Head Computed Tomography Imaging Research Articles

Deep learning architecture for scatter estimation in cone-beam computed tomography head imaging with varying field-of-measurement settings.

X-ray scatter causes considerable degradation in the cone-beam computed tomography (CBCT) image quality. To estimate the scatter, deep learning-based methods have been demonstrated to be effective. Modern CBCT systems can scan a wide range of field-of-measurement (FOM) sizes. Variations in the size of FOM can cause a major shift in the scatter-to-primary ratio in CBCT. However, the scatter estimation performance of deep learning networks has not been extensively evaluated under varying FOMs. Therefore, we train the state-of-the-art scatter estimation neural networks for varying FOMs and develop a method to utilize FOM size information to improve performance. We used FOM size information as additional features by converting it into two channels and then concatenating it to the encoder of the networks. We compared our approach for a U-Net, Spline-Net, and DSE-Net, by training them with and without the FOM information. We utilized a Monte Carlo-simulated dataset to train the networks on 18 FOM sizes and test on 30 unseen FOM sizes. In addition, we evaluated the models on the water phantoms and real clinical CBCT scans. The simulation study demonstrates that our method reduced average mean-absolute-percentage-error for U-Net by 38%, Spline-Net by 40%, and DSE-net by 33% for the scatter estimation in the 2D projection domain. Furthermore, the root-mean-square error on the 3D reconstructed volumes was improved for U-Net by 43%, Spline-Net by 30%, and DSE-Net by 23%. Furthermore, our method improved contrast and image quality on real datasets such as water phantom and clinical data. Providing additional information about FOM size improves the robustness of the neural networks for scatter estimation. Our approach is not limited to utilizing only FOM size information; more variables such as tube voltage, scanning geometry, and patient size can be added to improve the robustness of a single network.

Assessing the influence of dose modulation on water equivalent diameter estimation in pediatric head CT imaging

Pediatric head computed tomography (CT) scans necessitate precise determination of radiation dose to optimize image quality while minimizing patient exposure. This study explores the influence of head size variation on radiation dose and image quality using phantom and patients set, aiming to investigate the efficacy of Automatic Exposure Control (AEC) in dose modulation. Image noise and radiation dose were assessed using with and without AEC, maintaining a constant tube current of 180 mA. The correlation between CT dose index volume (CTDIvol) and water-equivalent diameter (Dw) was examined across 134 image sets, revealing an exponential relationship indicative of AEC's contribution to elevated radiation exposure in larger heads. Furthermore, a slice-based analysis demonstrated a consistent variation in tube current along the slice location, effectively modulated by Dw. However, discrepancies in tube current were observed in the caudal region, suggesting the influence of factors beyond section size. Refinement of analysis improved curve fitting, highlighting the importance of section size in dose modulation. AEC proved essential in maintaining image quality consistency across slices, with inactive AEC resulting in lower image noise in cranial slices. These findings highlight the significance of AEC-based dose modulation in pediatric head CT to ensure optimal image quality while minimizing radiation exposure, emphasizing the need for comprehensive understanding and implementation of AEC systems from different manufacturers.

Evaluation of Sphenoid Sinus Pneumatization Subtypes in Posteroanterior and Lateral Directions on CT Imaging

Background: Sphenoid sinus (SS) variations are diverse, and identifying them before transsphenoidal interventions is essential. The gold standard imaging method for assessing the morphological characteristics of the SS is computed tomography (CT), which yields valuable information. Objectives: This study used CT imaging to evaluate SS pneumatization in the Turkish population's posteroanterior (PA) and lateral directions. Methods: Head CT images of 509 cases in total were evaluated retrospectively. Initially, SS types were determined based on PA pneumatization. Afterward, the new classification was used, and SS pneumatizations were defined in the lateral direction accordingly. Results: According to PA pneumatization, conchal (0.6%), presellar (16.7%), sellar (66.4%), and post-sellar pneumatization (16.3%) were detected. Accordingly, the relationship between SS type and gender was insignificant (P = 0.354). When the lateral type pneumatization classification was used, the right lateral body (39.9%) and left lateral body (32.2%) were found to be the most common types. The relationship between SS lateral-type pneumatization and gender was insignificant (P = 0.388). Conclusions: Sphenoid sinus pneumatization shows significant individual variability, which is beneficial to evaluate before surgery. In addition, more information was obtained about pneumatization using lateral direction typing as a new classification.

