An Optimized Deep Learning Model for Automatic Diagnosis of COVID-19 Using Chest X-Ray Images

COVID-19 has caused havoc throughout the world in the last two years by infecting over 455 million people. Development of automatic diagnosis software tools for rapid screening of COVID-19 via clinical imaging such as X-ray is vital to combat this pandemic. An optimized deep learning model is designed in this paper to perform automatic diagnosis on the chest X-ray (CXR) images of patients and classify them into normal, pneumonia and COVID-19 cases. A convolutional neural network (CNN) is employed in optimized deep learning model given its excellent performances in feature extraction and classification. A particle swarm optimization with multiple chaotic initialization scheme (PSOMCIS) is also designed to fine tune the hyperparameters of CNN, ensuring the proper training of network. The proposed deep learning model, namely PSOMCIS-CNN, is evaluated using a public database consists of the CXR images with normal, pneumonia and COVID-19 cases. The proposed PSOMCIS-CNN is revealed to have promising performances for automatic diagnosis of COVID-19 cases by producing the accuracy, sensitivity, specificity, precision and F1 score values of 97.78%, 97.77%, 98.8%, 97.77% and 97.77%, respectively. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

A Case of Hypophysitis Associated With SARS-CoV2 Vaccination

Objective: Although SARS-CoV2 vaccines have been developed with multiple novel technologies and rapidly disseminated worldwide, the full profile of adverse effects has not been known. Recently, there are sporadic but increasing reports of endocrinopathies in relation to SARS-CoV2 vaccination. Here, we report a rare case of hypophysitis with acute onset of diabetes insipidus after SARS-CoV2 vaccination. Case Report: A 48-year-old female who had been in her usual state of health until she received the first SARS-CoV2 vaccine. Two days after the vaccination, she started to have flu-like symptoms including severe headache and myalgia as well as persistent headache, polydipsia and polyuria. On presentation, her vital signs were stable, with her pulse being 81 bpm, BP: 122/84 mm Hg, temperature: 98.6F. Her physical examination was unremarkable. Laboratory evaluation revealed normal values of the basic metabolic panel including sodium level of 142 mmol/L and normal complete blood count. Due to her prolonged and worsening headache, she underwent a brain MRI which revealed a 4 mm round shape of thickening pituitary stalk and partial empty sella. The polydipsia, polyuria, and the thickening of the pituitary stalk led to further pituitary work-up. She underwent the overnight water deprivation test followed by the desmopressin challenge test. Her overnight water deprivation test revealed hypernatremia (Na 147 mmol/L), elevated serum osmolarity (309 mmol/kg), and low urinary osmolarity (83 mmol/kg) which were compatible with diabetes insipidus. Her IGF1 level revealed low normal range (66 ng/mL). Her 250 mcg cosyntropin test showed appropriate response without adrenal insufficiency. DDAVP was started. 3 months after the vaccination, her symptoms have partially improved, and on repeating the MRI brain she has persistent pituitary stalk thickening. Discussion: We report a rare case of diabetes insipidus from hypophysitis associated with SARS-CoV2 vaccine. Mechanisms of SARS-CoV2 vaccination-associated endocrinopathy is unknown. From our literature search, we found increasing numbers of the cases of endocrinopathy reported after the SARS- CoV2 vaccination. The thyroid seems the most frequently reported endocrine organ (83%), followed by the pituitary (11%) and adrenal (6%). Average onset is 1-5 days after the vaccinations and reported with all types of SARS-CoV2 vaccines. More mid-age (average age 46) female (78%) cases have been reported. Although associations are not confirmed, endocrinopathies may be underestimated in the post vaccinated population. Further studies are warranted to better understand SARS-CoV2 vaccinations and potential associations of endocrinopathy.Presentation: Monday, June 13, 2022 12:30 p.m. - 2:30 p.m.

Detection of COVID-19 Using Deep Convolutional Neural Network on Chest X-Ray (CXR) Images

Our study aims to investigate the best performing Convolutional Neural Networks (CNN) suitable for COVID-19 detection on Chest X-Ray (CXR) images. We applied five state-of-art CNN models in this study: DarkNet-19, ResNet-101, SqueezeNet, VGG-16, and VGG-19. These CNN models were pre-trained with natural images for classification. Therefore, we used transfer learning to modify the fully connected layer and output layer for a binary classification between COVID-19 and normal lungs. The models were trained using our combined dataset of CXR images obtained from the public domain, COVIDx, and private domain, University of Malaya (UM). The CXR images were pre-processed with reflection along the horizontal and vertical axis before being fed into the CNN models. Then another combined dataset from both COVIDx and UM was used to test the performance of the models. The numbers of correctly and wrongly predicted classes were tallied and represented with a confusion matrix. Then, the specificity, sensitivity, precision, F1-score, and accuracy were measured to evaluate the performance of each model. Our study demonstrated an accuracy above 90% for all five models. Gradient-weighted Class Activation Mapping (Grad-CAM) was used to visualize the significant activation regions that contributed to the model's decision. We have also applied the COVID-Net-CXR-Large model to our combined dataset for testing to evaluate its performance in multiclass classification. The current CNN models require further improvement and modification before they can be applied clinically as a secondary tool for the diagnosis of COVID-19 cases. © 2021 IEEE.