Deep learning-based super-resolution of structural brain MRI at 1.5 T: application to quantitative volume measurement.

OBJECTIVE: This study investigated the feasibility of using deep learning-based super-resolution (DL-SR) technique on low-resolution (LR) images to generate high-resolution (HR) MR images with the aim of scan time reduction. The efficacy of DL-SR was also assessed through the application of brain volume measurement (BVM). MATERIALS AND METHODS: In vivo brain images acquired with 3D-T1W from various MRI scanners were utilized. For model training, LR images were generated by downsampling the original 1 mm-2 mm isotropic resolution images. Pairs of LR and HR images were used for training 3D residual dense net (RDN). For model testing, actual scanned 2 mm isotropic resolution 3D-T1W images with one-minute scan time were used. Normalized root-mean-square error (NRMSE), peak signal-to-noise ratio (PSNR), and structural similarity (SSIM) were used for model evaluation. The evaluation also included brain volume measurement, with assessments of subcortical brain regions. RESULTS: The results showed that DL-SR model improved the quality of LR images compared with cubic interpolation, as indicated by NRMSE (24.22% vs 30.13%), PSNR (26.19 vs 24.65), and SSIM (0.96 vs 0.95). For volumetric assessments, there were no significant differences between DL-SR and actual HR images (p > 0.05, Pearson's correlation > 0.90) at seven subcortical regions. DISCUSSION: The combination of LR MRI and DL-SR enables addressing prolonged scan time in 3D MRI scans while providing sufficient image quality without affecting brain volume measurement.

Improved reconstruction for highly accelerated propeller diffusion 1.5 T clinical MRI.

PURPOSE: Propeller fast-spin-echo diffusion magnetic resonance imaging (FSE-dMRI) is essential for the diagnosis of Cholesteatoma. However, at clinical 1.5 T MRI, its signal-to-noise ratio (SNR) remains relatively low. To gain sufficient SNR, signal averaging (number of excitations, NEX) is usually used with the cost of prolonged scan time. In this work, we leveraged the benefits of Locally Low Rank (LLR) constrained reconstruction to enhance the SNR. Furthermore, we enhanced both the speed and SNR by employing Convolutional Neural Networks (CNNs) for the accelerated PROPELLER FSE-dMRI on a 1.5 T clinical scanner. METHODS: Residual U-Net (RU-Net) was found to be efficient for propeller FSE-dMRI data. It was trained to predict 2-NEX images obtained by Locally Low Rank (LLR) constrained reconstruction and used 1-NEX images obtained via simplified reconstruction as the inputs. The brain scans from healthy volunteers and patients with cholesteatoma were performed for model training and testing. The performance of trained networks was evaluated with normalized root-mean-square-error (NRMSE), structural similarity index measure (SSIM), and peak SNR (PSNR). RESULTS: For 4 × under-sampled with 7 blades data, online reconstruction appears to provide suboptimal images-some small details are missing due to high noise interferences. Offline LLR enables suppression of noises and discovering some small structures. RU-Net demonstrated further improvement compared to LLR by increasing 18.87% of PSNR, 2.11% of SSIM, and reducing 53.84% of NRMSE. Moreover, RU-Net is about 1500 × faster than LLR (0.03 vs. 47.59 s/slice). CONCLUSION: The LLR remarkably enhances the SNR compared to online reconstruction. Moreover, RU-Net improves propeller FSE-dMRI as reflected in PSNR, SSIM, and NRMSE. It requires only 1-NEX data, which allows a 2 × scan time reduction. In addition, its speed is approximately 1500 times faster than that of LLR-constrained reconstruction.

Comparative analysis between different volumetric methods on measuring intracranial hemorrhage incorporating roundness index.

