Head computed tomography in the emergency department: a collection of easily missed findings that are life-threatening or life-changing

Artificial Intelligence Trained by Deep Learning Can Improve Computed Tomography Diagnosis of Nontraumatic Subarachnoid Hemorrhage by Nonspecialists

Subarachnoid hemorrhage (SAH) is a serious cerebrovascular disease with a high mortality rate and is known as a disease that is hard to diagnose because it may be overlooked by noncontrast computed tomography (NCCT) examinations that are most frequently used for diagnosis. To create a system preventing this oversight of SAH, we trained artificial intelligence (AI) with NCCT images obtained from 419 patients with nontraumatic SAH and 338 healthy subjects and created an AI system capable of diagnosing the presence and location of SAH. Then, we conducted experiments in which five neurosurgery specialists, five nonspecialists, and the AI system interpreted NCCT images obtained from 135 patients with SAH and 196 normal subjects. The AI system was capable of performing a diagnosis of SAH with equal accuracy to that of five neurosurgery specialists, and the accuracy was higher than that of nonspecialists. Furthermore, the diagnostic accuracy of four out of five nonspecialists improved by interpreting NCCT images using the diagnostic results of the AI system as a reference, and the number of oversight cases was significantly reduced by the support of the AI system. This is the first report demonstrating that an AI system improved the diagnostic accuracy of SAH by nonspecialists.

Blind spots on CT imaging of the head: Insights from 5 years of report addenda at a single institution

BACKGROUND

Errors of detection ("misses") are the major source of error in radiology. There is sparse prior literature describing patterns of detection error on CT head imaging.

PURPOSE

The objective of this study was to gain insight to areas on CT head imaging where radiologists are most likely to miss clinically relevant findings.

METHODS

We performed a cross-sectional study of consecutive reports of CT imaging of the head at a single institution spanning 5/1/2013-5/1/2018 (5 years). Detection errors described in addenda were categorized according to anatomic location, type of pathology, and potential impact on management. Blind spots were defined by the most common sites of missed findings.

RESULTS

A total of 165,943 reports for CT head imaging were obtained. Addenda were found in 1658 (~1%) of reports, of which 359 (21.7%) described errors of detection. Within the extracranial soft tissues (n = 73) the most common "misses" were at incidentally imaged parotid glands and the frontal scalp. Within osseous structures (n = 149), blind spots included the nasal and occipital bones. Vascular lesions (n = 47) which passed detection were most common at the distal MCA, carotid terminus and sigmoid sinus/jugular bulb. No predisposition was seen for anatomic subsites within the CSF space (n = 60) and brain parenchyma (n = 65).

CONCLUSIONS

Consistent patterns of blind spots are revealed. Radiologic teaching and search patterns to account for these sites of error may accelerate trainee competence and improve accuracy in the practice of radiology.

ACR Appropriateness Criteria® Acute Mental Status Change, Delirium, and New Onset Psychosis

Acute changes in mental status represent a broad collection of symptoms used to describe disorders in mentation and level of arousal, including the more narrowly defined diagnoses of delirium and psychosis. A wide range of precipitating factors may be responsible for symptom onset including infection, intoxication, and metabolic disorders. Neurologic causes that may be detected on neuroimaging include stroke, traumatic brain injury, nonconvulsive seizure, central nervous system infection, tumors, hydrocephalus, and inflammatory disorders. Not infrequently, two or more precipitating factors may be found. Neuroimaging with CT or MRI is usually appropriate if the clinical suspicion for an acute neurological cause is high, where the cause of symptoms is not found on initial assessment, and for patients whose symptoms do not respond appropriately to management. There was disagreement regarding the appropriateness of neuroimaging in cases where a suspected, nonneurologic cause is found on initial assessment. Neuroimaging with CT is usually appropriate for patients presenting with delirium, although the yield may be low in the absence of trauma or a focal neurological deficit. Neuroimaging with CT or MRI may be appropriate in the evaluation of new onset psychosis, although the yield may be low in the absence of a neurologic deficit. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

Stroke scan agnosia - What radiologists may not see

OBJECTIVE

Neuroimaging helps neurologists make accurate diagnoses. We used a multidisciplinary review system to search for perceptual discrepancies in stroke lesions. We sought to identify recurrent pitfalls in the detection of neuroimaging stroke lesions.

PATIENTS AND METHODS

Patients were selected from a neuroimaging database of second opinions if cerebrovascular lesions had been missed at initial reporting. Patient demographics, scanning modality and stroke type were recorded.

RESULTS

A neuroradiologist second opinion was provided for 1336 patients. Forty-four patients, 18 women and 26 men, mean age 59.9 (SD 14.2) years, were identified in whom a vascular lesion was not detected on initial reporting. The lesions included cerebellar infarcts in 17 patients (bilateral in 7), pontine infarction/ischaemia (n=5), pontine and cerebellar lesions (n=1) and spinal infarction (n=1). Supratentorial infarction occurred in 10 patients of which 3 were thalamic infarcts. Vessel abnormalities were present in 8 patients (hyperdense vessel n=3, dissection n=3, middle cerebral artery occlusion on CTA n=1 and cerebral venous sinus thrombosis n=1). Convexity subarachnoid hemorrhage was missed and a subdural hematoma was not identified in one patient. In 10 (23 %) patients the missed lesions occurred solely on CT brain scanning. The missed lesions were symptomatic in 28 (64 %) patients and presentations were acute in 14 (32 %) patients.

CONCLUSION

Some cerebrovascular lesions are prone to perceptual errors with CT and MRI brain scanning. Radiologists and neurologists should be aware that posterior fossa lesions (particularly in the cerebellum and pons) and hyperdense vessel signs may be missed. Better identification of radiological cerebrovascular lesions should enhance management of acute and chronic stroke patients.

Structural Neuroimaging Findings in Mild Traumatic Brain Injury

Common neuroimaging findings in mild traumatic brain injury (mTBI), including sport-related concussion (SRC), are reviewed based on computed tomography and magnetic resonance imaging (MRI). Common abnormalities radiologically identified on the day of injury, typically a computed tomographic scan, are in the form of contusions, small subarachnoid or intraparenchymal hemorrhages as well as subdural and epidural collections, edema, and skull fractures. Common follow-up neuroimaging findings with MRI include white matter hyperintensities, hypointense signal abnormalities that reflect prior hemorrhage, focal encephalomalacia, presence of atrophy and/or dilated Virchow-Robins perivascular space. The MRI findings from a large pediatric mTBI study show low frequency of positive MRI findings at 6 months postinjury. The review concludes with an examination of some of the advanced MRI-based image analysis methods that can be performed in the patient who has sustained an mTBI.