Impact of intelligent convolutional neural network -based algorithms on head computed tomography evaluation and comprehensive rehabilitation acupuncture therapy for patients with cerebral infarction

This work was to evaluate the impacts of comprehensive rehabilitation acupuncture therapy on the recovery of neurological function in cerebral infarction (CI) patients and to utilize convolutional neural network (CNN) intelligent algorithms to optimize head computed tomography (CT) images and improve lesion localization accuracy. 98 CI patients were divided into a control group (Ctrl group) and an experimental group (Exp group), with 48 patients in each group. The patients in the Ctrl group received CT evaluation combined with comprehensive rehabilitation acupuncture therapy. While, those in the Exp group received CT evaluation with the use of CNN algorithms for optimization, along with comprehensive rehabilitation acupuncture therapy. Acupuncture therapy included selecting acupoints on the patient’s head, selecting two horizontal needling needles from top to bottom at the acupoints on the front side of the lesion, and then horizontal needling along the top midline. The differences in treatment outcomes were compared between the two groups based on Fugl-Meyer upper limb assessment (FMA) scores, Barthel Index (BI) scores, National Institutes of Health Stroke Scale (NIHSS4) scores, Modified Edinburgh-Scandinavian Stroke Scale (MESSS) scores, and hemodynamics. Simultaneously, the CT images were optimized using CNN intelligent algorithms to improve image quality and lesion localization accuracy. The results showed that the CI CT images processed by the CNN-based intelligent algorithm showed significant improvements in clarity and contrast compared to conventional CT images. The CNN-based intelligent algorithm demonstrated higher sensitivity (97.5 %, 93.8 %), higher PSNR (30.14 dB, 24.72 dB), and lower missed detection rate (0.52 %, 1.88 %) in detecting CI lesions. The total effective rate in the Exp group was 95.83 %, which was significantly higher than the 85.42 % in the Ctrl group (P < 0.05). The Exp group showed significantly higher levels in FMA and BI scores (P < 0.05). After treatment, the NIHSS4 and MESSS scores in the Exp group were lower than those in the Ctrl group (P < 0.05). Additionally, post-treatment, the plasma concentrations and whole-blood viscosity (low shear and high shear) in the Exp group were lower than those in the Ctrl group, and the plasma concentration and whole-blood viscosity (high shear) were also lower than those in the Ctrl group (P < 0.05). In conclusion, comprehensive rehabilitation acupuncture therapy had a positive impact on the recovery of neurological function in CI patients. By applying CNN-based intelligent algorithms to optimize head CT images, lesion localization accuracy can be improved, thereby guiding rehabilitation treatment more effectively.

Estimate and compensate head motion in non-contrast head CT scans using partial angle reconstruction and deep learning.

