Detecting COVID-19 early may help in devising an appropriate treatment plan and disease containment decisions. In this study, we demonstrate how transfer learning from deep learning models can be used to perform COVID-19 detection using images from three most commonly used medical imaging modes X-Ray, Ultrasound, and CT scan. The aim is to provide over-stressed medical professionals a second pair of eyes through intelligent deep learning image classification models. We identify a suitable Convolutional Neural Network (CNN) model through initial comparative study of several popular CNN models. We then optimize the selected VGG19 model for the image modalities to show how the models can be used for the highly scarce and challenging COVID-19 datasets. We highlight the challenges (including dataset size and quality) in utilizing current publicly available COVID-19 datasets for developing useful deep learning models and how it adversely impacts the trainability of complex models. We also propose an image pre-processing stage to create a trustworthy image dataset for developing and testing the deep learning models. The new approach is aimed to reduce unwanted noise from the images so that deep learning models can focus on detecting diseases with specific features from them. Our results indicate that Ultrasound images provide superior detection accuracy compared to X-Ray and CT scans. The experimental results highlight that with limited data, most of the deeper networks struggle to train well and provides less consistency over the three imaging modes we are using. The selected VGG19 model, which is then extensively tuned with appropriate parameters, performs in considerable levels of COVID-19 detection against pneumonia or normal for all three lung image modes with the precision of up to 86% for X-Ray, 100% for Ultrasound and 84% for CT scans.
A 24-year-old previously healthy girl presented with persistent fever, headache, and jaundice. Rapid-test anti-dengue virus IgM antibody was positive but anti-dengue IgG was nonreactive, which is suggestive of primary dengue infection. There was clinical deterioration during empiric antibiotic and symptomatic therapy. Bone marrow examination demonstrated the presence of hemophagocytosis. Diagnosis of dengue fever with virus-associated hemophagocytic syndrome was made according to the diagnostic criteria of the HLH 2004 protocol of the Histiocyte Society. The patient recovered with corticosteroid therapy. A review of literature revealed only a handful of case reports that showed the evidence that this syndrome is caused by dengue virus. Our patient is an interesting case of hemophagocytic syndrome associated with classic dengue fever and contributes an additional case to the existing literature on this topic. This case highlights the need for increased awareness even in infections not typically associated with hemophagocytic syndrome.
CONTEXT:Prevalence of tuberculous pleural effusion is very high in the Asian subcontinent but very few studies have come up from this part of the world about the course of recovery of pulmonary functions after institution of anti-tubercular therapy (ATT) and thoracentesis.AIMS:To study initial lung function impairment, changes over time after institution of ATT and thoracentesis and residual abnormalities left at the end of six months of treatment.SETTINGS AND DESIGN:Randomized open level interventional study over two years in 52 patients at a tertiary level teaching hospital.METHODS:The study population was divided into two equal groups, A (therapeutic thoracentesis) and B (diagnostic thoracentesis). Spirometry, chest radiograph and ultrasonography of thorax were done initially and at each follow-up visit up to six months. Statistical analysis was done (P value < 0.05 considered significant).RESULTS:Both groups were comparable initially. After six months none in group A and five patients in group B had minimal pleural effusion. During follow up, mean percentage predicted of FEV1 and FVC increased more in A than in B and the differences were statistically significant (P < 0.05). Pleural thickening, initially absent in both groups, was found to be more in B as compared to A at subsequent follow-up visits and this was statistically significant (P < 0.05).CONCLUSIONS:Thoracentesis should be considered in addition to anti-TB treatment, especially in large effusions, in order to relieve dyspnea, avoid possibility of residual pleural thickening and risk of developing restrictive functional impairment.
