Tuberculosis (TB) is an infectious disease that can lead towards death if left untreated. TB detection involves extraction of complex TB manifestation features such as lung cavity, air space consolidation, endobronchial spread, and pleural effusions from chest x-rays (CXRs). Deep learning based approach named convolutional neural network (CNN) has the ability to learn complex features from CXR images. The main problem is that CNN does not consider uncertainty to classify CXRs using softmax layer. It lacks in presenting the true probability of CXRs by differentiating confusing cases during TB detection. This paper presents the solution for TB identification by using Bayesian-based convolutional neural network (B-CNN). It deals with the uncertain cases that have low discernibility among the TB and non-TB manifested CXRs. The proposed TB identification methodology based on B-CNN is evaluated on two TB benchmark datasets, i.e., Montgomery and Shenzhen. For training and testing of proposed scheme we have utilized Google Colab platform which provides NVidia Tesla K80 with 12 GB of VRAM, single core of 2.3 GHz Xeon Processor, 12 GB RAM and 320 GB of disk. B-CNN achieves 96.42% and 86.46% accuracy on both dataset, respectively as compared to the state-of-the-art machine learning and CNN approaches. Moreover, B-CNN validates its results by filtering the CXRs as confusion cases where the variance of B-CNN predicted outputs is more than a certain threshold. Results prove the supremacy of B-CNN for the identification of TB and non-TB sample CXRs as compared to counterparts in terms of accuracy, variance in the predicted probabilities and model uncertainty. INDEX TERMS Tuberculosis identification, computer-aided diagnostics, medical image analysis, Bayesian convolutional neural networks, model uncertainty.
BackgroundAcute Fibrinous and Organizing Pneumonia is a rare entity characterized by the histological pattern suggestive of diffuse alveolar damage, eosinophilic pneumonia and organizing pneumonia; the presence of intra alveolar “fibrin balls” distinguishes it from these conditions. Herein, we describe the association of acute fibrinous and organizing pneumonia with a respiratory tract infection. We believe that such an association has been extremely rarely described.Case presentationWe report the case of a 68 year old female patient of Afghan descent who presented with shortness of breath, cough and high grade fever not responding satisfactorily to standard antibiotic therapy. Imaging revealed bilateral basilar infiltrates and ground glass opacification of the right lower lung zone. Though the inflammatory markers decreased with antibiotic therapy, there was minimal improvement in the patient’s symptoms and radiological appearance of the lungs. Bronchoscopy was refused by the patient’s family and a Computed Tomography guided biopsy of the lung revealed a histological diagnosis of acute fibrinous and organizing pneumonia. Patient was initiated on high dose intravenous corticosteroid therapy followed by a maintenance dose of prednisolone at 40 mg/day. She recovered dramatically. However, due to poor compliance with treatment, she relapsed and was re-treated with the same regimen. Currently she is completely symptom free and is on a tapering corticosteroid dose.ConclusionWe conclude that AFOP may be a rare but under diagnosed entity and recommend that it should be considered in the differentials of a suspected pulmonary infection unresponsive to optimum antibiotic therapy.
BackgroundMetastatic involvement of the spleen by solid tumors is a rare clinical entity; those coming from endometrial adenocarcinomas are exceptionally rare. Spleen is an uncommon site for metastatic deposits due to its specific anatomy and microenvironment. Typically, splenic metastasis from endometrial carcinomas present months to years after curative surgery, chemotherapy or radiotherapy. The most common complaint in symptomatic patients is abdominal pain localized to the left hypochondrium. Most however, are asymptomatic only to be picked up on vigilant routine ultrasonography or computerized tomography during follow up. We report the case of a 54-year-old woman who presented to us after 50 months of total abdominal hysterectomy and bilateral salpingo-oophorectomy for an endometrial adenocarcinoma. She had severe abdominal pain localized to the left hypochondrium as the presenting complaint. To the best of our knowledge, this is the 1st case to be reported from Pakistan with 14 cases reported prior to our report. All past cases report the endometroid variant of endometrial adenocarcinoma as the primary tumor and our patient was a victim to the same variant.Case presentationA 54-year-old, nulliparous widowed woman presented with severe abdominal pain in the left hypochondrium for the last 4 months. The pain radiated to the left shoulder and was exacerbated with deep breathing. She had a history of total abdominal hysterectomy with bilateral salpingo-oophorectomy done 50 months back for stage 1a endometroid endometrial adenocarcinoma. Clinical examination revealed tenderness in the left hypochondrium but no visceromeglay was appreciable. Ultrasonography and computerized tomography revealed a space-occupying lesion within the spleen with associated splenomegaly. Computed tomography further suggested a large splenic abscess however the patient did not have fever, vomiting or leukocytosis which are the hallmarks of a splenic abscess. A splenectomy was performed for her complaints. On histopathology a metastatic adenocarcinoma was identified consistent with the primary tumor. The tumor was CK7, CA-125 and epithelial membrane antigen positive (EMA). The patient was then referred for further chemotherapy.ConclusionFrom this case we conclude, that although very rare, the spleen is a potential site for metastasis in endometroid endometrial adenocarcinoma. Since most patients are asymptomatic, routine examinations and imaging can identify its presence and avoid complications. If the practice is employed with vigilance, we may expect the clinical event to be diagnosed more frequently. The standard treatment is a classic splenectomy followed by chemotherapy.
Restless legs syndrome (RLS) is a common neurological disorder in hemodialysis (HD) patients. It is associated with poor sleep and decreased quality of life. The precipitants for the disorder are still poorly understood. The condition has not been studied extensively in Pakistan, which has a vast majority of end-stage renal disease patients on maintenance HD. We aimed to determine the prevalence of this condition in patients attending HD units of the largest renal dialysis center in Northern Pakistan. We also strived to determine any associations with dialysis inadequacy and the total duration of HD. This was an observational study comprising 279 patients. RLS was diagnosed using the International Restless Leg Syndrome Study Group criteria. Dialysis adequacy was determined using the Urea Reduction Ratio and the Kt/V technique. The prevalence of RLS in this large HD population was 24%. Our results show that a longer duration and greater number of HD sessions were significantly associated with the development of RLS (p<0.05). Dialysis inadequacy was not associated with the development of the disorder. These results may indicate that the pro-inflammatory nature of hemodialysis may have a role in the pathophysiology of RLS in HD patients and prolonged exposure to it may make them more prone to developing the disorder.
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.