Intracranial hematoma (ICH) volume is considered a predictor of clinical outcome and mortality rate in ICH patients with traumatic brain injury (TBI). The ABC/2 method for ICH volume is the standard method used to date, however, its level of accuracy has been questioned in some studies. This study compared the performance of the ABC/2 method with planimetry and truncated pyramidal methods to highlight the potential of the planimetry method applied with automatic segmentation for evaluation of epidural hematoma (EDH) and intraparenchymal hematoma (IPH) volume. Six different phantoms were designed to evaluate the accuracy of volume estimation methods. 221 hematoma regions extracted from CT scans of 125 patients with head injury were also used to analyze the efficiency. The roundness index was utilized for the quantification of the ellipsoid-like shape. Regions of EDH and IPH on the CT scans were annotated by radiologists. The estimation errors for each method were statistically analyzed and compared. In addition, the relationship between the errors and roundness index was examined. The planimetry method showed the lowest relative error on phantom data. In the case of the CT scan data, the truncated pyramidal method resulted in the underestimation of the volumes of EDH and IPH. Meanwhile, the ABC/2, through principal component analysis (PCA) in the two-dimensional and PCA in the three-dimensional methods, resulted in a significant overestimation. In addition, both these approaches produced relative errors that showed a correlation with the roundness indexes for IPH. In comparison to other methods, the planimetry method had the lowest level of error with regards to calculation of the volume and it was also independent of the hematoma shape. The planimetry method, therefore, has the potential to serve as a useful tool for the assessment of ICH volume in TBI patients by using a deep learning system.

The RSNA Cervical Spine Fracture CT Dataset.

This dataset is composed of cervical spine CT images with annotations related to fractures; it is available at https://www.kaggle.com/competitions/rsna-2022-cervical-spine-fracture-detection/.

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.

A Founder Intronic Variant in P3H1 Likely Results in Aberrant Splicing and Protein Truncation in Patients of Karen Descent with Osteogenesis Imperfecta Type VIII.

One of the most important steps in post-translational modifications of collagen type I chains is the hydroxylation of carbon-3 of proline residues by prolyl-3-hydroxylase-1 (P3H1). Genetic variants in P3H1 have been reported to cause autosomal recessive osteogenesis imperfecta (OI) type VIII. Clinical and radiographic examinations, whole-exome sequencing (WES), and bioinformatic analysis were performed in 11 Thai children of Karen descent affected by multiple bone fractures. Clinical and radiographic findings in these patients fit OI type VIII. Phenotypic variability is evident. WES identified an intronic homozygous variant (chr1:43212857A > G; NM_022356.4:c.2055 + 86A > G) in P3H1 in all patients, with parents in each patient being heterozygous for the variant. This variant is predicted to generate a new "CAG" splice acceptor sequence, resulting in the incorporation of an extra exon that leads to a frameshift in the final exon and subsequent non-functional P3H1 isoform a. Alternative splicing of P3H1 resulting in the absence of functional P3H1 caused OI type VIII in 11 Thai children of Karen descent. This variant appears to be specific to the Karen population. Our study emphasizes the significance of considering intronic variants.

Clinical and Genetic Studies of the First Monozygotic Twins with Pfeiffer Syndrome.

Objective: To report the clinical and radiographic findings and molecular etiology of the first monozygotic twins affected with Pfeiffer syndrome. Methods: Clinical and radiographic examination and whole exome sequencing were performed on two monozygotic twins with Pfeiffer syndrome. Results: An acceptor splice site mutation in FGFR2 (c.940-2A>G) was detected in both twins. The father and both twins shared the same haplotype, indicating that the mutant allele was from their father's chromosome who suffered severe upper airway obstruction and subsequent obstructive sleep apnea. Hypertrophy of nasal turbinates appears to be a newly recognized finding of Pfeiffer syndrome. Increased intracranial pressure in both twins were corrected early by fronto-orbital advancement with skull expansion and open osteotomy, in order to prevent the more severe consequences of increased intracranial pressure, including hydrocephalus, the bulging of the anterior fontanelle, and the diastasis of suture. Conclusions: Both twins carried a FGFR2 mutation and were discordant for lambdoid synostosis. Midface hypoplasia, narrow nasal cavities, and hypertrophic nasal turbinates resulted in severe upper airway obstruction and subsequent obstructive sleep apnea in both twins. Hypertrophy of the nasal turbinates appears to be a newly recognized finding of Pfeiffer syndrome. Fronto-orbital advancement with skull expansion and open osteotomy was performed to treat increased intracranial pressure in both twins. This is the first report of monozygotic twins with Pfeiffer syndrome.

Diagnostic Values of Red Flags and a Clinical Prediction Score for Emergent Intracranial Lesions in Non-Traumatic Pediatric Headaches.