In human body, several categories of degenerative processes are largely determined by free radicals originating in cell. Free radicals are also known to have correlated with a variety of cognitive disorders (CDs) resulting in neuronal injury and eventually to death. Alzheimer’s disease (AD) and Parkinson's disease (PD) are such kind of killer CDs that occur due to dysfunction of cholinergic and dopaminergic neurons. Plant parts of Ginkgo biloba, Bacopa monnieri etc. are being used for the treatment of cognitive disorders in several countries. The present study was aimed to explore the detailed antioxidant and anti-cholinesterase activity of Acaciacatechu leaf (ACL) over CDs. Gas chromatography-Mass spectroscopy (GC-MS) analysis and Nuclear Magnetic Resonance (NMR) were employed to identify the bioactive components present in ACL. Furthermore, the extract was evaluated to check the cytotoxic effects of ACL on normal cells. Amongst several antioxidant assays, DPPH assay, hydroxyl radical, nitric oxide radical and hypochlorous acid inhibitory activities were found to be greater in ACL than that of the respective standards while other assays exhibited a moderate or at per inhibitory activity with standards. Total phenolic and flavonoid content were also found to be present in decent amount. In addition, we found, a greater acetylcholinesterase (AChE) inhibitory activity of ACL when compared to other medicinally important plants, indicating its positive effect over CDs. Forty one bioactive components were explored through GC-MS. Of these, gallic acid, epicatechin, catechin, isoquercitrin etc. were found, which are potent antioxidant and a few of them have anti-neurodegenerative properties. Eventually, ACL was found to be nontoxic and safer to consume. Further studies with animal or human model however, would determine its efficacy as a potential anti-schizophrenic drug.
Context:Non-resolving pneumonia is often an area of concern for pulmonologists. Fiber optic bronchoscopy (FOB) may have a special role in etiologic evaluation of non-resolving pneumonias. There is paucity of recent studies in this field.Aims:This study aimed to assess the patients of non-resolving or slowly resolving pneumonia with special emphasis on efficacy of FOB and computed tomography (CT)-guided fine needle aspiration cytology (FNAC) in diagnosis.Settings and Design:Prospective, observational study conducted in a tertiary care institute over a period of one year.Materials and Methods:After fulfilling the definition of non-resolving pneumonia by clinical and radiological parameters, patients were evaluated by FOB with relevant microbiological, cytological, histopathological investigations and CT scan of thorax. CT-guided FNAC was done in selected cases where FOB was inconclusive.Results:Sixty patients were enrolled in the study. Mean age was 51.33 ± 1.71 years with male to female ratio 2:1. Right lung was more commonly involved (65%), and right upper lobe was the commonest site (25%). Pyogenic infection was the commonest etiology (53.3%), bronchogenic carcinoma and tuberculosis accounted for 26.7% and 16.7% cases, respectively. Both, FOB (85.7%) and CT-guided FNAC (91.67%) were very useful for etiological diagnosis of non-resolving pneumonia. Both the procedures were safe, and no major complication was observed.Conclusions:Because of the high yield of FOB, it is very useful and safe diagnostic tool for evaluation of non-resolving pneumonia. CT-guided FNAC also gives good yield when cases are properly selected.
Background:Vivax malaria is the most widely distributed human malaria and is responsible for up to 400 million infections every year. Recently, it has become evident that Plasmodium vivax monoinfection could also result in multiple organ dysfunction and severe life-threatening disease as seen in Plasmodium falciparum infection.Materials and Methods:The aim of this study was to note the different clinical and biochemical profiles of adult patients with the severe vivax malaria with regards to complications and outcome. This was a prospective observational study carried out at a tertiary care hospital in Kolkata over 9 month's period. Detailed history and examination findings were noted in all patients. Their clinical presentations, complications, course in ward until discharge or death was noted.Results:A total of 900 cases of vivax malaria were included in the study. Severe disease was present in 200 (22.2%) cases of malaria. There were 108 (54%) patients with single complication (SC) and 92 (46%) patients with the multiple complications (MC). Patients with SC had jaundice (48.1%) followed by cerebral involvement (25.9%), renal failure (7.4%), and pulmonary involvement (3.7%). The MC was found in various combinations and the majority (47.8%) had constellation of two different complications. The mortality rate of patients with the SC and MC was 7.4% and 34.8%. The overall mortality observed in severe vivax malaria was 20% (40/200).Conclusions:In recent years, the clinical pattern of vivax malaria has changed. Severe vivax malaria is now very common with increasing mortality. Not only the number, but also the type of complication influences the outcome of complicated malaria.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.