Ensemble of Deep Capsule Neural Networks: An Application to Pneumonia Prediction

Pneumonia is the primary cause of death in children under the age of 5 years. Faster and more accurate laboratory testing aids in the prescription of appropriate treatment for children suspected of having pneumonia, lowering mortality. In this work, we implement a deep neural network model to efficiently evaluate pediatric pneumonia from chest radio graph images. Our network uses a combination of convolutional and capsule layers to capture abstract details as well as low level hidden features from the the radio graphic images, allowing the model to generate more generic predictions. Furthermore, we combine several capsule networks by stacking them together and connected them with dense layers. The joint model is trained as a single model using joint loss and the weights of the capsule layers are updated using the dynamic routing algorithm. The proposed model is evaluated using benchmark pneumonia dataset\cite{kermany2018identifying}, and the outcomes of our experimental studies indicate that the capsules employed in the network enhance the learning of disease level features that are essential in diagnosing pneumonia. According to our comparison studies, the proposed model with Convolution base from InceptionV3 attached with Capsule layers at the end surpasses several existing models by achieving an accuracy of 94.84\%. The proposed model is superior in terms of various performance measures such as accuracy and recall, and is well suited to real-time pediatric pneumonia diagnosis, substituting manual chest radiography examination.

Novel Privacy Preserving Non-Invasive Sensing-Based Diagnoses of Pneumonia Disease Leveraging Deep Network Model

This article presents non-invasive sensing-based diagnoses of pneumonia disease, exploiting a deep learning model to make the technique non-invasive coupled with security preservation. Sensing and securing healthcare and medical images such as X-rays that can be used to diagnose viral diseases such as pneumonia is a challenging task for researchers. In the past few years, patients’ medical records have been shared using various wireless technologies. The wireless transmitted data are prone to attacks, resulting in the misuse of patients’ medical records. Therefore, it is important to secure medical data, which are in the form of images. The proposed work is divided into two sections: in the first section, primary data in the form of images are encrypted using the proposed technique based on chaos and convolution neural network. Furthermore, multiple chaotic maps are incorporated to create a random number generator, and the generated random sequence is used for pixel permutation and substitution. In the second part of the proposed work, a new technique for pneumonia diagnosis using deep learning, in which X-ray images are used as a dataset, is proposed. Several physiological features such as cough, fever, chest pain, flu, low energy, sweating, shaking, chills, shortness of breath, fatigue, loss of appetite, and headache and statistical features such as entropy, correlation, contrast dissimilarity, etc., are extracted from the X-ray images for the pneumonia diagnosis. Moreover, machine learning algorithms such as support vector machines, decision trees, random forests, and naive Bayes are also implemented for the proposed model and compared with the proposed CNN-based model. Furthermore, to improve the CNN-based proposed model, transfer learning and fine tuning are also incorporated. It is found that CNN performs better than other machine learning algorithms as the accuracy of the proposed work when using naive Bayes and CNN is 89% and 97%, respectively, which is also greater than the average accuracy of the existing schemes, which is 90%. Further, K-fold analysis and voting techniques are also incorporated to improve the accuracy of the proposed model. Different metrics such as entropy, correlation, contrast, and energy are used to gauge the performance of the proposed encryption technology, while precision, recall, F1 score, and support are used to evaluate the effectiveness of the proposed machine learning-based model for pneumonia diagnosis. The entropy and correlation of the proposed work are 7.999 and 0.0001, respectively, which reflects that the proposed encryption algorithm offers a higher security of the digital data. Moreover, a detailed comparison with the existing work is also made and reveals that both the proposed models work better than the existing work.

The Oral Complications of COVID-19

Background: COVID-19 is a novel coronavirus infectious disease associated with the severe acute respiratory syndrome. More and more patients are being cured due to the development of clinical guidelines for COVID-19 pneumonia diagnosis, treatment, and vaccines. However, the long-term impact of COVID-19 on patients after recovery is unclear. Currently available reports have shown that patients recovered from COVID-19 continue to experience health problems in respiratory and other organ systems. Oral problem is one of the important complications which has serious impacts on the rehabilitation and future quality of life, such as ageusia and macroglossia, but the oral complication is often being neglected.Aim of Review: From the perspective of stomatology, we summarized and elaborated in detail the types, pathogenesis of oral complications from COVID-19 patients after rehabilitation, and the reported prevention or treatment recommendations which may improve the COVID-19 patients associated oral diseases.Key Scientific Concepts of Review: 1) To understand the common oral complications and the mechanisms of the development of oral complications after the COVID-19 recovery; 2) To summary the practical strategies to prevent the oral complications and construct the rehabilitation plans for patients with oral complications.

A training plan to implement lung ultrasound for diagnosing pneumonia in children

Background Lung ultrasound (LUS) for critical patients requires trained operators to perform them, though little information exists on the level of training required for independent practice. The aims were to implement a training plan for diagnosing pneumonia using LUS and to analyze the inter-observer agreement between senior radiologists (SRs) and pediatric intensive care physicians (PICPs). Methods Prospective longitudinal and interventional study conducted in the Pediatric Intensive Care Unit of a tertiary hospital. Following a theoretical and practical training plan regarding diagnosing pneumonia using LUS, the concordance between SRs and the PICPs on their LUS reports was analyzed. Results Nine PICPs were trained and tested on both theoretical and practical LUS knowledge. The mean exam mark was 13.5/15. To evaluate inter-observer agreement, a total of 483 LUS were performed. For interstitial syndrome, the global Kappa coefficient (K) was 0.51 (95% CI 0.43–0.58). Regarding the presence of consolidation, K was 0.67 (95% CI 0.53–0.78), and for the consolidation pattern, K was 0.82 (95% CI 0.79–0.85), showing almost perfect agreement. Conclusions Our training plan allowed PICPs to independently perform LUS and might improve pneumonia diagnosis. We found a high inter-observer agreement between PICPs and SRs in detecting the presence and type of consolidation on LUS. Impact Lung ultrasound (LUS) has been proposed as an alternative to diagnose pneumonia in children. However, the adoption of LUS in clinical practice has been slow, and it is not yet included in general clinical guidelines. The results of this study show that the implementation of a LUS training program may improve pneumonia diagnosis in critically ill patients. The training program’s design, implementation, and evaluation are described. The high inter-observer agreement between LUS reports from the physicians trained and expert radiologists encourage the use of LUS not only for pneumonia diagnosis, but also for discerning bacterial and viral patterns.