INTRODUCTION: Diagnosis of emergent intracranial lesions that require emergency treatment either medically or surgically in non-traumatic pediatric headaches is important. Red-flag signs and symptoms are commonly used as justification for neuroimaging; however, evidence on its diagnostic values is limited. The study aims to identify diagnostic values of red-flags and develop a clinical prediction score to help improve the diagnostic yield of neuroimaging. METHODS: A retrospective review of 109 pediatric patients from 2006 to 2020 who presented with a non-traumatic headache was conducted. A clinical prediction score from red flags was developed using multivariate logistic regression. Discriminatory ability was examined using the area under the receiver operating characteristic curve. RESULTS: A total of 51 patients were diagnosed with emergent intracranial lesions. Four potential clinical red flag predictors were identified: (1) acute onset (less than 3 months), (2) altered conscious state, (3) focal motor abnormality, and (4) and ocular/pupillary abnormality or squint. A clinical prediction score was developed with good discriminatory properties (0.84). CONCLUSIONS: Clinical predictor scores from these four red flags may play an important role in maximizing neuroimaging and proper management for pediatric patients with non-traumatic headaches. Future validation studies are needed and could guide referrals and optimize the use of neuroimaging for these patients.

Diffuse neonatal hemangiomatosis with a single atypical cutaneous hemangioma.

Diffuse neonatal hemangiomatosis (DNH) is an extremely rare but deadly neonatal condition which presents as multiple cutaneous hemangiomas and hemangiomas in 3 or more visceral organs. DNH is usually suspected when multiple hemangiomas are found on the skin of the baby. We hereby present an interesting case in a newborn whose diagnosis was made from multiple intracranial, hepatic, and intramuscular hemangiomas, but with a single and unusual cutaneous manifestation over the right ankle. The patient was asymptomatic at the time of diagnosis. Due to the solitary nature of skin lesion, this report might contribute to a redefining of the term DNH.

Automatic hemorrhage segmentation on head CT scan for traumatic brain injury using 3D deep learning model.

The most common cause of long-term disability and death in young adults is a traumatic brain injury. The decision for surgical intervention for craniotomy is dependent on the injury type and the patient's neurologic exam. The potential subtypes of intracranial hemorrhage that may necessitate surgical intervention include subdural hemorrhage, epidural hemorrhage, and intraparenchymal hemorrhage. We proposed a novel automatic method for segmenting the hemorrhage subtypes on a CT scan by integrated CT scan with bone window as input of a deep learning model. Brain CT scans were collected from adult patients and annotated regions of subdural hemorrhage, epidural hemorrhage, and intraparenchymal hemorrhage by neuroradiologists. Their raw DICOM images were preprocessed by two different window settings i.e., subdural and bone windows. The collected CT scans were divided into two datasets namely training and test datasets. A deep-learning model was modified to segment regions of each hemorrhage subtype. The model is a three-dimensional convolutional neural network including four parallel pathways that process the input at different resolutions. It was trained by a training dataset. After the segmentation result was produced by the deep-learning model, it was then improved in the post-processing step. The size of the segmented lesion was considered, and a region-growing algorithm was applied. We evaluated the performance of the proposed method on the test dataset. The method reached the median Dice similarity coefficients higher than 0.37 for each hemorrhage subtype. The proposed method demonstrates higher Dice similarity coefficients and improved segmentation performance compared to previously published literature.

Circulating Lipocalin-2 level is positively associated with cognitive impairment in patients with metabolic syndrome.

The association between Lipocalin-2 (LCN2) and cognition in patients with metabolic syndrome (MetS) has not been thoroughly investigated. We aimed to evaluate whether serum LCN2 levels are associated with the alteration of cognitive function in patients with MetS. The total of 191 non-demented participants with MetS were enrolled onto the study in 2015, and a cohort study was conducted in a subpopulation in 2020. After adjustment for sex, age, waist circumference, creatinine levels, and HbA1C, an association between the higher serum LCN2 levels and the lower Montreal cognitive assessment (MoCA) scores was observed (B = - 0.045; 95%CI - 0.087, - 0.004; p 0.030). A total of 30 participants were followed-up in 2020. Serum LCN2 levels were decreased in correlation with age (23.31 ± 12.32 ng/ml in 2015 and 15.98 ± 11.28 ng/ml in 2020, p 0.024), while other metabolic parameters were unchanged. Magnetic resonance imaging studies were conducted on a subsample of patients in 2020 (n = 15). Associations between high serum LCN2 levels from 2015 and 2020 and changes in brain volume of hippocampus and prefrontal cortex from 2020 have been observed. These findings suggest a relationship between changes of the level of circulating LCN2, cognitive impairment, and changes in brain volume in patients with MetS. However, further investigation is still needed to explore the direct effect of circulating LCN2 on the cognition of MetS patients.

