Wenbao Wang and Linghua Kong
Jonathan Shapey, Guotai Wang, Reuben Dorent, Alexis Dimitriadis, Wenqi Li, Ian Paddick, Neil Kitchen, Sotirios Bisdas, Shakeel R. Saeed, Sebastien Ourselin, Robert Bradford, and Tom Vercauteren
Automatic segmentation of vestibular schwannomas (VSs) from MRI could significantly improve clinical workflow and assist in patient management. Accurate tumor segmentation and volumetric measurements provide the best indicators to detect subtle VS growth, but current techniques are labor intensive and dedicated software is not readily available within the clinical setting. The authors aim to develop a novel artificial intelligence (AI) framework to be embedded in the clinical routine for automatic delineation and volumetry of VS.
Imaging data (contrast-enhanced T1-weighted [ceT1] and high-resolution T2-weighted [hrT2] MR images) from all patients meeting the study’s inclusion/exclusion criteria who had a single sporadic VS treated with Gamma Knife stereotactic radiosurgery were used to create a model. The authors developed a novel AI framework based on a 2.5D convolutional neural network (CNN) to exploit the different in-plane and through-plane resolutions encountered in standard clinical imaging protocols. They used a computational attention module to enable the CNN to focus on the small VS target and propose a supervision on the attention map for more accurate segmentation. The manually segmented target tumor volume (also tested for interobserver variability) was used as the ground truth for training and evaluation of the CNN. We quantitatively measured the Dice score, average symmetric surface distance (ASSD), and relative volume error (RVE) of the automatic segmentation results in comparison to manual segmentations to assess the model’s accuracy.
Imaging data from all eligible patients (n = 243) were randomly split into 3 nonoverlapping groups for training (n = 177), hyperparameter tuning (n = 20), and testing (n = 46). Dice, ASSD, and RVE scores were measured on the testing set for the respective input data types as follows: ceT1 93.43%, 0.203 mm, 6.96%; hrT2 88.25%, 0.416 mm, 9.77%; combined ceT1/hrT2 93.68%, 0.199 mm, 7.03%. Given a margin of 5% for the Dice score, the automated method was shown to achieve statistically equivalent performance in comparison to an annotator using ceT1 images alone (p = 4e−13) and combined ceT1/hrT2 images (p = 7e−18) as inputs.
The authors developed a robust AI framework for automatically delineating and calculating VS tumor volume and have achieved excellent results, equivalent to those achieved by an independent human annotator. This promising AI technology has the potential to improve the management of patients with VS and potentially other brain tumors.
Naif M. Alotaibi, Justin Z. Wang, Christopher R. Pasarikovski, Daipayan Guha, Fawaz Al-Mufti, Muhammad Mamdani, Gustavo Saposnik, Tom A. Schweizer, and R. Loch Macdonald
Elevated intracranial pressure (ICP) is a well-recognized phenomenon in aneurysmal subarachnoid hemorrhage (aSAH) that has been demonstrated to lead to poor outcomes. Despite significant advances in clinical research into aSAH, there are no consensus guidelines devoted specifically to the management of elevated ICP in the setting of aSAH. To treat high ICP in aSAH, most centers extrapolate their treatment algorithms from studies and published guidelines for traumatic brain injury. Herein, the authors review the current management strategies for treating raised ICP within the aSAH population, emphasize key differences from the traumatic brain injury population, and highlight potential directions for future research in this controversial topic.
Radek Kolecki, Vikalpa Dammavalam, Abdullah Bin Zahid, Molly Hubbard, Osamah Choudhry, Marleen Reyes, ByoungJun Han, Tom Wang, Paraskevi Vivian Papas, Aylin Adem, Emily North, David T. Gilbertson, Douglas Kondziolka, Jason H. Huang, Paul P. Huang, and Uzma Samadani
The precise threshold differentiating normal and elevated intracranial pressure (ICP) is variable among individuals. In the context of several pathophysiological conditions, elevated ICP leads to abnormalities in global cerebral functioning and impacts the function of cranial nerves (CNs), either or both of which may contribute to ocular dysmotility. The purpose of this study was to assess the impact of elevated ICP on eye-tracking performed while patients were watching a short film clip.
Awake patients requiring placement of an ICP monitor for clinical purposes underwent eye tracking while watching a 220-second continuously playing video moving around the perimeter of a viewing monitor. Pupil position was recorded at 500 Hz and metrics associated with each eye individually and both eyes together were calculated. Linear regression with generalized estimating equations was performed to test the association of eye-tracking metrics with changes in ICP.
Eye tracking was performed at ICP levels ranging from −3 to 30 mm Hg in 23 patients (12 women, 11 men, mean age 46.8 years) on 55 separate occasions. Eye-tracking measures correlating with CN function linearly decreased with increasing ICP (p < 0.001). Measures for CN VI were most prominently affected. The area under the curve (AUC) for eye-tracking metrics to discriminate between ICP < 12 and ≥ 12 mm Hg was 0.798. To discriminate an ICP < 15 from ≥ 15 mm Hg the AUC was 0.833, and to discriminate ICP < 20 from ≥ 20 mm Hg the AUC was 0.889.
Increasingly elevated ICP was associated with increasingly abnormal eye tracking detected while patients were watching a short film clip. These results suggest that eye tracking may be used as a noninvasive, automatable means to quantitate the physiological impact of elevated ICP, which has clinical application for assessment of shunt malfunction, pseudotumor cerebri, concussion, and prevention of second-impact syndrome.