skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Optimized CNN-based Diagnosis System to Detect the Pneumonia from Chest Radiographs
Pneumonia is a high mortality disease that kills 50, 000 people in the United States each year. Children under the age of 5 and older population over the age of 65 are susceptible to serious cases of pneumonia. The United States spend billions of dollars fighting pneumonia-related infections every year. Early detection and intervention are crucial in treating pneumonia related infections. Since chest x-ray is one of the simplest and cheapest methods to diagnose pneumonia, we propose a deep learning algorithm based on convolutional neural networks to identify and classify pneumonia cases from these images. For all three models implemented, we obtained varying classification results and accuracy. Based on the results, we obtained better prediction with average accuracy of (68%) and average specificity of (69%) in contrast to the current state-of-the-art accuracy that is (51%) using the Visual Geometry Group (VGG16 also called OxfordNet), which is a convolutional neural network architecture developed by the Visual Geometry Group of Oxford. By implementing more novel lung segmentation techniques, reducing over fitting, and adding more learning layers, the proposed model has the potential to predict at higher accuracy than human specialists and will help subsidies and reduce the cost of diagnosis across the globe.  more » « less
Award ID(s):
1925960
PAR ID:
10140313
Author(s) / Creator(s):
; ; ;
Date Published:
Journal Name:
Proceedings of IEEE International Conference on Bioinformatics and Biomedicine (BIBM)
Page Range / eLocation ID:
2405 to 2412
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; ; (, Proceedings of the National Academy of Sciences)
    null (Ed.)
    By March 2020, COVID-19 led to thousands of deaths and disrupted economic activity worldwide. As a result of narrow case definitions and limited capacity for testing, the number of unobserved severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections during its initial invasion of the United States remains unknown. We developed an approach for estimating the number of unobserved infections based on data that are commonly available shortly after the emergence of a new infectious disease. The logic of our approach is, in essence, that there are bounds on the amount of exponential growth of new infections that can occur during the first few weeks after imported cases start appearing. Applying that logic to data on imported cases and local deaths in the United States through 12 March, we estimated that 108,689 (95% posterior predictive interval [95% PPI]: 1,023 to 14,182,310) infections occurred in the United States by this date. By comparing the model’s predictions of symptomatic infections with local cases reported over time, we obtained daily estimates of the proportion of symptomatic infections detected by surveillance. This revealed that detection of symptomatic infections decreased throughout February as exponential growth of infections outpaced increases in testing. Between 24 February and 12 March, we estimated an increase in detection of symptomatic infections, which was strongly correlated (median: 0.98; 95% PPI: 0.66 to 0.98) with increases in testing. These results suggest that testing was a major limiting factor in assessing the extent of SARS-CoV-2 transmission during its initial invasion of the United States. 
    more » « less
  2. ; ; ; ; ; (, Neurobiology of Aging)
    We explored neural processing differences associated with aging across four cognitive functions. In addition to ERP analysis, we included task-related microstate analyses, which identified stable states of neural activity across the scalp over time, to explore whole-head neural activation differences. Younger and older adults (YA, OA) completed face perception (N170), word-pair judgment (N400), visual oddball (P3), and flanker (ERN) tasks. Age-related effects differed across tasks. Despite age-related delayed latencies, N170 ERP and microstate analyses indicated no age-related differences in amplitudes or microstates. However, age-related condition differences were found for P3 and N00 amplitudes and scalp topographies: smaller condition differences were found for in OAs as well as broader centroparietal scalp distributions. Age group comparisons for the ERN revealed similar focal frontocentral activation loci, but differential activation patterns. Our findings of differential age effects across tasks are most consistent with the STAC-r framework which proposes that age-related effects differ depending on the resources available and the kinds of processing and cognitive load required of various tasks. 