Validation of expert system enhanced deep learning algorithm for automated screening for COVID-Pneumonia on chest X-rays

SARS-CoV2 pandemic exposed the limitations of artificial intelligence based medical imaging systems. Earlier in the pandemic, the absence of sufficient training data prevented effective deep learning (DL) solutions for the diagnosis of COVID-19 based on X-Ray data. Here, addressing the lacunae in existing literature and algorithms with the paucity of initial training data; we describe CovBaseAI, an explainable tool using an ensemble of three DL models and an expert decision system (EDS) for COVID-Pneumonia diagnosis, trained entirely on pre-COVID-19 datasets. The performance and explainability of CovBaseAI was primarily validated on two independent datasets. Firstly, 1401 randomly selected CxR from an Indian quarantine center to assess effectiveness in excluding radiological COVID-Pneumonia requiring higher care. Second, curated dataset; 434 RT-PCR positive cases and 471 non-COVID/Normal historical scans, to assess performance in advanced medical settings. CovBaseAI had an accuracy of 87% with a negative predictive value of 98% in the quarantine-center data. However, sensitivity was 0.66–0.90 taking RT-PCR/radiologist opinion as ground truth. This work provides new insights on the usage of EDS with DL methods and the ability of algorithms to confidently predict COVID-Pneumonia while reinforcing the established learning; that benchmarking based on RT-PCR may not serve as reliable ground truth in radiological diagnosis. Such tools can pave the path for multi-modal high throughput detection of COVID-Pneumonia in screening and referral.

Did the National Lockdown in Saudi Arabia Reduce Lower Respiratory Illnesses in Children?

Objectives: This study aims to explore the effect of lockdown and early precautionary measures implemented in Saudi Arabia on number of pediatric hospitalizations due to lower respiratory illnesses (bronchiolitis, asthma, and pneumonia).Methods: This is a retrospective cross-sectional study aims to review patients from four major hospitals in Saudi Arabia. All pediatric hospitalizations secondary to asthma, bronchiolitis, and pneumonia during the months of the lockdown (March, April, and May) in 2020 were documented. Then, they were compared to the previous 2 years. Variables like number of hospitalizations, oxygen requirement, mechanical ventilation, admission to the intensive care unit (ICU), length of stay, and results of viral studies were collected.Results: We included 1,003 children from four different centers. Males were slightly higher than females (55.8% vs. 44.2%). Total number of hospitalizations in 2020 was 201, significantly lower than 399 and 403 hospitalizations in 2019 and 2018, respectively (P < 0.01). The major drop happened on the months of April and May. Although bronchiolitis hospitalizations' dropped by more than half in 2020 compared to the previous 2 years, it was not statistically significant (P = 0.07). But, asthma hospitalizations were significantly less in 2020 compared to the previous 2 years (49–65% reduction, P = 0.003). Number of pneumonia cases were lowered in 2020 compared to the previous 2 years. However, proportion of pneumonia diagnosis to total hospitalizations increased in 2020 (55% compared to 50% and 35%). There was a surge of viral testing during a period of uncertainty in the early phase of the pandemic. This total reduction in hospitalization was not associated with higher oxygen requirements, mechanical ventilation, ICU admissions or longer hospital stay.Conclusions: Lockdown and precautionary measures executed during the early phase of COVID-19 pandemic helped decrease the number of hospitalizations due to lower respiratory illnesses in Saudi Arabia. Reduction in hospitalizations seems less likely to be secondary to hospital avoidance or delayed presentations as number of ICU admission and oxygen requirements did not increase. The post pandemic pattern of respiratory illnesses among children needs further research.

Discordant diagnostic criteria for pneumonia in COPD trials: a review

Inhaled corticosteroids (ICS) have a class effect of increasing pneumonia risk in patients with COPD. However, pneumonia incidence varies widely across clinical trials of ICS use in COPD. This review clarifies methodological differences in defining and recording pneumonia events in these trials and discusses factors that could contribute to the varying pneumonia incidence. Literature searches and screening yielded 40 relevant references for inclusion. Methods used to capture pneumonia events in these studies included investigator-reported pneumonia adverse events, standardised list of signs or symptoms, radiographic confirmation of suspected cases and/or confirmation by an independent clinical end-point committee. In general, more stringent pneumonia diagnosis criteria led to lower reported pneumonia incidence rates. In addition, studies varied in design and population characteristics, including exacerbation history and lung function, factors that probably contribute to the varying pneumonia incidence. As such, cross-trial comparisons are problematic. A minimal set of standardised criteria for diagnosis and reporting of pneumonia should be used in COPD studies, as well as reporting of patients’ pneumonia history at baseline, to allow comparison of pneumonia rates between trials. Currently, within-trial comparison of ICS-containing versus non-ICS-containing treatments is the appropriate method to assess the influence of ICS on pneumonia incidence.