An optimal deep learning framework for multi-type hemorrhagic lesions detection and quantification in head CT images for traumatic brain injury.

Traumatic Brain Injury (TBI) could lead to intracranial hemorrhage (ICH), which has now been identified as a major cause of death after trauma if it is not adequately diagnosed and properly treated within the first 24 hours. CT examination is widely preferred for urgent ICH diagnosis, which enables the fast identification and detection of ICH regions. However, the use of it requires the clinical interpretation by experts to identify the subtypes of ICH. Besides, it is unable to provide the details needed to conduct quantitative assessment, such as the volume and thickness of hemorrhagic lesions, which may have prognostic importance to the decision-making on emergency treatment. In this paper, an optimal deep learning framework is proposed to assist the quantitative assessment for ICH diagnosis and the accurate detection of different subtypes of ICH through head CT scan. Firstly, the format of raw input data is converted from 3D DICOM to NIfTI. Secondly, a pre-trained multi-class semantic segmentation model is applied to each slice of CT images, so as to obtain a precise 3D mask of the whole ICH region. Thirdly, a fine-tuned classification neural network is employed to extract the key features from the raw input data and identify the subtypes of ICH. Finally, a quantitative assessment algorithm is adopted to automatically measure both thickness and volume via the 3D shape mask combined with the output probabilities of the classification network. The results of our extensive experiments demonstrate the effectiveness of the proposed framework where the average accuracy of 96.21 percent is achieved for three types of hemorrhage. The capability of our optimal classification model to distinguish between different types of lesion plays a significant role in reducing the false-positive rate in the existing work. Furthermore, the results suggest that our automatic quantitative assessment algorithm is effective in providing clinically relevant quantification in terms of volume and thickness. It is more important than the qualitative assessment conducted through visual inspection to the decision-making on emergency surgical treatment.

Feasibility of accelerated 3D T1-weighted MRI using compressed sensing: application to quantitative volume measurements of human brain structures.

OBJECTIVE: Scan time reduction is necessary for volumetric acquisitions to improve workflow productivity and to reduce motion artifacts during MRI procedures. We explored the possibility that Compressed Sensing-4 (CS-4) can be employed with 3D-turbo-field-echo T1-weighted (3D-TFE-T1W) sequence without compromising subcortical measurements on clinical 1.5 T MRI. MATERIALS AND METHODS: Thirty-three healthy volunteers (24 females, 9 males) underwent imaging scans on a 1.5 T MRI equipped with a 12-channel head coil. 3D-TFE-T1W for whole-brain coverage was performed with different acceleration factors, including SENSE-2, SENSE-4, CS-4. Freesurfer, FSL's FIRST, and volBrain packages were utilized for subcortical segmentation. All processed data were assessed using the Wilcoxon signed-rank test. RESULTS: The results obtained from SENSE-2 were considered as references. For SENSE-4, the maximum signal-to-noise ratio (SNR) drop was detected in the Accumbens (51.96%). For CS-4, the maximum SNR drop was detected in the Amygdala (10.55%). Since the SNR drop in CS-4 is relatively small, the SNR in all of the subcortical volumes obtained from SENSE-2 and CS-4 are not statistically different (P > 0.05), and their Pearson's correlation coefficients are larger than 0.90. The maximum biases of SENSE-4 and CS-4 were found in the Thalamus with the mean of differences of 1.60 ml and 0.18 ml, respectively. CONCLUSION: CS-4 provided sufficient quality of 3D-TFE-T1W images for 1.5 T MRI equipped with a 12-channel receiver coil. Subcortical volumes obtained from the CS-4 images are consistent among different post-processing packages.

A novel P3H1 mutation is associated with osteogenesis imperfecta type VIII and dental anomalies.

