The brain-computer interface (BCI) is a cutting-edge technology that has the potential to change the world. Electroencephalogram (EEG) motor imagery (MI) signal has been used extensively in many BCI ...applications to assist disabled people, control devices or environments, and even augment human capabilities. However, the limited performance of brain signal decoding is restricting the broad growth of the BCI industry. In this article, we propose an attention-based temporal convolutional network (ATCNet) for EEG-based motor imagery classification. The ATCNet model utilizes multiple techniques to boost the performance of MI classification with a relatively small number of parameters. ATCNet employs scientific machine learning to design a domain-specific deep learning model with interpretable and explainable features, multihead self-attention to highlight the most valuable features in MI-EEG data, temporal convolutional network to extract high-level temporal features, and convolutional-based sliding window to augment the MI-EEG data efficiently. The proposed model outperforms the current state-of-the-art techniques in the BCI Competition IV-2a dataset with an accuracy of 85.38% and 70.97% for the subject-dependent and subject-independent modes, respectively.
To describe an infant with Adams Oliver syndrome (AOS) with ocular signs similar to familial exudative vitreoretinopathy.
A full-term female infant presented with a congenital scalp defect, ...hypoplasia of the fingers and toes along with a radial retinal fold in the right eye and tractional retinal detachment in the left eye. Fluorescein angiography findings included peripheral retinal nonperfusion, irregular vascular sprouting beyond the vascular-avascular junction, pinpoint areas of hyperfluorescence as well as late peripheral and posterior vascular leakage. The patient was clinically diagnosed with Adams Oliver syndrome based on the collective findings. Laser photocoagulation to the avascular retina was performed in both eyes which resulted in stabilization of the condition after 2 years of follow up.
The ocular phenotype in AOS may be similar to familial exudative vitreoretinopathy. Therefore, suspicion of the diagnosis should prompt ophthalmic evaluation including fluorescein angiography to detect and possibly treat the ischemic retinopathy.
Human activity recognition (HAR) remains a challenging yet crucial problem to address in computer vision. HAR is primarily intended to be used with other technologies, such as the Internet of Things, ...to assist in healthcare and eldercare. With the development of deep learning, automatic high-level feature extraction has become a possibility and has been used to optimize HAR performance. Furthermore, deep-learning techniques have been applied in various fields for sensor-based HAR. This study introduces a new methodology using convolution neural networks (CNN) with varying kernel dimensions along with bi-directional long short-term memory (BiLSTM) to capture features at various resolutions. The novelty of this research lies in the effective selection of the optimal video representation and in the effective extraction of spatial and temporal features from sensor data using traditional CNN and BiLSTM. Wireless sensor data mining (WISDM) and UCI datasets are used for this proposed methodology in which data are collected through diverse methods, including accelerometers, sensors, and gyroscopes. The results indicate that the proposed scheme is efficient in improving HAR. It was thus found that unlike other available methods, the proposed method improved accuracy, attaining a higher score in the WISDM dataset compared to the UCI dataset (98.53% vs. 97.05%).
Background/PurposeWhether YouTube videos contain precise and adequate information on certain orthodontic procedures remains unclear. This study aimed to investigate the content and quality of YouTube ...videos on orthodontic elastics and identify the predictors of high-level content YouTube videos.Materials and MethodsTwo hundred YouTube videos were screened for eligibility, and after applying the inclusion criteria, 133 videos were excluded. Student's t-test was used to compare the characteristics, quality parameters, and total content of the low-level and high-level content videos. Chi-square or Fisher's exact tests were implemented to identify the source and content element differences across low-level and high-level content videos. Pearson's correlation coefficients were used to determine the relationship between the total content score, video information and quality index (VIQI), and YouTube characteristics. Stepwise linear multiple regressions with forward selection were used to test the association of the YouTube characteristics and VIQI with the total content score.ResultsAmong 67 included videos, only 19.4% of videos were classified as high-level content videos. High-level content videos had significantly higher mean number of likes (MD = 4041.7; SD = 4680.7; P-value=0.0068), VIQI score (MD = 4.17; SD = 4.87; P-value=0.0073), and total content score (MD = 4.04; SD = 1.23; P-value=<0.0001). The adjusted linear regression model demonstrated a significant association between the total content score and VIQI, where 1 unit increase in the VIQI was significantly associated with a 0.16 increase in the total content score (B = 0.16; standard error SE=0.04; P = 0.0003). Further, a significant association was observed between the total content score and video duration, where 1 minute increase in the video duration was significantly associated with a 0.15 increase in the total content score (B = 0.15; SE = 0.05; P = 0.008).ConclusionThis study demonstrated that YouTube content quality concerning orthodontic elastics is poor. Thus, future implementation of online visual content provided by certified orthodontists will ensure accurate and thorough information delivery.
•Bitewing radiographs for proximal resin composite restorations performed by dental students were assessed for different procedural errors.•The most observed procedural errors were internal gaps at ...the bonding interface and internal voids.•Molars had 0.39 the risk of internal voids, 0.41 the risk of sharp angle, and 0.57 the risk of open contact compared to premolars.•This study reinforces the conclusions drawn from previous articles, which indicate a strong correlation between dental students and procedural errors in proximal resin composite restorations.•Dental schools may consider adopting new tools and approaches to teach proximal resin composite restoration placement effectively.