Patient head motion is a common source of image artifacts in computed tomography (CT) of the head, leading to degraded image quality and potentially incorrect diagnoses. The partial angle reconstruction (PAR) means dividing the CT projection into several consecutive angular segments and reconstructing each segment individually. Although motion estimation and compensation using PAR has been developed and investigated in cardiac CT scans, its potential for reducing motion artifacts in head CT scans remains unexplored. To develop a deep learning (DL) model capable of directly estimating head motion from PAR images of head CT scans and to integrate the estimated motion into an iterative reconstruction process to compensate for the motion. Head motion is considered as a rigid transformation described by six time-variant variables, including the three variables for translation and three variables for rotation. Each motion variable is modeled using a B-spline defined by five control points (CP) along time. We split the full projections from 360° into 25 consecutive PARs and subsequently input them into a convolutional neural network (CNN) that outputs the estimated CPs for each motion variable. The estimated CPs are used to calculate the object motion in each projection, which are incorporated into the forward and backprojection of an iterative reconstruction algorithm to reconstruct the motion-compensated image. The performance of our DL model is evaluated through both simulation and phantom studies. The DL model achieved high accuracy in estimating head motion, as demonstrated in both the simulation study (mean absolute error (MAE) ranging from 0.28 to 0.45mm or degree across different motion variables) and the phantom study (MAE ranging from 0.40 to 0.48mm or degree). The resulting motion-corrected image, , exhibited a significant reduction in motion artifacts when compared to the traditional filtered back-projection reconstructions, which is evidenced both in the simulation study (image MAE drops from 178 33HU to 37 9HU, structural similarity index (SSIM) increases from 0.60 0.06 to 0.98 0.01) and the phantom study (image MAE drops from 117 17HU to 42 19HU, SSIM increases from 0.83 0.04 to 0.98 0.02). We demonstrate that using PAR and our proposed deep learning model enables accurate estimation of patient head motion and effectively reduces motion artifacts in the resulting head CT images.

Advancing pediatric head CT dose prediction: correlating patient size metrics with AP and LAT dimensions in Moroccan population

This study aims to establish comprehensive relationships between patient size, focusing on effective diameter (Deff) and water-equivalent diameter (Dw), in conjunction with anterior-posterior (AP) and lateral (LAT) dimensions. The primary goal is to refine precise patient dose estimations in pediatric head computed tomography (CT) examinations. Our examination involved a dataset of 105 pediatric head CT images sourced from a Moroccan pediatric university hospital. These images, selected from the central scanning range, underwent quantification for LAT, AP, Deff, and Dw, with measurements carried out using RadiAnt Dicom Viewer for lateral and anteroposterior dimensions. Our analytical approach encompassed modeling Deff and Dw as functions of LAT, AP, and combined AP + LAT dimensions. Additionally, Dw was modeled as a function of Deff and calculated Size Specific Dose Estimates (SSDE) based on Dw was compared to SSDE computed from one single dimension. The outcomes revealed strong correlations among Deff, Dw, and fundamental geometric dimensions (LAT, AP, and AP + LAT), as indicated by consistently high coefficients of determination (R² > 0.8). SSDE calculated from Dw shows impressive correlation with SSDE computed from a single dimension (R² > 0.9). This study offers insights into the relationships between patient size and critical dimensions, identifying high correlations that can represent a significant advancement in predicting doses for pediatric head CT examinations. This introduces the potential for streamlining the determination of head diameter and dose by relying on a single dimension, thereby enhancing the efficiency of dose predictions by facilitating the monitoring of radiation exposure of this population in clinical routine.

Motion Artifact Detection Based on Regional–Temporal Graph Attention Network from Head Computed Tomography Images

Artifacts are the main cause of degradation in CT image quality and diagnostic accuracy. Because of the complex texture of CT images, it is a challenging task to automatically detect artifacts from limited image samples. Recently, graph convolutional networks (GCNs) have achieved great success and shown promising results in medical imaging due to their powerful learning ability. However, GCNs do not take the attention mechanism into consideration. To overcome their limitations, we propose a novel Regional–Temporal Graph Attention Network for motion artifact detection from computed tomography images (RT-GAT). In this paper, head CT images are viewed as a heterogeneous graph by taking regional and temporal information into consideration, and the graph attention network is utilized to extract the features of the constructed graph. Then, the feature vector is input into the classifier to detect the motion artifacts. The experimental results demonstrate that our proposed RT-GAT method outperforms the state-of-the-art methods on a real-world CT dataset.

VOARiability: Interobserver and intermodality variability analysis in OAR contouring from head and neck CT and MR images.