    more » « less
  3. ; ; ; ; ; ; ; (, Communications Biology)
    Abstract Incomplete documentary evidence, variable biomolecular preservation, and limited skeletal responses have hindered assessment of acute infections in the past. This study was initially developed to explore the diagnostic potential of dental calculus to identify infectious diseases, however, the breadth and depth of information gained from a particular individual, St. Louis Individual (St.LI), enabled an individualized assessment and demanded broader disciplinary introspection of ethical research conduct. Here, we document the embodiment of structural violence in a 23-year-old Black and/or African American male, who died of lobar pneumonia in 1930s St. Louis, Missouri. St.LI exhibits evidence of systemic poor health, including chronic oral infections and a probable tuberculosis infection. Metagenomic sequencing of dental calculus recovered three pre-antibiotic era pathogen genomes, which likely contributed to the lobar pneumonia cause of death (CoD):Klebsiella pneumoniae(13.8X);Acinetobacter nosocomialis(28.4X); andAcinetobacter junii(30.1X). Ante- and perimortem evidence of St.LI’s lived experiences chronicle the poverty, systemic racism, and race-based structural violence experienced by marginalized communities in St. Louis, which contributed to St.LI’s poor health, CoD, anatomization, and inclusion in the Robert J. Terry Anatomical Collection. These same embodied inequalities continue to manifest as health disparities affecting many contemporary communities in the United States. 
    more » « less
  4. ; (, AIAA AVIATION Forum)
    null (Ed.)
    Machine learning techniques have received attention in fluid dynamics in terms of predicting, clustering and classifying complex flow physics. One application has been the classification or clustering of various wake structures that emanate from blu˙ bodies such as cylinders or flapping foils, creating a rich diversity of vortex formations specific to flow conditions, geometry, and/or kinematics of the body. When utilizing oscillating foils to harvest energy from tidal or river flows, it is critical to understand the intricate and nonlinear relationship between flapping kinematics and the downstream vortex wake structure for optimal siting and operation of arrays. This paper develops a classification model to obtain groups of kinematics that contain similar wake patterns within the energy harvesting regime. Data is obtained through simulations of 27 unique oscillating foil kinematics for a total of 13,650 samples of the wake vorticity field. Within these samples three groups are visually labeled based on the relative angle of attack. A machine learning approach combining a convolutional neural network (CNN) with long short-term memory (LSTM) units is utilized to automatically classify the wakes into the three groups. The average accuracy on five test data subsets is 80% when the three visually labeled groups are used for classification. After analyzing the test subset with lowest accuracy, an update on the group division boundaries is proposed. With this update, the algorithm achieves an average accuracy of 90%, demonstrating that the three groups are able to discern distinct wake structures within a range of energy harvesting kinematics. 
    more » « less
  5. ; ; ; ; ; ; ; ; ; ; et al (, Frontiers in Artificial Intelligence)
    null (Ed.)
    The newly discovered Coronavirus Disease 2019 (COVID-19) has been globally spreading and causing hundreds of thousands of deaths around the world as of its first emergence in late 2019. The rapid outbreak of this disease has overwhelmed health care infrastructures and arises the need to allocate medical equipment and resources more efficiently. The early diagnosis of this disease will lead to the rapid separation of COVID-19 and non-COVID cases, which will be helpful for health care authorities to optimize resource allocation plans and early prevention of the disease. In this regard, a growing number of studies are investigating the capability of deep learning for early diagnosis of COVID-19. Computed tomography (CT) scans have shown distinctive features and higher sensitivity compared to other diagnostic tests, in particular the current gold standard, i.e., the Reverse Transcription Polymerase Chain Reaction (RT-PCR) test. Current deep learning-based algorithms are mainly developed based on Convolutional Neural Networks (CNNs) to identify COVID-19 pneumonia cases. CNNs, however, require extensive data augmentation and large datasets to identify detailed spatial relations between image instances. Furthermore, existing algorithms utilizing CT scans, either extend slice-level predictions to patient-level ones using a simple thresholding mechanism or rely on a sophisticated infection segmentation to identify the disease. In this paper, we propose a two-stage fully automated CT-based framework for identification of COVID-19 positive cases referred to as the “COVID-FACT”. COVID-FACT utilizes Capsule Networks, as its main building blocks and is, therefore, capable of capturing spatial information. In particular, to make the proposed COVID-FACT independent from sophisticated segmentations of the area of infection, slices demonstrating infection are detected at the first stage and the second stage is responsible for classifying patients into COVID and non-COVID cases. COVID-FACT detects slices with infection, and identifies positive COVID-19 cases using an in-house CT scan dataset, containing COVID-19, community acquired pneumonia, and normal cases. Based on our experiments, COVID-FACT achieves an accuracy of 90.82 % , a sensitivity of 94.55 % , a specificity of 86.04 % , and an Area Under the Curve (AUC) of 0.98, while depending on far less supervision and annotation, in comparison to its counterparts. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email