To identify the potential factors that induce procedural errors during posterior proximal resin composite restorations placed by dental students.
This retrospective study evaluated 803 bitewing radiographs of posterior proximal resin composite restorations placed by dental students at Imam Abdulrahman bin Faisal University. Atypical radiographic signs of failure were screened, and different patient-, operator-, and clinical-related factors were recorded. Chi-square test was used to examine the relationship between procedural errors and recorded factors. Stepwise adjusted logistic regression model was performed to identify predictors of procedural errors.
The most observed errors were internal gaps at the bonding interface and internal voids. Molars had 0.39 the risk of internal voids (odds ratio OR = 0.39; confidence interval CI = 0.25–0.60; P = <0.0001), 0.41 the risk of sharp angle (OR = 0.41; CI = 0.24–0.68; P = <0.001), and 0.57 the risk of open contact (OR = 0.57; CI = 0.34–0.97; P = 0.04) compared to premolars. Those who were >40 years of age had 1.79 the risk of overhang compared to younger patients (OR = 1.79; CI = 1.04–3.11; P = <0.04). First molars and premolars had 0.64 the risk of overhang compared to second molars and premolars (OR = 0.64; CI = 0.41–1.00; P = 0.04). Junior students had 1.97 the risk of internal gap compared to their senior counterparts (OR = 1.97; CI = 1.20–3.21; P = 0.008). Mesial restorations had 0.38 the risk of external gap compared to mesio-occluso-distal (MOD) restorations (OR = 0.38; CI = 0.19–0.78; P = 0.003). Restorations with a margin coronal to the cemento-enamel junction (CEJ) had 0.44 the risk of external gap compared to those restorations with a margin apical to the CEJ (OR = 0.44; CI = 0.29–0.66; P = <0.0001).
Our findings suggested a higher incidence of procedural errors in restoring premolars and MOD cavity preparations. Therefore, it is crucial to enhance the comprehensiveness of laboratory training and expose students to diverse clinical scenarios and various techniques.
Abstract
A recently developed
Phox2a::Cre
mouse line has been shown to capture anterolateral system (ALS) projection neurons. Here, we used this line to test whether Phox2a-positive cells represent a ...distinct subpopulation among lamina I ALS neurons. We show that virtually all lamina I Phox2a cells can be retrogradely labelled from injections targeted on the lateral parabrachial area (LPb), and that most of those in the cervical cord also belong to the spinothalamic tract. Phox2a cells accounted for ~ 50–60% of the lamina I cells retrogradely labelled from LPb or thalamus. Phox2a was preferentially associated with smaller ALS neurons, and with those showing relatively weak neurokinin 1 receptor expression. The Phox2a cells were also less likely to project to the ipsilateral LPb. Although most Phox2a cells phosphorylated extracellular signal-regulated kinases following noxious heat stimulation, ~ 20% did not, and these were significantly smaller than the activated cells. This suggests that those ALS neurons that respond selectively to skin cooling, which have small cell bodies, may be included among the Phox2a population. Previous studies have defined neurochemical populations among the ALS cells, based on expression of Tac1 or Gpr83. However, we found that the proportions of Phox2a cells that expressed these genes were similar to the proportions reported for all lamina I ALS neurons, suggesting that Phox2a is not differentially expressed among cells belonging to these populations. Finally, we used a mouse line that resulted in membrane labelling of the Phox2a cells and showed that they all possess dendritic spines, although at a relatively low density. However, the distribution of the postsynaptic protein Homer revealed that dendritic spines accounted for a minority of the excitatory synapses on these cells. Our results confirm that Phox2a-positive cells in lamina I are ALS neurons, but show that the
Phox2a::Cre
line preferentially captures specific types of ALS cells.
BACKGROUND Transfusion therapy has a well-established role in the management of several sickle cell disease (SCD)-related complications. Nevertheless, the benefits of transfusion must outweigh the ...possible risks, including iron overload, infections, and transfusion reactions. Alloimmunization is the underlying etiology of most delayed hemolytic transfusion reactions (DHTR). DHTR is often underestimated and underdiagnosed in sickle cell disease patients as it mimics a vaso-occlusive crisis in presentation. Alloimmunization to RBC antigens can be a serious complication of transfusion, which is of particular interest in individuals with SCD, as the occurrence rate is higher in this population. This complication represents a secondary immunological phenomenon that typically arises after the emergence of an alloantibody to which the patient had been previously sensitized to. CASE REPORT Here, we report 2 cases of delayed hemolytic transfusion reaction (DHTR) in which the patients showed evidence of alloimmunization from previous blood transfusions. The patients were managed with a variety of medications, including supportive treatments, utilization of immunosuppressive agents, and enhancement of erythropoiesis. Both patients had evidence of clinical and laboratory improvement following the management. CONCLUSIONS DHTR is considered one of the most deleterious complications of transfusion in SCD patients. The diagnosis and management of DHTR is very challenging, especially because it can present differently in this population. A high index of clinical suspicion is needed in addition to the laboratory criteria.
