A comparison of performance between a deep learning model with residents for localization and classification of intracranial hemorrhage.

Intracranial hemorrhage (ICH) from traumatic brain injury (TBI) requires prompt radiological investigation and recognition by physicians. Computed tomography (CT) scanning is the investigation of choice for TBI and has become increasingly utilized under the shortage of trained radiology personnel. It is anticipated that deep learning models will be a promising solution for the generation of timely and accurate radiology reports. Our study examines the diagnostic performance of a deep learning model and compares the performance of that with detection, localization and classification of traumatic ICHs involving radiology, emergency medicine, and neurosurgery residents. Our results demonstrate that the high level of accuracy achieved by the deep learning model, (0.89), outperforms the residents with regard to sensitivity (0.82) but still lacks behind in specificity (0.90). Overall, our study suggests that the deep learning model may serve as a potential screening tool aiding the interpretation of head CT scans among traumatic brain injury patients.

Benefits of Combined MRI Sequences in Meningioma Consistency Prediction: A Prospective Study of 287 Consecutive Patients.

Objective Consistency of meningiomas is one of the most important factors affecting the completeness of removal and major risks of meningioma surgery. This study used preoperative magnetic resonance imaging (MRI) sequences in single and in combination to predict meningioma consistency. Methods The prospective study included 287 intracranial meningiomas operated on by five attending neurosurgeons at Chiang Mai University Hospital from July 2012 through June 2020. The intraoperative consistency was categorized in four grades according to the method of surgical removal and intensity of ultrasonic aspirator, then correlated with preoperative tumor signal intensity pattern on MRI including T1-weighted image, T2-weighted image (T2WI), fluid-attenuated inversion recovery (FLAIR), and diffusion-weighted image (DWI), which were described as hypointensity, isointensity, and hyperintensity signals which were blindly interpreted by one neuroradiologist. Results Among 287 patients, 29 were male and 258 female. The ages ranged from 22 to 83 years. A total of 189 tumors were situated in the supratentorial space and 98 were in the middle fossa and infratentorial locations. Note that 125 tumors were found to be of soft consistency (grades 1, 2) and 162 tumors of hard consistency (grades 3, 4). Hyperintensity signals on T2WI, FLAIR, and DWI were significantly associated with soft consistency of meningiomas (relative risk [RR] 2.02, 95% confidence interval [CI] 1.35-3.03, p = 0.001, RR 2.19, 95% CI 1.43-3.35, p < 0.001, and RR 1.47, 95% CI 1.02-2.11, p = 0.037, respectively). Further, chance to be soft consistency significantly increased when two and three hyperintensity signals were combined (RR 2.75, 95% CI 1.62-4.65, p ≤ 0.001, RR 2.79, 95% CI 1.58-4.93, p < 0.001, respectively). Hypointensity signals on T2WI, FLAIR, and DWI were significantly associated with hard consistency of meningiomas (RR 1.82, 95% CI 1.18-2.81, p = 0.007, RR 1.80, 95% CI 1.15-2.83, p = 0.010, RR 1.67, 95% CI 1.07-2.59, p = 0.023, respectively) and chance to be hard consistency significantly increased when three hypointensity signals were combined (RR 1.82, 95% CI 1.11-2.97, p = 0.017). Conclusion T2WI, FLAIR, and DWI hyperintensity signals of the meningiomas was solely significantly associated with soft consistency and predictive value significantly increased when two and three hyperintensity signals were combined. Each of hypointensity signals on T2WI, FLAIR, and DWI was significantly associated with hard consistency of tumors and tendency to be hard consistency significantly increased when hypointensity was found in all three sequences.

Atypical Posterior Reversible Encephalopathy Syndrome in Intraorbital Lymphoma after Tumor Biopsy: An Illustrative Case.

Introduction A 63-year-old male presented with visual loss and left eye proptosis. Magnetic resonance imaging revealed a left orbital tumor, measuring 1.4 cm × 0.9 cm. The patient underwent left frontotemporal craniotomy to perform a biopsy of the tumor. During the postoperative period, the patient developed the first episode of a generalized tonic-clonic seizure. Case Summary Computed tomography of the brain showed hypodensity of the bilateral basal ganglia and thalami with associated edematous white matter hypodensity of bilateral temporo-occipital lobes compatible with atypical posterior reversible encephalopathy syndrome (PRES). The patient received antiepileptic medication and was observed for clinical seizure. One week later, computed tomography of the brain showed the reversible process of PRES. The pathology report revealed diffuse large B cell lymphoma. Following pathological diagnosis, the patient received treatment with whole-brain radiotherapy. Conclusion This is the first reported case of atypical PRES associated with orbital lymphoma following craniotomy for the purpose of tumor biopsy. Early detection as well as seizure and blood pressure control, is essential for the proper treatment of PRES.

Fornix Integrity Is Differently Associated With Cognition in Healthy Aging and Non-amnestic Mild Cognitive Impairment: A Pilot Diffusion Tensor Imaging Study in Thai Older Adults.

Damage to the fornix leads to significant memory impairment and executive dysfunction and is associated with dementia risk. We sought to identify if fornix integrity and fiber length are disrupted in mild cognitive impairment (MCI) and how they associate with cognition. Data from 14 healthy older adult controls (HCs) and 17 subjects with non-amnestic MCI (n-aMCI) were analyzed. Diffusion tensor imaging (DTI) at 1.5 Tesla MRI was performed to enable manual tracing of the fornix and calculation of DTI parameters. Higher fractional anisotropy of body and column of the fornix was associated with better executive functioning and memory, more strongly in the HC than in the n-aMCI group. Fornix fiber tract length (FTL) was associated with better executive function, more strongly in the n-aMCI than in the HC group, and with better memory, more strongly in the HC than in the n-aMCI group. These results highlight a decline in the contributions of the fornix to cognition in n-aMCI and suggest that maintenance of fornix FTL is essential for sustaining executive functioning in people with n-aMCI.

Spinal arteriovenous shunts: accuracy of shunt detection, localization, and subtype discrimination using spinal magnetic resonance angiography and manual contrast injection using a syringe.

OBJECT: The object of this study was to evaluate the accuracy of fast 3D contrast-enhanced spinal MR angiography (MRA) using a manual syringe contrast injection technique for detecting and evaluating spinal arteriovenous shunts (AVSs). METHODS: This was a retrospective study of 15 patients and 20 spinal MRA and catheter angiography studies. The accuracy of using spinal MRA to detect spinal AVS, localize shunts, and discriminate the subtype and dominant arterial feeder of the AVS were studied. RESULTS: There were 14 pretherapeutic and 6 posttherapeutic follow-up spinal MRA and catheter spinal angiography studies. The spinal AVS was demonstrated in 17 of 20 studies. Spinal MRA demonstrated 100% sensitivity for detecting spinal AVS with no false-negative results. A 97% accuracy rate for AVS subtype discrimination and shunt level localization was achieved using this study's diagnostic criteria. The detection of the dominant arterial feeder was limited to 9 of these 17 cases (53%). CONCLUSIONS: The fast 3D contrast-enhanced MRA technique performed using manual syringe contrast injection can detect the presence of a spinal AVS, locate the shunt level, and discriminate AVS subtype in most cases, but is limited when detecting small arterial feeders.