OBJECTIVE: Our objective was to investigate the molecular etiology of osteogenesis imperfecta type VIII and dental anomalies in 4 siblings of a Karen tribe family. MATERIALS AND METHODS: Four patients and their unaffected parents were studied by clinical and radiographic examination. In situ hybridization of P3h1 during early murine tooth development, whole-exome sequencing, and Sanger direct sequencing were performed. RESULTS: A novel homozygous missense P3H1 mutation (NM_001243246.1; c.2141A>G; NP_001230175.1; p.Lys714Arg) was identified in all patients. Their unaffected parents were heterozygous for the mutation. The mutation is hypothesized to belong to isoform c of P3H1. Mutations in P3H1 are associated with autosomal recessive osteogenesis imperfecta type VIII. Hypodontia, a mesiodens, and single-rooted permanent second molars found in our patients have never been reported in patients with P3H1 mutations. Single-rooted second permanent molars or failure to form multiple roots implies effects of the P3H1 mutation on root development. CONCLUSIONS: We report a novel P3H1 mutation as the underlying cause of osteogenesis imperfecta type VIII with dental anomalies. Our study suggests that isoform c of P3H1 is also a functional isoform of P3H1. We report, for the first time, to our knowledge, the association of P3H1 mutation and osteogenesis imperfecta type VIII with dental anomalies.

The relationship between external bony defects and widened lateral interorbital distance in frontoethmoidal encephalomeningocele.

PURPOSE: The aim of this study was to identify the prevalence of the abnormal lateral interorbital distance (LIOD), micropthalmia or anopthalmia and incidence of hydrocephalus in frontoethmoidal encephalomeningocele (FEEM), and to identify the correlation between the external bony defects and a widened lateral orbital wall. MATERIALS AND METHODS: FEEM patients who were treated in the Faculty of Medicine, Chiang Mai University between January 1, 2006, and December 31, 2016, were retrospectively reviewed from the medical records. We included the patients who had undergone head computed tomography (CT) of the facial bones. Exclusion criteria were any patients who had undergone prior facial surgery or had maxillofacial bone trauma before they underwent a CT scan, craniofacial disease such as facial clefts, frontonasal dysplasia, craniosynostosis, and inter-frontal encephalocele. Measurements were taken from the CT images to establish the diameter of the external bony defect, medial interorbital distance and lateral interorbital distance. The relationships between the external bony defect and widened LIOD were analyzed. Patients were divided into two groups: those with normal LIOD and those with wide LIOD. RESULTS: A total of 94 FEEM patients were evaluated. Thirteen patients (13.83%) had a wider than normal LIOD. The incidence of micropthalmia or anopthalmia was 9.57% and hydrocephalus was 45.74%. From a univariable model, there were no statistically significant differences in terms of patient characteristics between the LIOD groups, except for sex. The proportion of wide LIOD was higher in male patients (male vs female; 20.34% vs 2.86%, p = 0.027). The proportion between the external bony defect and the LIOD (EL ratio) associated with a wide LIOD was shown in a cutoff point of 23% (AuROC = 73.27%, 95%CI = 59.87%-86.66%). The multivariable analysis indicated factors associated with abnormal LIOD were an EL ratio of 23% (adjusted RR = 1.98, 95% CI = 1.58-2.51) and male sex (adjusted RR = 6.85, 95% CI = 2.00-23.55). CONCLUSION: There is an association between the EL ratio and widening of the LIOD. An EL ratio greater than 23% could be used to moderately predict a wide LIOD. A large external bony defect may be associated with lateral displacement of the orbit causing true hypertelorism. Therefore, orbital translocation must be considered in patients with a wide LIOD.

Sphenoid Sinus Diseases: A Review of 1,442 Patients.

OBJECTIVE: To review and report diseases of the sphenoid sinus from the literature and from a university hospital. METHODS: Inpatients' data were retrospectively gathered and reviewed from January 2006 to June 2016. Clinical data, imaging, organisms, and pathological reports were collected. Pathology was divided into infection/inflammation, tumor, and miscellaneous. A literature review was performed with the search term "isolated sphenoid disease" in PubMed. Original primary studies with 20 patients or more were reviewed. RESULTS AND DISCUSSION: One hundred and twenty-two patients were enrolled. Seventy-two subjects were female (59%). The average age was 54.3 years (±18.0). Imaging abnormalities were found incidentally in 27 patients (22.1%). The most common symptom was headache (63.9%). Visual loss, the second most common symptom, was more frequent in the tumor group (30.6% versus 54.2%). From the literature review, 21 primary studies with 1,320 total patients were included. From all studies and the present study, infection/inflammation was the most common pathology (75%) [95% confidence interval (CI): 0.696, 0.804]. Overall, tumors were found in 18.9% and malignant tumors in 7.0% [95% CI: 0.045, 0.095]. CONCLUSION: A specific diagnosis of a sphenoid lesion is needed during active investigation. Infection/inflammation was the most common pathology and malignancy was found in 7%.