The classification of ship images has become a significant area of research within the remote sensing community due to its potential applications in maritime security, traffic monitoring, and ...environmental protection. Traditional monitoring methods like the Automated Identification System (AIS) and the Constant False Alarm Rate (CFAR) have their limitations, such as challenges with sea clutter and the problem of ships turning off their transponders. Additionally, classifying ship images in remote sensing is a complex task due to the spatial arrangement of geospatial objects, complex backgrounds, and the resolution limitations of sensor platforms. To address these challenges, this paper introduces a novel approach that leverages a unique dataset termed Heterogeneous Ship data and a new technique called the Spatial–Channel Attention with Bilinear Pooling Network (SCABPNet). First, we introduce the Heterogeneous Ship data, which combines Synthetic Aperture Radar (SAR) and optical satellite imagery, to leverage the complementary features of the SAR and optical modalities, thereby providing a richer and more-diverse set of features for ship classification. Second, we designed a custom layer, called the Spatial–Channel Attention with Bilinear Pooling (SCABP) layer. This layer sequentially applies the spatial attention, channel attention, and bilinear pooling techniques to enhance the feature representation by focusing on extracting informative and discriminative features from input feature maps, then classify them. Finally, we integrated the SCABP layer into a deep neural network to create a novel model named the SCABPNet model, which is used to classify images in the proposed Heterogeneous Ship data. Our experiments showed that the SCABPNet model demonstrated superior performance, surpassing the results of several state-of-the-art deep learning models. SCABPNet achieved an accuracy of 97.67% on the proposed Heterogeneous Ship dataset during testing. This performance underscores SCABPNet’s capability to focus on ship-specific features while suppressing background noise and feature redundancy. We invite researchers to explore and build upon our work.
To describe neuroimaging findings and to report the epidemiologic and clinical characteristics of patients with coronavirus disease 2019 (COVID-19) with neurologic manifestations.
In this ...retrospective multicenter study (11 hospitals), we included 64 patients with confirmed COVID-19 with neurologic manifestations who underwent a brain MRI.
The cohort included 43 men (67%) and 21 women (33%); their median age was 66 (range 20-92) years. Thirty-six (56%) brain MRIs were considered abnormal, possibly related to severe acute respiratory syndrome coronavirus. Ischemic strokes (27%), leptomeningeal enhancement (17%), and encephalitis (13%) were the most frequent neuroimaging findings. Confusion (53%) was the most common neurologic manifestation, followed by impaired consciousness (39%), presence of clinical signs of corticospinal tract involvement (31%), agitation (31%), and headache (16%). The profile of patients experiencing ischemic stroke was different from that of other patients with abnormal brain imaging: the former less frequently had acute respiratory distress syndrome (
= 0.006) and more frequently had corticospinal tract signs (
= 0.02). Patients with encephalitis were younger (
= 0.007), whereas agitation was more frequent for patients with leptomeningeal enhancement (
= 0.009).
Patients with COVID-19 may develop a wide range of neurologic symptoms, which can be associated with severe and fatal complications such as ischemic stroke or encephalitis. In terms of meningoencephalitis involvement, even if a direct effect of the virus cannot be excluded, the pathophysiology seems to involve an immune or inflammatory process given the presence of signs of inflammation in both CSF and neuroimaging but the lack of virus in CSF.
NCT04368390.
•In a multicentric cohort of 59 patients suffering from severe COVID-19, ASL perfusion was abnormal in 53, hypoperfusion occurred in 48 (with a predominant fronto-temporo-polar pattern), whereas ...hyperperfusion occurred in 9.•Post-contrast FLAIR leptomeningeal enhancement was the only parameter with a significant association with cerebral perfusion, as it was tightly linked to hyperperfusion.•Frontotemporal hypoperfusion was also found in quantitative subgroup analysis compared to healthy subjects.
Cerebral hypoperfusion has been reported in patients with COVID-19 and neurological manifestations in small cohorts. We aimed to systematically assess changes in cerebral perfusion in a cohort of 59 of these patients, with or without abnormalities on morphological MRI sequences.
Patients with biologically-confirmed COVID-19 and neurological manifestations undergoing a brain MRI with technically adequate arterial spin labeling (ASL) perfusion were included in this retrospective multicenter study. ASL maps were jointly reviewed by two readers blinded to clinical data. They assessed abnormal perfusion in four regions of interest in each brain hemisphere: frontal lobe, parietal lobe, posterior temporal lobe, and temporal pole extended to the amygdalo-hippocampal complex.
Fifty-nine patients (44 men (75%), mean age 61.2 years) were included. Most patients had a severe COVID-19, 57 (97%) needed oxygen therapy and 43 (73%) were hospitalized in intensive care unit at the time of MRI. Morphological brain MRI was abnormal in 44 (75%) patients. ASL perfusion was abnormal in 53 (90%) patients, and particularly in all patients with normal morphological MRI. Hypoperfusion occurred in 48 (81%) patients, mostly in temporal poles (52 (44%)) and frontal lobes (40 (34%)). Hyperperfusion occurred in 9 (15%) patients and was closely associated with post-contrast FLAIR leptomeningeal enhancement (100% 66.4%-100% of hyperperfusion with enhancement versus 28.6% 16.6%-43.2% without, p = 0.002). Studied clinical parameters (especially sedation) and other morphological MRI anomalies had no significant impact on perfusion anomalies.
Brain ASL perfusion showed hypoperfusion in more than 80% of patients with severe COVID-19, with or without visible lesion on conventional MRI abnormalities.