Skull Base-related Lesions at Routine Head CT from the Emergency Department: Pearls, Pitfalls, and Lessons Learned

Potential strength and weakness of artificial intelligence integration in emergency radiology: a review of diagnostic utilizations and applications in patient care optimization

Artificial intelligence (AI) and its recent increasing healthcare integration has created both new opportunities and challenges in the practice of radiology and medical imaging. Recent advancements in AI technology have allowed for more workplace efficiency, higher diagnostic accuracy, and overall improvements in patient care. Limitations of AI such as data imbalances, the unclear nature of AI algorithms, and the challenges in detecting certain diseases make it difficult for its widespread adoption. This review article presents cases involving the use of AI models to diagnose intracranial hemorrhage, spinal fractures, and rib fractures, while discussing how certain factors like, type, location, size, presence of artifacts, calcification, and post-surgical changes, affect AI model performance and accuracy. While the use of artificial intelligence has the potential to improve the practice of emergency radiology, it is important to address its limitations to maximize its advantages while ensuring the safety of patients overall.

Image Quality Assessment of a Deep Learning-Based Automatic Bone Removal Algorithm for Cervical CTA

BACKGROUND

The present study aims to evaluate the postprocessing image quality of a deep-learning (DL)-based automatic bone removal algorithm in the real clinical practice for cervical computed tomography angiography (CTA).

MATERIALS AND METHODS

A total of 100 patients (31 females, 61.4 ± 12.4 years old) who had performed cervical CTA from January 2022 to July 2022 were included retrospectively. Three different types of scanners were used. Ipsilateral cervical artery was divided into 10 segments. The performance of the DL algorithm and conventional algorithm in terms of bone removal and vascular integrity was independently evaluated by two radiologists for each segment. The difference in the performance between the two algorithms was compared. Inter- and intrarater consistency were assessed, and the correlation between the degree of carotid artery stenosis and the rank of bone removal and vascular integrity was analyzed.

RESULTS

Significant differences were observed in the rankings of bone removal and vascular integrity between the two algorithms on most segments on both sides. Compared to DL algorithm, the conventional algorithm showed a higher correlation between the degree of carotid artery stenosis and vascular integrity ( r = -0.264 vs r = -0.180). The inter- and intrarater consistency of DL algorithm were found to be higher than or equal to those of conventional algorithm.

CONCLUSIONS

The DL algorithm for bone removal in cervical CTA demonstrated significantly better performance than conventional postprocessing method, particularly in the segments with complex anatomical structures and adjacent to bone.

Impact of a deep learning-based brain CT interpretation algorithm on clinical decision-making for intracranial hemorrhage in the emergency department

Intracranial hemorrhage is a critical emergency that requires prompt and accurate diagnosis in the emergency department (ED). Deep learning technology can assist in interpreting non-enhanced brain CT scans, but its real-world impact on clinical decision-making is uncertain. This study assessed a deep learning-based intracranial hemorrhage detection algorithm (DLHD) in a simulated clinical environment with ten emergency medical professionals from a tertiary hospital's ED. The participants reviewed CT scans with clinical information in two steps: without and with DLHD. Diagnostic performance was measured, including sensitivity, specificity, accuracy, and the area under the receiver operating characteristic curve. Consistency in clinical decision-making was evaluated using the kappa statistic. The results demonstrated that DLHD minimally affected experienced participants' diagnostic performance and decision-making. In contrast, inexperienced participants exhibited significantly increased sensitivity (59.33-72.67%, p < 0.001) and decreased specificity (65.49-53.73%, p < 0.001) with the algorithm. Clinical decision-making consistency was moderate among inexperienced professionals (k = 0.425) and higher among experienced ones (k = 0.738). Inexperienced participants changed their decisions more frequently, mainly due to the algorithm's false positives. The study highlights the need for thorough evaluation and careful integration of deep learning tools into clinical workflows, especially for less experienced professionals.