Accurate and consistent contouring of organs-at-risk (OARs) from medical images is a key step of radiotherapy (RT) cancer treatment planning. Most contouring approaches rely on computed tomography (CT) images, but the integration of complementary magnetic resonance (MR) modality is highly recommended, especially from the perspective of OAR contouring, synthetic CT and MR image generation for MR-only RT, and MR-guided RT. Although MR has been recognized as valuable for contouring OARs in the head and neck (HaN) region, the accuracy and consistency of the resulting contours have not been yet objectively evaluated. To analyze the interobserver and intermodality variability in contouring OARs in the HaN region, performed by observers with different level of experience from CT and MR images of the samepatients. In the final cohort of 27 CT and MR images of the same patients, contours of up to 31 OARs were obtained by a radiation oncology resident (junior observer, JO) and a board-certified radiation oncologist (senior observer, SO). The resulting contours were then evaluated in terms of interobserver variability, characterized as the agreement among different observers (JO and SO) when contouring OARs in a selected modality (CT or MR), and intermodality variability, characterized as the agreement among different modalities (CT and MR) when OARs were contoured by a selected observer (JO or SO), both by the Dice coefficient (DC) and 95-percentile Hausdorff distance (HD ). The mean (±standard deviation) interobserver variability was 69.0 ± 20.2% and 5.1 ± 4.1mm, while the mean intermodality variability was 61.6 ± 19.0% and 6.1 ± 4.3mm in terms of DC and HD , respectively, across all OARs. Statistically significant differences were only found for specific OARs. The performed MR to CT image registration resulted in a mean target registration error of 1.7 ± 0.5mm, which was considered as valid for the analysis of intermodalityvariability. The contouring variability was, in general, similar for both image modalities, and experience did not considerably affect the contouring performance. However, the results indicate that an OAR is difficult to contour regardless of whether it is contoured in the CT or MR image, and that observer experience may be an important factor for OARs that are deemed difficult to contour. Several of the differences in the resulting variability can be also attributed to adherence to guidelines, especially for OARs with poor visibility or without distinctive boundaries in either CT or MR images. Although considerable contouring differences were observed for specific OARs, it can be concluded that almost all OARs can be contoured with a similar degree of variability in either the CT or MR modality, which works in favor of MR images from the perspective of MR-only and MR-guidedRT.

Comparison of Image Quality and Radiation Dose between Filtered Back Projection and Sinogram Affirmed Iterative Reconstruction in CT Head: A Cross-sectional Study

Introduction: Computed Tomography (CT) scans account for only 15% of total diagnostic procedures but for over half of the collective dose of radiation. The increasing awareness about the harmful effects of radiation has also created a need for developing techniques that decrease radiation exposure while at the same time providing a reasonably good image quality. The majority of the CT units today use Filtered Back Projection (FBP) as the means of image reconstruction. However, FBP leads to coupling between image noise and radiation exposure and limits the dose reduction possible while providing diagnostic quality images. Second-generation Iterative Reconstruction (IR) techniques have caught the attention of medical researchers because they provide a superior image quality than FBP at the same radiation dose showing potential for dose reduction. Aim: To compare the image quality and radiation dose in standard dose CT head reconstructed with FBP and low dose CT head reconstructed with Sinogram Affirmed Iterative Reconstruction (SAFIRE) and FBP both. Materials and Methods: A cross-sectional analytical study was conducted on 100 patients in the emergency department of Vardhman Mahavir Medical College and Safdarjung Hospital in India from November 2018 to April 2020. Patients referred for emergency Non-contrast Computed Tomography (NCCT) head for any indication excluding post-op cases and those with metallic artifacts were included. Fifty patients chosen using systematic random sampling underwent low-dose CT head using the Care Dose 4D Automatic Exposure Control (AEC) system with reconstruction using SAFIRE (Group-A) and FBP (Group-B). Another 50 patients who were within five years of the corresponding Group-A and B patients underwent standard dose head CT reconstruction using FBP (Group-C). CTDI vol, DLP, and effective dose were recorded in all patients. Image quality was assessed objectively using the unpaired t-test for Group-A and C and paired t-test for Group-A and B. Subjective image analysis between the groups was done using a 4-point Likert scale. Results: Image quality parameters were found to be better in Group-A compared to Group-C (p&lt;0.05). The mean values in low dose SAFIRE group, low dose FBP group, and standard dose FBP group were as follows: Grey Matter (GM) Signal-to-Noise Ratio (SNR) (14±3.29, 8.96±2 and 9.24±1.96), White Matter (WM) SNR (14.6 ±3.73, 6.9±3.79 and 7.11±1.68), Cerebrospinal Fluid (CSF) SNR (1.65±1.12, 0.86±0.80 and 1.08±0.61) and Contrastto-Noise Ratio (CNR) (3.06±0.94, 1.81±0.69 and 1.74±0.69), respectively. Group-A had significantly improved image quality parameters than Group-B and Group-C. Radiation dose reduction of 42%, 39.34%, and 42.5% was achieved in CTDIvol, DLP, and effective dose respectively in low dose group. Conclusion: Low dose CT head reconstructed with SAFIRE were significantly better in image quality compared to standarddose CT head images reconstructed using FBP, while allowing for up to 42% reduction in dose.

