Speakers

Machine learning in action: Revolutionizing intracranial hematoma
1109 09:00-09:10
AI & New Technology/304B
Objectives: Traumatic intracranial hematomas represent a critical clinical situation where early detection and management are of utmost importance. Machine learning has been recently used in the detection of neuroradiological findings. Hence, it can be used in the detection of intracranial hematomas and furtherly initiate a management cascade of patient transfer, diagnostics, admission, and emergency intervention. We aim, here, to develop a diagnostic tool based on artificial intelligence to detect hematomas instantaneously, and automatically start a cascade of actions that support the management protocol depending on the early diagnosis. Materials and Methods: A plot was designed as a staged model: The first stage of initiating and training the machine with the provisional evaluation of its accuracy and the second stage of supervised use in a tertiary care hospital and a third stage of its generalization in primary and secondary care hospitals. Two datasets were used: CQ500, a public dataset, and our dataset collected retrospectively from our tertiary hospital. Results: A mean dice score of 0.83 was achieved on the validation set of CQ500. Moreover, the detection of intracranial hemorrhage was successful in 94% of cases for the CQ500 test set and 93% for our local institute cases. Poor detection was present in only 6–7% of the total test set. Moderate false-positive results were encountered in 18% and major false positives reached 5% for the total test set. Conclusion: The proposed approach for the early detection of acute intracranial hematomas provides a reliable outset for generating an automatically initiated management cascade in high-flow hospitals. Keywords: Artificial intelligence, Deep learning, Automatic intracranial hematoma detection, Patient selection, Patient referral system.

Extent of Hyperostotic Bone Resection in Convexity Meningioma to Achieve Pathologically Free Margins
1108 TBD
Neuro-oncology/305
Objective Hyperostosis in meningiomas can be present in 4.5% to 44% of cases. Radical resection should include aggressive removal of invaded bone. It is not clear however to what extent bone removal should be carried to achieve pathologically free margins, especially that in many cases, there is a T2 hyperintense signal that extends beyond the hyperostotic bone. In this study we try to investigate the perimeter of tumour cells outside the visible nidus of hyperostotic bone and to what extent they are present outside this nidus. This would serve as an initial step for setting guidelines on dealing with hyperostosis in meningioma surgery. Methods This is a prospective case series that included 14 patients with convexity meningiomas and hyperostosis during the period from March 2017 to August 2018 in two university hospitals. Patients demographics, clinical, imaging characteristics, intraoperative and postoperative data were collected and analysed. In all cases, all visible abnormal bone was excised bearing in mind to also include the hyperintense diploe in magnetic resonance imaging (MRI) T2 weighted images after careful preoperative assessment. To examine bony tumour invasion, five marked bone biopsies were taken from the craniotomy flap for histopathological examinations. These include one from the centre of hyperostotic nidus and the other four from the corners at a 2-cm distance from the margin of the nidus. Results Our study included five males (35.7%) and nine females (64.3%) with a mean age of 43.75 years (33-55). Tumor site was parietal in seven cases (50%), fronto-parietal in three cases (21.4%), parieto-occipital in two cases (14.2%), frontal region in one case and bicoronal (midline) in one case. Tumour pathology revealed a World Health Organization (WHO) grade I in seven cases (50%), atypical meningioma (WHO II) in five cases (35.7%) and anaplastic meningioma (WHO III) in two cases (14.2%). In all grade I and II meningiomas, bone biopsies harvested from the nidus revealed infiltration with tumour cells while all other bone biopsies from the four corners (2 cm from nidus) were free. In cases of anaplastic meningiomas, all five biopsies were positive for tumour cells. Conclusion Removal of the gross epicentre of hyperostotic bone with the surrounding 2 cm is adequate to ensure radical excision and free bone margins in grade I and II meningiomas. Hyperintense signal change in MRI T2 weighted images, even beyond visible hypersototic areas, doesn’t necessarily represent tumour invasion. Keywords: Hyperostosis, Meningioma