Prognostic Value of Optic Nerve Sheath Diameters after Acute Ischemic Stroke According to Slice Thickness on Computed Tomography

The optic nerve sheath diameter (ONSD) can predict intracranial pressure and outcomes in neurological disease, but it remains unclear whether a small ONSD can be accurately measured on routine CT images with a slice thickness of approximately 4-5 mm. We measured the ONSD and ONSD/eyeball transverse diameter (ETD) ratio on routine-slice (4 mm) and thin-slice (0.6-0.75 mm) brain CT images from initial scans of acute ischemic stroke (AIS) patients. ONSD-related variables, National Institutes of Health Stroke Scale (NIHSS) scores, and age were compared between good (modified Rankin Scale [mRS] ≤ 2) and poor (mRS > 2) outcomes at discharge. Among 155 patients, 38 had poor outcomes. The thin-slice ONSD was different between outcome groups (p = 0.047), while the routine-slice ONSD showed no difference. The area under the curve (AUC) values for the ONSD and ONSD/ETD were 0.58 (95% CI, 0.49-0.66) and 0.58 (95% CI, 0.50-0.66) on the routine-slice CT, and 0.60 (95% CI, 0.52-0.68) and 0.62 (95% CI, 0.54-0.69) on the thin-slice CT. The thin-slice ONSD/ETD ratio correlated with initial NIHSS scores (r = 0.225, p = 0.005). After adjusting for NIHSS scores and age, ONSD-related variables were not associated with outcomes, and adding them to a model with NIHSS scores and age did not improve performance (all p-values > 0.05). Although ONSD measurements were not an independent outcome predictor, they correlated with stroke severity, and the thin-slice ONSD provided a slightly better prognostic performance than the routine-slice ONSD.

Checklist: Neck computed tomography in non-traumatic emergencies

Patients attending the emergency department (ED) with cervical inflammatory/infectious symptoms or presenting masses that may involve the aerodigestive tract or vascular structures require a contrast-enhanced computed tomography (CT) scan of the neck. Its radiological interpretation is hampered by the anatomical complexity and pathophysiological interrelationship between the different component systems in a relatively small area. Recent studies propose a systematic evaluation of the cervical structures, using a 7-item checklist, to correctly identify the pathology and detect incidental findings that may interfere with patient management. As a conclusion, the aim of this paper is to review CT findings in non-traumatic pathology of the neck in the ED, highlighting the importance of a systematic approach in its interpretation and synthesis of a structured, complete, and concise radiological report.

Diagnostic Accuracy and Performance Analysis of a Scanner-Integrated Artificial Intelligence Model for the Detection of Intracranial Hemorrhages in a Traumatology Emergency Department

Intracranial hemorrhages require an immediate diagnosis to optimize patient management and outcomes, and CT is the modality of choice in the emergency setting. We aimed to evaluate the performance of the first scanner-integrated artificial intelligence algorithm to detect brain hemorrhages in a routine clinical setting. This retrospective study includes 435 consecutive non-contrast head CT scans. Automatic brain hemorrhage detection was calculated as a separate reconstruction job in all cases. The radiological report (RR) was always conducted by a radiology resident and finalized by a senior radiologist. Additionally, a team of two radiologists reviewed the datasets retrospectively, taking additional information like the clinical record, course, and final diagnosis into account. This consensus reading served as a reference. Statistics were carried out for diagnostic accuracy. Brain hemorrhage detection was executed successfully in 432/435 (99%) of patient cases. The AI algorithm and reference standard were consistent in 392 (90.7%) cases. One false-negative case was identified within the 52 positive cases. However, 39 positive detections turned out to be false positives. The diagnostic performance was calculated as a sensitivity of 98.1%, specificity of 89.7%, positive predictive value of 56.7%, and negative predictive value (NPV) of 99.7%. The execution of scanner-integrated AI detection of brain hemorrhages is feasible and robust. The diagnostic accuracy has a high specificity and a very high negative predictive value and sensitivity. However, many false-positive findings resulted in a relatively moderate positive predictive value.

A Rare Case of Benign Long-Standing Ecchordosis Physaliphora

Ecchordosis physaliphora (EP) is a rare benign lesion arising from embryonic notochordal remnants, typically located in the retroclival region. This case report presents a 46-year-old male patient experiencing intermittent headaches and occipital pain. Imaging revealed a well-defined, smoothly corticated bony lesion on the left side of the clivus, accompanied by a characteristic bony stalk devoid of any aggressive features. A review of the patient's medical records indicated stable imaging findings of the lesion over six years. Clinicians and radiologists should be familiar with EP as a benign entity and differentiate it from aggressive pathologies.