Artifact Assessment in the Skull Base Region of Head CT Images with Various Tilt Angles Relative to the Orbitomeatal Line

The aim of this study was to evaluate artifacts in the skull base region of head computed tomography (CT) images with various tilt angles relative to the orbitomeatal line. CT images of a head phantom acquired by helical and non-helical scanning with the tilt angles set from 0 to 20 degrees in 5-degree increments were evaluated in this study. Regions of interest (ROIs) were set at the cerebellum, temporal lobe, frontal lobe, and basal ganglia in the phantom images. Artifacts were evaluated by the coefficient of variation (CV) of the mean CT value between ROIs and the location parameter (β) of the Gumbel method. The CV and β values increased with increasing tilt angle for both helical and nonhelical images in the frontal lobes, but both decreased in the cerebellar region. In the temporal lobe and basal ganglia, there was no trend of change with tilt angle. Increasing the tilt angle relative to the OM line increased artifacts at the frontal lobes and decreased artifacts at the cerebellar region.

Enhancing facial feature de-identification in multiframe brain images: A generative adversarial network approach.

The collection of head images for public datasets in the field of brain science has grown remarkably in recent years, underscoring the need for robust de-identification methods to adhere with privacy regulations. This paper elucidates a novel deep learning-based approach to deidentifying facial features in brain images using a generative adversarial network to synthesize new facial features and contours. We employed the precision of the three-dimensional U-Net model to detect specific features such as the ears, nose, mouth, and eyes. Results: Our method diverges from prior studies by highlighting partial regions of the head image rather than comprehensive full-head images. We trained and tested our model on a dataset comprising 490 cases from a publicly available head computed tomography image dataset and an additional 70 cases with head MR images. Integrated data proved advantageous, with promising results. The nose, mouth, and eye detection achieved 100% accuracy, while ear detection reached 85.03% in the training dataset. In the testing dataset, ear detection accuracy was 65.98%, and the validation dataset ear detection attained 100%. Analysis of pixel value histograms demonstrated varying degrees of similarity, as measured by the Structural Similarity Index (SSIM), between raw and generated features across different facial features. The proposed methodology, tailored for partial head image processing, is well suited for real-world imaging examination scenarios and holds potential for future clinical applications contributing to the advancement of research in de-identification technologies, thus fortifying privacy safeguards.