Severe acute respiratory syndrome coronavirus 2 may inflict a post‐viral condition known as post‐COVID‐19 syndrome (PCS) or long‐COVID. Studies measuring levels of inflammatory and vascular ...biomarkers in blood, serum, or plasma of COVID‐19 survivors with PCS versus non‐PCS controls have produced mixed findings. Our review sought to meta‐analyse those studies. A systematic literature search was performed across five databases until 25 June 2022, with an updated search on 1 November 2022. Data analyses were performed with Review Manager and R Studio statistical software. Twenty‐four biomarkers from 23 studies were meta‐analysed. Higher levels of C‐reactive protein (Standardized mean difference (SMD) = 0.20; 95% CI: 0.02–0.39), D‐dimer (SMD = 0.27; 95% CI: 0.09–0.46), lactate dehydrogenase (SMD = 0.30; 95% CI: 0.05–0.54), and leukocytes (SMD = 0.34; 95% CI: 0.02–0.66) were found in COVID‐19 survivors with PCS than in those without PCS. After sensitivity analyses, lymphocytes (SMD = 0.30; 95% CI: 0.12–0.48) and interleukin‐6 (SMD = 0.30; 95% CI: 0.12–0.49) were also significantly higher in PCS than non‐PCS cases. No significant differences were noted in the remaining biomarkers investigated (e.g., ferritin, platelets, troponin, and fibrinogen). Subgroup analyses suggested the biomarker changes were mainly driven by PCS cases diagnosed via manifestation of organ abnormalities rather than symptomatic persistence, as well as PCS cases with duration of <6 than ≥6 months. In conclusion, our review pinpointed certain inflammatory and vascular biomarkers associated with PCS, which may shed light on potential new approaches to understanding, diagnosing, and treating PCS.
PURPOSE: To evaluate the incidence of pediatric Haemophilus influenzae related orbital cellulitis after the routine administration of the H. influenzae vaccine in Saudi Arabia and to identify the ...most common pathogens, predisposing factors, related complications, and the need for surgical intervention.
METHODS: A retrospective chart review was performed of patients with a discharge diagnosis of orbital cellulitis who presented to the King Abdul Aziz University Hospital, Riyadh, Saudi Arabia. Saudi children born between 2000 and 2016 were included.
RESULTS: Thirty-one cases were included and 8 had positive cultures. No cases of H. influenzae were found. Of all the positive culture, the most common pathogens isolated were Staphylococcus aureus, Streptococcus pneumoniae and group A Streptococcus in 3 cases (37.5%), 2 cases (25%) and 2 cases (25%). The most common predisposing factor was sinusitis in 90.3% of cases. Sixteen patients (51.6%) developed complications. Fifteen patients (48.3%) developed subperiosteal abscesses, 1 patient (3.22%) developed a lid abscess, 2 patients (6.45%) developed recurrent orbital cellulitis, and 1 patient (3.22%) developed persistent strabismus. Only 10 (31.25%) patients underwent surgical intervention during admission.
CONCLUSION: The vaccine seems to be effective at eliminating the orbital cellulitis potentially related to H. influenzae. The most common current pathogens were Gram-positive Staphylococcus and Streptococcus species. Sinusitis remains the most common cause of orbital cellulitis in children. In select cases, medical treatment can resolve the infections however, close observation is warranted.
Electroencephalography (EEG) motor imagery (MI) signals have recently gained a lot of attention as these signals encode a person’s intent of performing an action. Researchers have used MI signals to ...help disabled persons, control devices such as wheelchairs and even for autonomous driving. Hence decoding these signals accurately is important for a Brain–Computer interface (BCI) system. But EEG decoding is a challenging task because of its complexity, dynamic nature and low signal to noise ratio. Convolution neural network (CNN) has shown that it can extract spatial and temporal features from EEG, but in order to learn the dynamic correlations present in MI signals, we need improved CNN models. CNN can extract good features with both shallow and deep models pointing to the fact that, at different levels relevant features can be extracted. Fusion of multiple CNN models has not been experimented for EEG data. In this work, we propose a multi-layer CNNs method for fusing CNNs with different characteristics and architectures to improve EEG MI classification accuracy. Our method utilizes different convolutional features to capture spatial and temporal features from raw EEG data. We demonstrate that our novel MCNN and CCNN fusion methods outperforms all the state-of-the-art machine learning and deep learning techniques for EEG classification. We have performed various experiments to evaluate the performance of the proposed CNN fusion method on public datasets. The proposed MCNN method achieves 75.7% and 95.4% on the BCI Competition IV-2a dataset and the High Gamma Dataset respectively. The proposed CCNN method based on autoencoder cross-encoding achieves more than 10% improvement for cross-subject EEG classification.
•Multi CNN models with different layers and filters for robust EEG feature extraction.•Fusion model for merging multiple CNNs for EEG classification.•Use of transfer learning and pretraining to further improve EEG decoding accuracy.•Autoencoders cross-subject feature reconstruction to achieve better results.