotics in spinal surgery has significant potential benefits for both surgeons and patients, including reduced surgeon fatigue, improved screw accuracy, decreased radiation exposure, greater options for minimally invasive surgery, and less time required to train residents on techniques that can have steep learning curves. However, previous robotic systems have several drawbacks, which are addressed by the innovative ExcelsiusGPSTM robotic system. The robot is secured to the operating room floor, not the patient. It has a rigid external arm that facilitates direct transpedicular drilling and screw placement, without requiring K-wires. In addition, the ExcelsisuGPSTM has integrated neuronavigation, not present in other systems. It also has surveillance marker that immediately alerts the surgeon in the event of loss of registration, and a lateral force meter to alert the surgeon in the event of skiving. Here, we present the first spinal surgery performed with the assistance of this newly approved robot. The surgery was performed with excellent screw placement, minimal radiation exposure to the patient and surgeon, and the patient had a favorable outcome. We report the first operative case with the ExcelsisuGPSTM, and the first spine surgery utilizing real-time image-guided robotic assistance.

Founder of modern hydrocephalus diagnosis and therapy: Walter Dandy at the Johns Hopkins Hospital.

Because of his exceptional and pioneering contributions to the understanding and treatment of neurosurgical conditions, Walter Dandy is considered to be one of the founders of both neurosurgery and neuroradiology. In the field of hydrocephalus, Dandy developed revolutionary research models, imaging modalities, and operative procedures. His laboratory and clinical experiences at the Johns Hopkins Hospital, including the surgical treatment of hydrocephalus, are well illustrated in the publications he authored. Archival materials housed at the Johns Hopkins University School of Medicine provide a window into Dandy's clinical experience and supplement the work published during his lifetime. His operative experience with hydrocephalus spanned 1915-1946 and comprised 381 surgeries. From this clinical experience, Dandy created much of the framework for modern diagnostic imaging and treatment of hydrocephalus.

Ventricular Volume Dynamics During the Development of Adult Chronic Communicating Hydrocephalus in a Rodent Model.

INTRODUCTION: The pathophysiology of normal-pressure hydrocephalus and the correlation with its symptomatology is not well understood. OBJECTIVE: To monitor and evaluate the enlargement patterns of the ventricular system for each ventricle and its correlation with the presenting symptoms. METHODS: Bilateral kaolin injection into the subarachnoid space overlying the cranial convexities was done in 18 adult rats. Magnetic resonance imaging was performed on an 11.7-T scanner 15, 60, 90, and 120 days after injection. Volumes of the ventricular system were measured for each ventricle and correlated with biweekly behavioral findings. RESULTS: There was a progressive increase in the ventricular volume for the lateral ventricles since day 15 in the kaolin-injected animals. There was a nonsignificant trend in volume growth for the third ventricle, but its enlargement was synchronous with the lateral ventricles. No significant change for the fourth ventricle. No symptoms were detected in the first 60 days. Association was found between the ventricular volume and locomotor changes. In addition, the odds of locomotor symptoms increased by 3% for every additional cubic millimeter of volume in the left (P < 0.001) and right (P = 0.023) ventricles, and for the total magnetic resonance imaging volume by 1% (P = 0.013). CONCLUSIONS: Expansion of the lateral ventricles maintained similar proportions over time, accompanied by a synchronous third ventricular expansion with less proportion and a nonsignificant fourth enlargement. Lateral ventricles enlarged most in those animals that were to develop late locomotor deterioration. Further research using this animal model combined with different radiologic imaging techniques, such as diffusion tensor imaging and perfusion studies, is recommended.

Endonasal endoscopic resection of olfactory neuroblastoma: an 11-year experience.

OBJECTIVE: Olfactory neuroblastoma (ONB) is a rare malignant neoplasm of the sinonasal cavity. Surgery has been and remains a mainstay of treatment for patients with this tumor. Open craniofacial resections have been the treatment of choice for many decades. More recently, experience has been growing with endoscopic approaches in the management of patients with ONB. The object of this study is to report the authors' experience over the past 11 years with ONB patients treated with purely endonasal endoscopic techniques. METHODS: The authors performed a retrospective chart review of 20 consecutive patients with ONB who underwent a completely endonasal endoscopic approach for an oncological tumor resection at their institution between January 2006 and January 2017. Patient demographics, tumor stage, pathological grade, frozen section analysis, permanent margin assessment, perioperative complications, postoperative therapy, length of follow-up, and outcomes at last follow-up were collected and analyzed. RESULTS: Eighteen patients presented with newly diagnosed disease, with a modified Kadish stage of A in 2 cases, B in 3, C in 11, and D in 2. Two patients presented with recurrent tumors. An average of 25.3 specimens per patient were examined by frozen section analysis. Although analysis of intraoperative frozen section margins was negative in all but 1 case, microscopic foci of tumor were found in 7 cases (35%) on permanent histopathological analysis. Perioperative complications occurred in 7 patients (35%) including 1 patient who developed a cerebrospinal fluid leak; there were no episodes of meningitis. All but 1 patient received postoperative radiotherapy, and 5 patients received postoperative chemotherapy. With a mean follow-up of over 5 years, 19 patients were alive and 1 patient died from an unrelated cause. There were 2 cases of tumor recurrence. The 5-year overall, disease-specific, and recurrence-free survival rates were 92.9%, 100%, and 92.9%, respectively. CONCLUSIONS: The current results provide additional evidence for the continued use of endoscopic procedures in the management of this malignancy.

Evaluation of an in vivo model for ventricular shunt infection: a pilot study using a novel antimicrobial-loaded polymer.