Toward Head Computed Tomography Image Reconstruction Standardization With Deep-Learning-Assisted Automatic Detection

Toward Head Computed Tomography Image Reconstruction Standardization With Deep-Learning-Assisted Automatic Detection

Risk factors of prognosis in older patients with severe brain injury after surgical intervention

BackgroundOlder patients (aged ≥ 60 years) with severe brain injury have a high mortality and disability rate. The objective of this retrospective study was to assess the clinical risk factors of prognosis in older patients with severe brain injury after surgical intervention and to analyze the prognosis of the surviving group of patients 1 year after discharge.MethodsClinical data of older patients with severe brain injury who were admitted to two neurosurgical centers between January 2010 and December 2020 were collected. Patient age, sex, Glasgow Coma Scale (GCS) score at admission, underlying disease, mechanisms of injury, abnormal pupillary reflex, head computed tomography imaging findings (such as hematoma type),intraoperative brain swelling and other factors were reviewed. All the patients were categorized into a good prognosis (survival) group and a poor prognosis (death) group by the Glasgow Outcome Score (GOS); also, the related factors affecting the prognosis were screened and the independent risk factors were identified by the Binary logistic regression analysis. GOS was used to evaluate the prognosis of the surviving group of patients 1 year after discharge.ResultsOut of 269 patients, 171 (63.6%) survived, and 98 (36.4%) died during hospitalization. Univariate analysis showed that age, GCS score at admission, underlying diseases, abnormal pupillary reflex, the disappearance of ambient cistern, the midline structure shift, intraoperative brain swelling, oxygen saturation < 90%, and cerebral hernia were risk factors for the prognosis of older patients with severe brain injury after surgical intervention. Multivariate analysis showed that age, underlying diseases, disappearance of ambient cistern, Oxygen saturation < 90% and intraoperative brain swelling were independent risk factors of the prognosis in the population. The effect of surgical intervention differed among various age groups at 1-year follow-up after surgery.ConclusionsThe results of this retrospective study confirmed that age, underlying diseases, disappearance of ambient cistern, intraoperative brain swelling, and oxygen saturation < 90% are associated with poor prognosis in older postoperative patients with severe brain injury. Surgical intervention may improve prognosis and reduce mortality in older patients (age < 75 years). But for those patients (age ≥ 75 years), the prognosis was poor after surgical intervention.

Anatomic Variations of Foramen Ovale as a Predictor of Successful Cannulation in Percutaneous Trigeminal Rhizotomies.

Percutaneous trigeminal rhizotomies are common treatment modalities for medically refractory trigeminal neuralgia (TN). Failure of these procedures is frequently due to surgical inability to cannulate the foramen ovale (FO) and is thought to be due to variations in anatomy. The purpose of this study is to characterize the relationships between anatomic features surrounding FO and investigate the association between anatomic morphology and successful cannulation of FO in patients undergoing percutaneous trigeminal rhizotomy. A retrospective analysis was conducted of all patients undergoing percutaneous trigeminal rhizotomy for TN at our academic center between January 1, 2010, and July 31, 2022. Preoperative 1-mm thin-cut computed tomography head imaging was accessed to perform measurements surrounding the FO, including inlet width, outlet width, interforaminal distance (a representation of the lateral extent of FO along the middle fossa), and sella-sphenoid angle (a representation of the coronal slope of FO). Mann-Whitney U tests assessed the difference in measurements for patients who succeeded and failed cannulation. Among 37 patients who met inclusion criteria, 34 (91.9%) successfully underwent cannulation. Successful cannulation was associated with larger inlet widths (median = 5.87 vs 3.67 mm, U = 6.0, P = .006), larger outlet widths (median = 7.13 vs 5.10 mm, U = 14.0, P = .040), and smaller sella-sphenoid angles (median = 52.00° vs 111.00°, U = 0.0, P < .001). Interforaminal distances were not associated with the ability to cannulate FO surgically. We have identified morphological characteristics associated with successful cannulation in percutaneous rhizotomies for TN. Preoperative imaging may optimize surgical technique and predict cannulation failure.

Reproducibility of Facial Information in Three-Dimensional Reconstructed Head Images: An Exploratory Study.

