Recurrent Sub-volume Analysis of Head CT Scans for the Detection of Intracranial Hemorrhage

2019 ◽  
pp. 864-872 ◽  
Author(s):  
M. S. Vidya ◽  
Yogish Mallya ◽  
Arun Shastry ◽  
J. Vijayananda
Keyword(s):  
Intracranial Hemorrhage ◽  
Ct Scans ◽  
Volume Analysis ◽  
Head Ct

scholarly journals Expert-level detection of acute intracranial hemorrhage on head computed tomography using deep learning

2019 ◽  
Vol 116 (45) ◽  
pp. 22737-22745 ◽  
Author(s):  
Weicheng Kuo ◽  
Christian Hӓne ◽  
Pratik Mukherjee ◽  
Jitendra Malik ◽  
Esther L. Yuh
Keyword(s):  
Computed Tomography ◽  
San Francisco ◽  
Intracranial Hemorrhage ◽  
Imaging Modality ◽  
Area Under The Curve ◽  
High Specificity ◽  
Ct Scans ◽  
Cross Sectional ◽  
Unique Challenge ◽  
Head Ct

Computed tomography (CT) of the head is used worldwide to diagnose neurologic emergencies. However, expertise is required to interpret these scans, and even highly trained experts may miss subtle life-threatening findings. For head CT, a unique challenge is to identify, with perfect or near-perfect sensitivity and very high specificity, often small subtle abnormalities on a multislice cross-sectional (three-dimensional [3D]) imaging modality that is characterized by poor soft tissue contrast, low signal-to-noise using current low radiation-dose protocols, and a high incidence of artifacts. We trained a fully convolutional neural network with 4,396 head CT scans performed at the University of California at San Francisco and affiliated hospitals and compared the algorithm’s performance to that of 4 American Board of Radiology (ABR) certified radiologists on an independent test set of 200 randomly selected head CT scans. Our algorithm demonstrated the highest accuracy to date for this clinical application, with a receiver operating characteristic (ROC) area under the curve (AUC) of 0.991 ± 0.006 for identification of examinations positive for acute intracranial hemorrhage, and also exceeded the performance of 2 of 4 radiologists. We demonstrate an end-to-end network that performs joint classification and segmentation with examination-level classification comparable to experts, in addition to robust localization of abnormalities, including some that are missed by radiologists, both of which are critically important elements for this application.

scholarly journals Enlarged subarachnoid spaces and intracranial hemorrhage in children with accidental head trauma

2017 ◽  
Vol 19 (2) ◽  
pp. 254-258 ◽  
Author(s):  
Amanda K. Fingarson ◽  
Maura E. Ryan ◽  
Suzanne G. McLone ◽  
Corey Bregman ◽  
Emalee G. Flaherty
Keyword(s):  
Intracranial Hemorrhage ◽  
Head Circumference ◽  
Child Protection ◽  
Ct Scanning ◽  
Ct Scans ◽  
Bivariate Analysis ◽  
Subdural Hemorrhage ◽  
Head Ct ◽  
Subarachnoid Spaces ◽  
Clinical Measurements

OBJECTIVE Benign external hydrocephalus (BEH) is an enlargement of the subarachnoid spaces (SASs) that can be seen in young children. It is controversial whether children with BEH are predisposed to developing subdural hemorrhage (SDH) with or without trauma. This issue is clinically relevant as a finding of unexplained SDH raises concerns about child abuse and often prompts child protection and law enforcement investigations. METHODS This retrospective study included children (1–24 months of age) who underwent head CT scanning after an accidental fall of less than 6 feet. Head CT scans were reviewed, cranial findings were documented, and the SAS was measured and qualitatively evaluated. Enlarged SAS was defined as an extraaxial space (EAS) greater than 4 mm on CT scans. Clinical measurements of head circumference (HC) were noted, and the head circumference percentile was calculated. The relationship between enlarged SAS and HC percentile, and enlarged SAS and intracranial hemorrhage (ICH), were investigated using bivariate analysis. RESULTS Of the 110 children included in this sample, 23 had EASs greater than 4 mm. The mean patient age was 6.8 months (median 6.0 months). Thirty-four patients (30.9%) had ICHs, including subarachnoid/subpial (6.2%), subdural (6.2%), epidural (5.0%), and unspecified extraaxial hemorrhage (16.5%). Enlarged SAS was positively associated with subarachnoid/subpial hemorrhage; there was no association between enlarged SASs and either SDH or epidural hemorrhage. A larger SAS was positively associated with larger HC percentile; however, HC percentile was not independently associated with ICH. CONCLUSIONS Enlarged SAS was not associated with SDH, but was associated with other ICHs. The authors' findings do not support the theory that BEH predisposes children to SDH with minor accidental trauma.