Breaking Barriers in Cranioplasty: 3D Printing in Low and Middle-Income Settings-Insights from Zenica, Bosnia and Herzegovina

Background and Objectives: Cranial defects pose significant challenges in low and middle-income countries (LIMCs), necessitating innovative and cost-effective craniofacial reconstruction strategies. The purpose of this study was to present the Bosnia and Herzegovina model, showcasing the potential of a multidisciplinary team and 3D-based technologies, particularly PMMA implants, to address cranial defects in a resource-limited setting. Materials and Methods: An observational, non-experimental prospective investigation involved three cases of cranioplasty at the Department of Neurosurgery, Cantonal Hospital Zenica, Bosnia and Herzegovina, between 2019 and 2023. The technical process included 3D imaging and modeling with MIMICS software (version 10.01), 3D printing of the prototype, mold construction and intraoperative modification for precise implant fitting. Results: The Bosnia and Herzegovina model demonstrated successful outcomes in cranioplasty, with PMMA implants proving cost-effective and efficient in addressing cranial defects. Intraoperative modification contributed to reduced costs and potential complications, while the multidisciplinary approach and 3D-based technologies facilitated accurate reconstruction. Conclusions: The Bosnia and Herzegovina model showcases a cost-effective and efficient approach for craniofacial reconstruction in LIMICs. Collaborative efforts, 3D-based technologies, and PMMA implants contribute to successful outcomes. Further research is needed to validate sustained benefits and enhance craniofacial reconstruction strategies in resource-constrained settings.

Artificial Intelligence Model Trained with Sparse Data to Detect Facial and Cranial Bone Fractures from Head CT

The presence of cranial and facial bone fractures is an important finding on non-enhanced head computed tomography (CT) scans from patients who have sustained head trauma. Some prior studies have proposed automatic cranial fracture detections, but studies on facial fractures are lacking. We propose a deep learning system to automatically detect both cranial and facial bone fractures. Our system incorporated models consisting of YOLOv4 for one-stage fracture detection and improved ResUNet (ResUNet++) for the segmentation of cranial and facial bones. The results from the two models mapped together provided the location of the fracture and the name of the fractured bone as the final output. The training data for the detection model were the soft tissue algorithm images from a total of 1,447 head CT studies (a total of 16,985 images), and the training data for the segmentation model included 1,538 selected head CT images. The trained models were tested on a test dataset consisting of 192 head CT studies (a total of 5,890 images). The overall performance achieved a sensitivity of 88.66%, a precision of 94.51%, and an F1 score of 0.9149. Specifically, the cranial and facial regions were evaluated and resulted in a sensitivity of 84.78% and 80.77%, a precision of 92.86% and 87.50%, and F1 scores of 0.8864 and 0.8400, respectively. The average accuracy for the segmentation labels concerning all predicted fracture bounding boxes was 80.90%. Our deep learning system could accurately detect cranial and facial bone fractures and identify the fractured bone region simultaneously.

A Deep Learning Approach for Automated Bone Removal from Computed Tomography Angiography of the Brain

Advanced visualization techniques such as maximum intensity projection (MIP) and volume rendering (VR) are useful for evaluating neurovascular anatomy on CT angiography (CTA) of the brain; however, interference from surrounding osseous anatomy is common. Existing methods for removing bone from CTA images are limited in scope and/or accuracy, particularly at the skull base. We present a new brain CTA bone removal tool, which addresses many of these limitations. A deep convolutional neural network was designed and trained for bone removal using 72 brain CTAs. The model was tested on 15 CTAs from the same data source and 17 CTAs from an independent external dataset. Bone removal accuracy was assessed quantitatively, by comparing automated segmentation results to manual segmentations, and qualitatively by evaluating VR visualization of the carotid siphons compared to an existing method for automated bone removal. Average Dice overlap between automated and manual segmentations from the internal and external test datasets were 0.986 and 0.979 respectively. This was superior compared to a publicly available noncontrast head CT bone removal algorithm which had a Dice overlap of 0.947 (internal dataset) and 0.938 (external dataset). Our algorithm yielded better VR visualization of the carotid siphons than the publicly available bone removal tool in 14 out of 15 CTAs (93%, chi-square statistic of 22.5, p-value of < 0.00001) from the internal test dataset and 15 out of 17 CTAs (88%, chi-square statistic of 23.1, p-value of < 0.00001) from the external test dataset. Bone removal allowed subjectively superior MIP and VR visualization of vascular anatomy/pathology. The proposed brain CTA bone removal algorithm is rapid, accurate, and allows superior visualization of vascular anatomy and pathology compared to other available techniques and was validated on an independent external dataset.