Facial information acquired via three-dimensional reconstruction of head computed tomography (CT) data may be considered personal information, which can be problematic for neuroimaging studies. However, no study has verified the relationship between slice thickness and face reproducibility. This study determined the relationship and match rate between image slice thickness and face detection accuracy of face-recognition software in facial reconstructed models. Head CT data of 60 cases comprising entire faces obtained under conditions of non-contrast and 1-mm slice thickness were resampled to obtain 2-10-mm slice-thickness data. Facial models, reconstructed by image thresholding, were acquired from the data. We performed face detection tests per slice thickness on the models and calculated the face detection rate. The reconstructed facial models created from 1-mm slice-thickness data and other slice thicknesses were used as training and test data, respectively. Match confidence scores were obtained via three programs, match rates were calculated per slice thickness, and generalized estimating equations were used to evaluate the match rate trend. In general, the face detection rates for the 1-10-mm slice thicknesses were 100, 100, 98.3, 98.3, 95.0, 91.7, 86.7, 78.3, 68.3, and 61.7 %, respectively. The match rates for the 2-10-mm slice thicknesses were 100, 98.3, 98.3, 95.0, 85.0, 71.7, 53.3, 28.3, and 16.7 %, respectively. The reconstructed models tended to have higher match rates as the slice thickness decreased. Thus, thin-slice head CT imaging data may increase the possibility of the information becoming personally identifiable health information.

Association between Bilateral Selective Antegrade Cerebral Perfusion and Postoperative Ischemic Stroke in Patients with Emergency Surgery for Acute Type A Aortic Dissection-Single Centre Experience.

Acute type A aortic dissection (ATAAD) is a surgical emergency with a mortality of 1-2% per hour. Since its discovery over 200 years ago, surgical techniques for repairing a dissected aorta have evolved, and with the introduction of hypothermic circulatory arrest and cerebral perfusion, complex techniques for replacing the entire aortic arch were possible. However, postoperative neurological complications contribute significantly to mortality in this group of patients. The aim of this study was to determine the association between different bilateral selective antegrade cerebral perfusion (ACP) times and the incidence of postoperative ischemic stroke in patients with emergency surgery for ATAAD. Patients with documented hemorrhagic or ischemic stroke, clinical signs of stroke or neurological dysfunction prior to surgery, that died on the operating table or within 48 h after surgery, from whom the postoperative neurological status could not be assessed, and with incomplete medical records were excluded from this study. The diagnosis of postoperative stroke was made using head computed tomography imaging (CT) when clinical suspicion was raised by a neurologist in the immediate postoperative period. For selective bilateral antegrade cerebral perfusion, we used two balloon-tipped cannulas inserted under direct vision into the innominate artery and the left common carotid artery. Each cannula is connected to a separate pump with an independent pressure line. Near-infrared spectroscopy was used in all cases for cerebral oxygenation monitoring. The circulatory arrest was initiated after reaching a target core temperature of 25-28 °C. In total, 129 patients were included in this study. The incidence of postoperative ischemic stroke documented on a head CT was 24.8% (31 patients), and postoperative death was 20.9% (27 patients). The most common surgical technique performed was supravalvular ascending aorta and Hemiarch replacement with a Dacron graft in 69.8% (90 patients). The mean cardiopulmonary bypass time was 210 +/- 56.874 min, the mean aortic cross-clamp time was 114.775 +/- 34.602 min, and the mean cerebral perfusion time was 37.837 +/- 18.243 min. Using logistic regression, selective ACP of more than 40 min was independently associated with postoperative ischemic stroke (OR = 3.589; 95%CI = 1.418-9.085; p = 0.007). Considering the high incidence of postoperative stroke in our study population, we concluded that bilateral selective ACP should be used with caution, especially in patients with severely calcified ascending aorta and/or aortic arch and supra-aortic vessels. All efforts should be made to minimize the duration of circulatory arrest when using bilateral selective ACP with a target of less than 30 min, in hypothermia, at a body temperature of 25-28 °C.

Trade-off between the radiation parameters and image quality using iterative reconstruction techniques in head computed tomography: a phantom study.