OBJECTIVE: Ventricular shunt infection remains an issue leading to high patient morbidity and cost, warranting further investigation. The authors sought to create an animal model of shunt infection that could be used to evaluate possible catheter modifications and innovations. METHODS: Three dogs underwent bilateral ventricular catheter implantation and inoculation with methicillin-sensitive Staphylococcus aureus (S. aureus). In 2 experimental animals, the catheters were modified with a polymer containing chemical "pockets" loaded with vancomycin. In 1 control animal, the catheters were polymer coated but without antibiotics. Animals were monitored for 9 to 11 days, after which the shunts were explanted. MRI was performed after shunt implantation and prior to catheter harvest. The catheters were sonicated prior to microbiological culture and also evaluated by electron microscopy. The animals' brains were evaluated for histopathology. RESULTS: All animals underwent successful catheter implantation. The animals developed superficial wound infections, but no neurological deficits. Imaging demonstrated ventriculitis and cerebral edema. Harvested catheters from the control animal demonstrated > 104 colony-forming units (CFUs) of S. aureus. In the first experimental animal, one shunt demonstrated > 104 CFUs of S. aureus, but the other demonstrated no growth. In the second experimental animal, one catheter demonstrated no growth, and the other grew trace S. aureus. Brain histopathology revealed acute inflammation and ventriculitis in all animals, which was more severe in the control. CONCLUSIONS: The authors evaluated an animal model of ventricular shunting and reliably induced features of shunt infection that could be microbiologically quantified. With this model, investigation of pathophysiological and imaging correlates of infection and potentially beneficial shunt catheter modifications is possible.

Twenty-Five Diagnoses on Midline Images of the Brain: From Fetus to Child to Adult.

Midsagittal images of the brain provide a wealth of anatomic information and may show abnormalities that are pathognomonic for particular diagnoses. Using an anatomy-based approach, the authors identify pertinent anatomic structures to serve as a checklist when evaluating these structures. Subregions evaluated include the corpus callosum, pituitary gland and sellar region, pineal gland and pineal region, brainstem, and cerebellum. The authors present 25 conditions with characteristic identifiable abnormalities at midsagittal imaging. Midsagittal views from multiple imaging modalities are shown, including computed tomography, ultrasonography, and magnetic resonance (MR) imaging. Standard MR imaging sequences are shown, as well as fetal MR and sagittal diffusion-weighted images. To demonstrate these conditions, fetal, neonatal, childhood, adolescent, and young adulthood images are reviewed. The differentiation of normal variants is guided by the understanding of anatomy and pathology. When a specific diagnosis is not possible, the authors present information to evaluate differential considerations and discuss when follow-up imaging may be indicated. The authors hope each case will clarify a pertinent differential diagnosis, appropriately guide patient management, and improve understanding of normal anatomy and identification of pathologic entities. It is in these hopes that the authors have presented a checklist of pertinent anatomy and pathologic entities that can build on existing search patterns. Improved confidence and accuracy in the evaluation of midsagittal images will benefit physicians and patients. ©RSNA, 2018.

Pedicle screw accuracy assessment in ExcelsiusGPS® robotic spine surgery: evaluation of deviation from pre-planned trajectory.

BACKGROUND: The ExcelsiusGPS® (Globus Medical, Inc., Audubon, PA) is a next-generation spine surgery robotic system recently approved for use in the United States. The objective of the current study is to assess pedicle screw accuracy and clinical outcomes among two of the first operative cases utilizing the ExcelsiusGPS® robotic system and describe a novel metric to quantify screw deviation. METHODS: Two patients who underwent lumbar fusion at a single institution with the ExcelsiusGPS® surgical robot were included. Pre-operative trajectory planning was performed from an intra-operative CT scan using the O-arm (Medtronic, Inc., Minneapolis, MN). After robotic-assisted screw implantation, a post-operative CT scan was obtained to confirm ideal screw placement and accuracy with the planned trajectory. A novel pedicle screw accuracy algorithm was devised to measure screw tip/tail deviation distance and angular offset on axial and sagittal planes. Screw accuracy was concurrently determined by a blinded neuroradiologist using the traditional Gertzbein-Robbins method. Clinical variables such as symptomatology, operative data, and post-operative follow-up were also collected. RESULTS: Eight pedicle screws were placed in two L4-L5 fusion cases. Mean screw tip deviation was 2.1 mm (range 0.8-5.2 mm), mean tail deviation was 3.2 mm (range 0.9-5.4 mm), and mean angular offset was 2.4 degrees (range 0.7-3.8 degrees). All eight screws were accurately placed based on the Gertzbein-Robbins scale (88% Grade A and 12% Grade B). There were no cases of screw revision or new post-operative deficit. Both patients experienced improvement in Frankel grade and Karnofsky Performance Status (KPS) score by 6 weeks post-op. CONCLUSION: The ExcelsiusGPS® robot allows for precise execution of an intended pre-planned trajectory and accurate screw placement in the first patients to undergo robotic-assisted fusion with this technology.

Shortcomings of Ventricle Segmentation Using Deep Convolutional Networks.

Normal Pressure Hydrocephalus (NPH) is a brain disorder that can present with ventriculomegaly and dementia-like symptoms, which often can be reversed through surgery. Having accurate segmentation of the ventricular system into its sub-compartments from magnetic resonance images (MRI) would be beneficial to better characterize the condition of NPH patients. Previous segmentation algorithms need long processing time and often fail to accurately segment severely enlarged ventricles in NPH patients. Recently, deep convolutional neural network (CNN) methods have been reported to have fast and accurate performance on medical image segmentation tasks. In this paper, we present a 3D U-net CNN-based network to segment the ventricular system in MRI. We trained three networks on different data sets and compared their performances. The networks trained on healthy controls (HC) failed in patients with NPH pathology, even in patients with normal appearing ventricles. The network trained on images from HC and NPH patients provided superior performance against state-of-the-art methods when evaluated on images from both data sets.