Morphological Measurements of Innominate Foramina and Bony Spurs along the Base of Sphenoid as a Potential Risk Factor for Neurovascular Entrapment, Radiological Interpretation and Surgical Access

Background

Restricted access and compression of neurovascular structures at various anatomic variations at the skull base poses a challenge to surgeons, neurologists and anesthetists. The present study was performed with the objective of providing morphometric analysis of innominate foramina, and anomalous bony bars and spurs along the infratemporal surface of the greater wing of the sphenoid and reviewing the practical significance of dealing with this region.

Methods

A total of 100 dry-aged human adult skulls from the archives of the osteology library of the Department of Anatomy were studied. A detailed morphometric analysis of such innominate foramina and anomalous osseous structures along the base of the sphenoid was performed using a sliding digital vernier caliper.

Results

Anomalous bony bar was found in 22 skulls (25.28%). A complete bar was observed at eight (9.1%). An innominate foramen was located inferomedial to foramen ovale (5 unilateral and 3 bilateral) with a mean anteroposterior diameter of 3.44 mm and a mean transverse diameter of 3.16 mm.

Conclusion

Neurovascular structures may be compressed by abnormal bony outgrowths or while traversing through such unnamed bony foramina. The latter may also be overlooked and mistaken during radiological interpretation leading to delayed diagnosis. Such unnamed foramina and bony outgrowths need to be documented in the literature due to their surgical, and radiological implications and limited citations.

Intracranial Hemorrhage Detection Using Parallel Deep Convolutional Models and Boosting Mechanism

Intracranial hemorrhage (ICH) can lead to death or disability, which requires immediate action from radiologists. Due to the heavy workload, less experienced staff, and the complexity of subtle hemorrhages, a more intelligent and automated system is necessary to detect ICH. In literature, many artificial-intelligence-based methods are proposed. However, they are less accurate for ICH detection and subtype classification. Therefore, in this paper, we present a new methodology to improve the detection and subtype classification of ICH based on two parallel paths and a boosting technique. The first path employs the architecture of ResNet101-V2 to extract potential features from windowed slices, whereas Inception-V4 captures significant spatial information in the second path. Afterwards, the detection and subtype classification of ICH is performed by the light gradient boosting machine (LGBM) using the outputs of ResNet101-V2 and Inception-V4. Thus, the combined solution, known as ResNet101-V2, Inception-V4, and LGBM (Res-Inc-LGBM), is trained and tested over the brain computed tomography (CT) scans of CQ500 and Radiological Society of North America (RSNA) datasets. The experimental results state that the proposed solution efficiently obtains 97.7% accuracy, 96.5% sensitivity, and 97.4% F1 score using the RSNA dataset. Moreover, the proposed Res-Inc-LGBM outperforms the standard benchmarks for the detection and subtype classification of ICH regarding the accuracy, sensitivity, and F1 score. The results prove the significance of the proposed solution for its real-time application.

Can Optic Nerve Sheath Images on a Thin-Slice Brain Computed Tomography Reconstruction Predict the Neurological Outcomes in Cardiac Arrest Survivors?

We analyzed the prognostic performance of optic nerve sheath diameter (ONSD) on thin-slice (0.6 mm) brain computed tomography (CT) reconstruction images as compared to routine-slice (4 mm) images. We conducted a retrospective analysis of brain CT images taken within 2 h after cardiac arrest. The maximal ONSD (mONSD) and optic nerve sheath area (ONSA) were measured on thin-slice images, and the routine ONSD (rONSD) and gray-to-white matter ratio (GWR) were measured on routine-slice images. We analyzed their area under the receiver operator characteristic curve (AUC) and the cutoff values for predicting a poor 6-month neurological outcome (a cerebral performance category score of 3–5). Of the 159 patients analyzed, 113 patients had a poor outcome. There was no significant difference in rONSD between the outcome groups (p = 0.116). Compared to rONSD, mONSD (AUC 0.62, 95% CI: 0.54–0.70) and the ONSA (AUC 0.63, 95% CI: 0.55–0.70) showed better prognostic performance and had higher sensitivities to determine a poor outcome (mONSD, 20.4% [95% CI, 13.4–29.0]; ONSA, 16.8% [95% CI, 10.4–25.0]; rONSD, 7.1% [95% CI, 3.1–13.5]), with specificity of 95.7% (95% CI, 85.2–99.5). A combined cutoff value obtained by both the mONSD and GWR improved the sensitivity (31.0% [95% CI, 22.6–40.4]) of determining a poor outcome, while maintaining a high specificity. In conclusion, rONSD was clinically irrelevant, but the mONSD had an increased sensitivity in cutoff having acceptable specificity. Combination of the mONSD and GWR had an improved prognostic performance in these patients.

