A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Pneumonia Disease Detection Using Varied-complexity Machine Learning Algorithms
Pneumonia, a severe respiratory infection, typically induced by viruses or bacteria, stands as a major global concern. It was responsible for the deaths of over 5,900,000 children under five years in 2015, equating to 16% of total fatalities, and continued to claim more than 808,000 lives in 2017, representing 15 % of all deaths within the same age group. Recently, the and artificial intelligence (AI) has brought about a remarkable change in the healthcare sector, leading to new advancements in medical treatments and diagnostic procedures. This study proposed two improved and varied complexity Convolutional Neural Networks (CNNs) for early detection of pneumonia from X-ray images of the chest, a task that traditionally relies on the expertise of skilled radiologists. By applying Convolutional Neural Networks (CNNs) and comparing two models of varying complexity, the research not only advances the precision of medical diagnostics but also illustrates that complexity in deep learning models does not necessarily equate to superior performance. Both models, the first complex model and the second simpler model, achieved training accuracies of 94.38 % and 99.64%, validation accuracies of 92.57% and 91.35%, and test accuracies of 91.5 % and 88.62 %, respectively. The standard comprehensive evaluation criteria, including a confusion matrix, accuracy, precision, recall, and F1 score, were applied to assess the classification capability of each model.
Pneumonia Disease Detection Using Varied-complexity Machine Learning Algorithms
Pneumonia, a severe respiratory infection, typically induced by viruses or bacteria, stands as a major global concern. It was responsible for the deaths of over 5,900,000 children under five years in 2015, equating to 16% of total fatalities, and continued to claim more than 808,000 lives in 2017, representing 15 % of all deaths within the same age group. Recently, the and artificial intelligence (AI) has brought about a remarkable change in the healthcare sector, leading to new advancements in medical treatments and diagnostic procedures. This study proposed two improved and varied complexity Convolutional Neural Networks (CNNs) for early detection of pneumonia from X-ray images of the chest, a task that traditionally relies on the expertise of skilled radiologists. By applying Convolutional Neural Networks (CNNs) and comparing two models of varying complexity, the research not only advances the precision of medical diagnostics but also illustrates that complexity in deep learning models does not necessarily equate to superior performance. Both models, the first complex model and the second simpler model, achieved training accuracies of 94.38 % and 99.64%, validation accuracies of 92.57% and 91.35%, and test accuracies of 91.5 % and 88.62 %, respectively. The standard comprehensive evaluation criteria, including a confusion matrix, accuracy, precision, recall, and F1 score, were applied to assess the classification capability of each model.
Pneumonia Disease Detection Using Varied-complexity Machine Learning Algorithms
Alobaid, Ahmad (author) / Bonny, Talal (author) / Al-Shabi, Mohammad (author)
2024-06-03
1225404 byte
Conference paper
Electronic Resource
English
Reliability analysis of portal frame subjected to varied lateral loads using machine learning
| Springer Verlag | 2024
Reliability analysis of portal frame subjected to varied lateral loads using machine learning
| Springer Verlag | 2024
Application of gradually-varied flow algorithms to simulate buried streams
| British Library Online Contents | 2002
| Springer Verlag | 2023