Iterative reconstruction techniques (IRTs) are commonly used in computed tomography (CT) and help to reduce image noise. To determine the minimum radiation dose while preserving image quality in head CT using IRTs. The anthropomorphic phantom was used to scan nine head CT image series with varied radiation parameters. CT dose parameters, including volume CT dose index (CTDIvol [in mGy]) and dose length product (DLP [in mGy/cm]), were recorded for each scan series. Different noise levels (iDoseL1-6) were used in IRT reconstructions for soft and bone tissues. In total, 15 measurements were taken from five regions of interest (ROI) with an area of 10 mm2. The signal-to-noise ratio (SNR) and noise values obtained at different ROIs were compared among various reconstruction methods with repeated measures of statistical analysis. In the head CT scan, applying IRT iDoseL5 had the lowest noise and highest SNR for soft tissue (P < 0.05), and increased iDose can decrease CT dose by 54.6% without compromising image quality. While for bone tissue reconstruction, no clear association was found between the level of iDose and noise. However, when CTDIvol is >20 mGy, iDoseL4 is slightly superior to other reconstruction methods (P < 0.065). Using IRTs in head CTs reduces radiation dose while maintaining image quality. IDoseL5 provided optimal balance for soft tissue.

Extraction of Radiological Characteristics From Free-Text Imaging Reports Using Natural Language Processing Among Patients With Ischemic and Hemorrhagic Stroke: Algorithm Development and Validation.

Neuroimaging is the gold-standard diagnostic modality for all patients suspected of stroke. However, the unstructured nature of imaging reports remains a major challenge to extracting useful information from electronic health records systems. Despite the increasing adoption of natural language processing (NLP) for radiology reports, information extraction for many stroke imaging features has not been systematically evaluated. In this study, we propose an NLP pipeline, which adopts the state-of-the-art ClinicalBERT model with domain-specific pretraining and task-oriented fine-tuning to extract 13 stroke features from head computed tomography imaging notes. We used the model to generate structured data sets with information on the presence or absence of common stroke features for 24,924 patients with strokes. We compared the survival characteristics of patients with and without features of severe stroke (eg, midline shift, perihematomal edema, or mass effect) using the Kaplan-Meier curve and log-rank tests. Pretrained on 82,073 head computed tomography notes with 13.7 million words and fine-tuned on 200 annotated notes, our HeadCT_BERT model achieved an average area under receiver operating characteristic curve of 0.9831, F1-score of 0.8683, and accuracy of 97%. Among patients with acute ischemic stroke, admissions with any severe stroke feature in initial imaging notes were associated with a lower probability of survival (P<.001). Our proposed NLP pipeline achieved high performance and has the potential to improve medical research and patient safety.

Should Infants with Blunt Traumatic Brain Injuries and Intracranial Hemorrhage Have Routine Repeat Imaging?

The practice of repeat head CT imaging in infants as a distinct population is poorly studied. The purpose of this study was to evaluate the incidence and utility of repeat head CT in the infant population. A 10-year retrospective review was conducted of infants with blunt traumatic head injuries (N = 50) that presented to a trauma center. Information from the hospital trauma registry and patient medical records were extracted regarding the size and type of injury, number and results of computed tomography (CT) imaging, changes in neurological exams, and any interventions that were required. Most patients (68%) had at least one repeat CT, with 26% showing progression of hemorrhage. Decreased Glasgow Coma Scale was associated with having repeat CT scans. Nearly one in four infants had a change in management associated with repeat imaging. Repeat CT scans resulted in operative interventions in 11.8% of cases and longer intensive care unit (ICU) stays in 8.8% of cases. Repeat CT scans were associated with increased hospital length of stay, but not with increased ventilator days, ICU length of stay, or mortality. Worsening bleeds were associated with mortality, but not with other hospital outcomes. Changes in management following repeat CT appeared to be more common in this population than in older children or adults. Findings from this study supported repeat CT imaging in infants, however, further research is needed to validate results of this study.

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.