Checklists to reduce diagnostic error: a systematic review of the literature using a human factors framework

OBJECTIVES

To apply a human factors framework to understand whether checklists reduce clinical diagnostic error have (1) gaps in composition; and (2) components that may be more likely to reduce errors.

DESIGN

Systematic review.

DATA SOURCES

PubMed, EMBASE, Scopus and Web of Science were searched through 15 February 2022.

ELIGIBILITY CRITERIA

Any article that included a clinical checklist aimed at improving the diagnostic process. Checklists were defined as any structured guide intended to elicit additional thinking regarding diagnosis.

DATA EXTRACTION AND SYNTHESIS

Two authors independently reviewed and selected articles based on eligibility criteria. Each extracted unique checklist was independently characterised according to the well-established human factors framework: Systems Engineering Initiative for Patient Safety 2.0 (SEIPS 2.0). If reported, checklist efficacy in reducing diagnostic error (eg, diagnostic accuracy, number of errors or any patient-related outcomes) was outlined. Risk of study bias was independently evaluated using standardised quality assessment tools in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

RESULTS

A total of 30 articles containing 25 unique checklists were included. Checklists were characterised within the SEIPS 2.0 framework as follows: Work Systems subcomponents of Tasks (n=13), Persons (n=2) and Internal Environment (n=3); Processes subcomponents of Cognitive (n=20) and Social and Behavioural (n=2); and Outcomes subcomponents of Professional (n=2). Other subcomponents, such as External Environment or Patient outcomes, were not addressed. Fourteen checklists examined effect on diagnostic outcomes: seven demonstrated improvement, six were without improvement and one demonstrated mixed results. Importantly, Tasks-oriented studies more often demonstrated error reduction (n=5/7) than those addressing the Cognitive process (n=4/10).

CONCLUSIONS

Most diagnostic checklists incorporated few human factors components. Checklists addressing the SEIPS 2.0 Tasks subcomponent were more often associated with a reduction in diagnostic errors. Studies examining less explored subcomponents and emphasis on Tasks, rather than the Cognitive subcomponents, may be warranted to prevent diagnostic errors.

Radiological Diagnosis of a Rare Prepontine Lesion: Ecchordosis Physaliphora

Ecchordosis physaliphora (EP) is a notochordal remnant tissue rarely encountered during routine clinical practice. These lesions usually do not produce any significant symptoms as they are slow-growing and mostly small in size. Symptoms are due to mass effects on adjacent structures when they are large or extra-tumoral hemorrhage. Because of histological similarity with chordoma, diagnosis is challenging, and this differentiation is essential as the disease course and treatment differ significantly. Imaging plays a crucial role in identifying and distinguishing these lesions.

We report the case of a 16-year-old male who presented with intermittent headache and neck pain for six months. His routine clinical examinations were within normal limits. On neurological assessment, there was no focal neurodeficit. Evaluation of cranial nerves did not reveal any evidence of palsy. Routine hematological tests were also normal. A computed tomography (CT) scan of the brain revealed a mass in front of the pons. Magnetic resonance imaging (MRI) for further evaluation revealed a T1 hypointense and T2/fluid-attenuated inversion recovery hyperintense lesion in the pre-pontine cistern. There was no enhancement in the mass either in the post-contrast CT or MRI scans. There was no bony erosion and clivus was normal. Based on the location and characteristic imaging features, a diagnosis of EP was made.

There may be several other lesions that may present as a mass in the pre-pontine region. Histopathological tests may find it difficult to distinguish between lesions that originate from notochord remnants. Imaging studies play a vital role in confirming the diagnosis and help in planning treatment and follow-up.

Deep learning versus iterative image reconstruction algorithm for head CT in trauma

PURPOSE

To compare the image quality between a deep learning-based image reconstruction algorithm (DLIR) and an adaptive statistical iterative reconstruction algorithm (ASiR-V) in noncontrast trauma head CT.