Etiology and Pathogenesis of Intraventricular Hemorrhage in Newborns

PEDIATRICS ◽  
1989 ◽  
Vol 84 (2) ◽  
pp. 382-385
Author(s):  
KAREN E. PAPE
Keyword(s):  
Intracranial Hemorrhage ◽  
Intraventricular Hemorrhage ◽  
Ct Scanning ◽  
Postmortem Examination ◽  
Brain Lesions ◽  
Ct Scans ◽  
Clinical Criteria ◽  
The Status

The article by Miall-Allen et al is an important contribution to the status of BP control and intracranial hemorrhage in the newborn. Until 10 years ago, there were limited methods of assessment. Hemorrhagic brain lesions were diagnosed mainly by clinical criteria and postmortem examination. In the late 1970s, CT scans provided much more in vivo information. However, CT scanning is inherently difficult in the newborn. Early machines produced a significant amount of radiation and the infants had to be transported to a radiology unit. These difficulties meant that a limited number of scans were done and children were rarely sequentially exposed.

Head CT Decision Rules and Intracranial Hemorrhage

2013 ◽  
pp. 47-55
Author(s):  
Brandon Allen ◽  
Latha Ganti ◽  
Bobby Desai
Keyword(s):  
Intracranial Hemorrhage ◽  
Decision Rules ◽  
Head Ct

Abstract WP68: Interpretable Deep Learning-based Characterization of Intracranial Hemorrhage on Head CT

Stroke ◽  
2019 ◽  
Vol 50 (Suppl_1) ◽  
Author(s):  
Bernardo Bizzo ◽  
Behrooz Hashemian ◽  
Troy McNitt ◽  
Michael T Caton ◽  
Walter Wiggins ◽  
...  
Keyword(s):  
Deep Learning ◽  
Intracranial Hemorrhage ◽  
Head Ct

Size-specific dose estimations for pediatric chest, abdomen/pelvis and head CT scans with the use of GATE

2019 ◽  
Vol 65 ◽  
pp. 181-190 ◽  
Author(s):  
Theodora Kostou ◽  
Panagiotis Papadimitroulas ◽  
Pavlos Papaconstadopoulos ◽  
Slobodan Devic ◽  
Jan Seuntjens ◽  
...  
Keyword(s):  
Ct Scans ◽  
Head Ct ◽  
Pediatric Chest

Relevance of early head CT scans following neurosurgical procedures: an analysis of 892 intracranial procedures at Rush University Medical Center

2014 ◽  
Vol 121 (2) ◽  
pp. 307-312 ◽  
Author(s):  
Ricardo B. V. Fontes ◽  
Adam P. Smith ◽  
Lorenzo F. Muñoz ◽  
Richard W. Byrne ◽  
Vincent C. Traynelis
Keyword(s):  
Operating Room ◽  
Neurological Examination ◽  
Medical Center ◽  
Ct Scanning ◽  
Ct Scans ◽  
Type I ◽  
Type Ii ◽  
Head Ct ◽  
Ct Findings ◽  
Neurological Change

Object Early postoperative head CT scanning is routinely performed following intracranial procedures for detection of complications, but its real value remains uncertain: so-called abnormal results are frequently found, but active, emergency intervention based on these findings may be rare. The authors' objective was to analyze whether early postoperative CT scans led to emergency surgical interventions and if the results of neurological examination predicted this occurrence. Methods The authors retrospectively analyzed 892 intracranial procedures followed by an early postoperative CT scan performed over a 1-year period at Rush University Medical Center and classified these cases according to postoperative neurological status: baseline, predicted neurological change, unexpected neurological change, and sedated or comatose. The interpretation of CT results was reviewed and unexpected CT findings were classified based on immediate action taken: Type I, additional observation and CT; Type II, active nonsurgical intervention; and Type III, surgical intervention. Results were compared between neurological examination groups with the Fisher exact test. Results Patients with unexpected neurological changes or in the sedated or comatose group had significantly more unexpected findings on the postoperative CT (p < 0.001; OR 19.2 and 2.3, respectively) and Type II/III interventions (p < 0.001) than patients at baseline. Patients at baseline or with expected neurological changes still had a rate of Type II/III changes in the 2.2%–2.4% range; however, no patient required an immediate return to the operating room. Conclusions Over a 1-year period in an academic neurosurgery service, no patient who was neurologically intact or who had a predicted neurological change required an immediate return to the operating room based on early postoperative CT findings. Obtaining early CT scans should not be a priority in these patients and may even be cancelled in favor of MRI studies, if the latter have already been planned and can be performed safely and in a timely manner. Early postoperative CT scanning does not assure an uneventful course, nor should it replace accurate and frequent neurological checks, because operative interventions were always decided in conjunction with the neurological examination.

Sign in / Sign up

Export Citation Format

Share Document