METHODS

Head CT scans from 94 consecutive trauma patients were included. Images were reconstructed with ASiR-V 50% and the DLIR strengths: low (DLIR-L), medium (DLIR-M), and high (DLIR-H). The image quality was assessed quantitatively and qualitatively and compared between the different reconstruction algorithms. Inter-reader agreement was assessed by weighted kappa.

RESULTS

DLIR-M and DLIR-H demonstrated lower image noise (p < 0.001 for all pairwise comparisons), higher SNR of up to 82.9% (p < 0.001), and higher CNR of up to 53.3% (p < 0.001) compared to ASiR-V. DLIR-H outperformed other DLIR strengths (p ranging from < 0.001 to 0.016). DLIR-M outperformed DLIR-L (p < 0.001) and ASiR-V (p < 0.001). The distribution of reader scores for DLIR-M and DLIR-H shifted towards higher scores compared to DLIR-L and ASiR-V. There was a tendency towards higher scores with increasing DLIR strengths. There were fewer non-diagnostic CT series for DLIR-M and DLIR-H compared to ASiR-V and DLIR-L. No images were graded as non-diagnostic for DLIR-H regarding intracranial hemorrhage. The inter-reader agreement was fair-good between the second most and the less experienced reader, poor-moderate between the most and the less experienced reader, and poor-fair between the most and the second most experienced reader.

CONCLUSION

The image quality of trauma head CT series reconstructed with DLIR outperformed those reconstructed with ASiR-V. In particular, DLIR-M and DLIR-H demonstrated significantly improved image quality and fewer non-diagnostic images. The improvement in qualitative image quality was greater for the second most and the less experienced readers compared to the most experienced reader.

Deep learning algorithm in detecting intracranial hemorrhages on emergency computed tomographies

BACKGROUND

Highly accurate detection of intracranial hemorrhages (ICH) on head computed tomography (HCT) scans can prove challenging at high-volume centers. This study aimed to determine the number of additional ICHs detected by an artificial intelligence (AI) algorithm and to evaluate reasons for erroneous results at a level I trauma center with teleradiology services.

METHODS

In a retrospective multi-center cohort study, consecutive emergency non-contrast HCT scans were analyzed by a commercially available ICH detection software (AIDOC, Tel Aviv, Israel). Discrepancies between AI analysis and initial radiology report (RR) were reviewed by a blinded neuroradiologist to determine the number of additional ICHs detected and evaluate reasons leading to errors.

RESULTS

4946 HCT (05/2020-09/2020) from 18 hospitals were included in the analysis. 205 reports (4.1%) were classified as hemorrhages by both radiology report and AI. Out of a total of 162 (3.3%) discrepant reports, 62 were confirmed as hemorrhages by the reference neuroradiologist. 33 ICHs were identified exclusively via RRs. The AI algorithm detected an additional 29 instances of ICH, missed 12.4% of ICH and overcalled 1.9%; RRs missed 10.9% of ICHs and overcalled 0.2%. Many of the ICHs missed by the AI algorithm were located in the subarachnoid space (42.4%) and under the calvaria (48.5%). 85% of ICHs missed by RRs occurred outside of regular working-hours. Calcifications (39.3%), beam-hardening artifacts (18%), tumors (15.7%), and blood vessels (7.9%) were the most common reasons for AI overcalls. ICH size, image quality, and primary examiner experience were not found to be significantly associated with likelihood of incorrect AI results.

CONCLUSION

Complementing human expertise with AI resulted in a 12.2% increase in ICH detection. The AI algorithm overcalled 1.9% HCT.

TRIAL REGISTRATION

German Clinical Trials Register (DRKS-ID: DRKS00023593).

Ecchordosis physaliphora: Case report and brief review of the literature

Ecchordosis physaliphora is a rare congenital benign hamartomatous lesion originating from nodal cord remnants. This is histopathologically indistinguishable from chordoma, and hence imaging plays a key role in diagnosis. These lesions are hypointense on T1-weighted and hyperintense on T2-weighted images, and follow CSF signal. In contrast to chordoma, Ecchordosis Physaliphora does not demonstrate contrast enhancement. Here, we present a case of 32-year-old male with no prior medical history, who presented to an outside facility for chronic headache workup and incidentally detected indeterminate lytic defect in the bony clivus with a well demarcated smoothly corticated margin. Further assessment with MRI brain showed findings characteristic of Ecchordosis physaliphora, a benign congenital hamartomatous lesion originating from nodal cord remnants requiring no additional follow-up imaging or intervention.

A deep learning algorithm for automatic detection and classification of acute intracranial hemorrhages in head CT scans

Acute Intracranial hemorrhage (ICH) is a life-threatening disease that requires emergency medical attention, which is routinely diagnosed using non-contrast head CT imaging. The diagnostic accuracy of acute ICH on CT varies greatly among radiologists due to the difficulty of interpreting subtle findings and the time pressure associated with the ever-increasing workload. The use of artificial intelligence technology may help automate the process and assist radiologists for more prompt and better decision-making. In this work, we design a deep learning approach that mimics the interpretation process of radiologists, and combines a 2D CNN model and two sequence models to achieve accurate acute ICH detection and subtype classification. Being developed using the extensive 2019-RSNA Brain CT Hemorrhage Challenge dataset with over 25000 CT scans, our deep learning algorithm can accurately classify the acute ICH and its five subtypes with AUCs of 0.988 (ICH), 0.984 (EDH), 0.992 (IPH), 0.996 (IVH), 0.985 (SAH), and 0.983 (SDH), respectively, reaching the accuracy level of expert radiologists. Our method won 1st place among 1345 teams from 75 countries in the RSNA challenge. We have further evaluated our algorithm on two independent external validation datasets with 75 and 491 CT scans, respectively, and our method maintained high AUCs of 0.964 and 0.949 for acute ICH detection. These results have demonstrated the high performance and robust generalization ability of our proposed method, which makes it a useful second-read or triage tool that can facilitate routine clinical applications.

Radiology and A Radiologist: A Keystone in the Turmoil of Trauma Setting

Traumatic injuries are one of the leading causes of morbidity and mortality. Precise diagnosis and management in the golden hour are key to decrease morbidity and mortality. History and physical examination alone are insufficient to avoid misdiagnosis. In this article, we tried to determine the role of a radiologist and an appropriate imaging modality in a trauma setting. We conducted a literature review of published research articles. We used the keywords imaging, trauma, imaging and trauma, and trauma imaging essentials were used on PubMed and Google Scholar. The articles published in the English language from 2015 to 2020 with full free text available were included. Using the medical subject heading (MeSH) strategy, "diagnostic imaging" (Major {Majr}) and "multiple trauma/diagnostic imaging" (Mesh) on PubMed, we identified 34 papers after applying the inclusion and exclusion criteria. Twenty articles were finally selected which included studies from 2015 to 2020 with articles focusing on the adult population and acute cases. A radiologist and imaging modalities are the essential parts of a trauma setting to lower morbidity and mortality. X-rays and Extended Focussed Assessment with Sonography for Trauma (eFAST) are the first-line imaging modality in the acute trauma setting. However, the CT scan is the most sensitive modality that should be done to avoid misdiagnosis depending upon the patient's history and physical examination.

Split hypoglossal facial anastomosis for facial nerve palsy due to skull base fractures: A case report

Introduction

Traumatic brain injury (TBI) is the most prevalent causes of morbidity and mortality worldwide. The biomechanics of primary TBI involve a direct impact, practically extended to the base of the skull, and most of the skull base fractures (SBF) are identified in anterior and medial cranial fossa. Furthermore, those predicted in the medial area are related to fissures from temporal bones.

Presentation of case

We report two cases of right facial nerve palsy initiated by SBF's, which were diagnosed and treated at our institution. The 3D CT evaluation in our first case showed a longitudinal fracture of the right petrosal bone, which was longitudinal and transverse for the second case. Two cases of facial nerve palsy were managed with split hypoglossal facial anastomosis to restore functional reanimation. All patients were adequately achieved after the procedure, and the hypoglossal nerve function was preserved.

Conclusion

Split hypoglossal facial anastomosis technique was used to treat patients with facial nerve paralysis resulting from SBF's. This was to achieve good recovery outcome, in terms of facial reanimation and preservation of tongue function.

•

A skull base fracture (SBF) is about 4% of all cases Traumatic brain injury (TBI).

•

SBF which frequently occurs in the petrous part of the temporal bone, is implicated in facial nerve palsy.

•

Split hypoglossal facial anastomosis technique showed good recovery of facial reanimation with HB scale assessment.