Deep learning-based automated detection of pulmonary embolism on CT pulmonary angiograms: No significant effects on report communication times and patient turnaround in the emergency department nine months after technical implementation

Schmuelling, Lena; Franzeck, Fabian; Nickel, Christian H.; Mansella, Gregory; Bingisser, Roland; Schmidt, Noemi; Stieltjes, Bram; Bremerich, Jens; Sauter, Alexander; Weikert, Thomas; Sommer, Gregor

doi:10.1016/j.ejrad.2021.109816

Cited by 23 publications

(4 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such automatization provides the opportunity for future training and validation on much larger data sets. A critical factor for the successful implementation of artificial intelligence systems into hospital workflows besides factors such as integration into the local standard operating procedures and clinical systems is a high robustness and reliability to establish the necessary trust into AI systems [39]. The LRP analysis presented here is able to generate a graphical analysis of the decision relevant areas of the network in the images and therefore, could be used for validation.…”

Section: Discussionmentioning

confidence: 99%

Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning

et al. 2021

View full text Add to dashboard Cite

Abdominal aortic aneurysms (AAA) may remain clinically silent until they enlarge and patients present with a potentially lethal rupture. This necessitates early detection and elective treatment. The goal of this study was to develop an easy-to-train algorithm which is capable of automated AAA screening in CT scans and can be applied to an intra-hospital environment. Three deep convolutional neural networks (ResNet, VGG-16 and AlexNet) were adapted for 3D classification and applied to a dataset consisting of 187 heterogenous CT scans. The 3D ResNet outperformed both other networks. Across the five folds of the first training dataset it achieved an accuracy of 0.856 and an area under the curve (AUC) of 0.926. Subsequently, the algorithms performance was verified on a second data set containing 106 scans, where it ran fully automated and resulted in an accuracy of 0.953 and an AUC of 0.971. A layer-wise relevance propagation (LRP) made the decision process interpretable and showed that the network correctly focused on the aortic lumen. In conclusion, the deep learning-based screening proved to be robust and showed high performance even on a heterogeneous multi-center data set. Integration into hospital workflow and its effect on aneurysm management would be an exciting topic of future research.

show abstract

Section: Discussionmentioning

confidence: 99%

Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning

et al. 2021

View full text Add to dashboard Cite

show abstract

“…AI improved the sensitivity of physicians by 9% and the specificity by 4% and reduced the average number of false-positive fractures per patient by 42% and mean reading time by 15%. Schmuelling et al [31] assessed the impact of a triage system that detected and alerted radiologists about ED cases with suspected pulmonary embolism on CT angiograms. While the study demonstrated good diagnostic accuracy (sensitivity 80%, specificity 95%, PPV 82%, and NPV 94%), there was no effect on report communication times and patient turnaround 9-months post-implementation.…”

Section: Radiologymentioning

confidence: 99%

Automation in Contemporary Clinical Information Systems: a Survey of AI in Healthcare Settings

Magrabi,

Lyell,

Coiera

2023

Yearb Med Inform

View full text Add to dashboard Cite

Aims and objectives: To examine the nature and use of automation in contemporary clinical information systems by reviewing studies reporting the implementation and evaluation of artificial intelligence (AI) technologies in healthcare settings. Method: PubMed/MEDLINE, Web of Science, EMBASE, the tables of contents of major informatics journals, and the bibliographies of articles were searched for studies reporting evaluation of AI in clinical settings from January 2021 to December 2022. We documented the clinical application areas and tasks supported, and the level of system autonomy. Reported effects on user experience, decision-making, care delivery and outcomes were summarised. Results: AI technologies are being applied in a wide variety of clinical areas. Most contemporary systems utilise deep learning, use routinely collected data, support diagnosis and triage, are assistive (requiring users to confirm or approve AI provided information or decisions), and are used by doctors in acute care settings in high-income nations. AI systems are integrated and used within existing clinical information systems including electronic medical records. There is limited support for One Health goals. Evaluation is largely based on quantitative methods measuring effects on decision-making. Conclusion: AI systems are being implemented and evaluated in many clinical areas. There remain many opportunities to understand patterns of routine use and evaluate effects on decision-making, care delivery and patient outcomes using mixed-methods. Support for One Health including integrating data about environmental factors and social determinants needs further exploration.

show abstract

“…Moreover, the neural network has high inclusiveness and scalability, which can continuously expand data processing capacity by increasing the number of hidden layers. In addition, the neural network has enormous development potential embedded in its continuous improvement by optimization algorithms, which empower the network with different functions and adaptability to different sample data and application scenarios (Schmuelling et al, 2021). With the constant expansion of the application field of DL, the neural network will further enlarge the application range.…”

Section: Feasibility Analysis Of Neural Network Modellingmentioning

confidence: 99%

The evaluation of enterprise supply chain intelligent manufacturing system for agricultural interconnection data based on machine learning

Yu¹,

Li²,

Li³

et al. 2023

Expert Systems

View full text Add to dashboard Cite

The present work aims to promote the modernization of the real economy and facilitate the manufacturing industry to construct an intelligent manufacturing system. Analysis and evaluation are performed on the intelligent manufacturing system (IMS) of enterprises under the Internet of Things (IoT). First, the connotation and characteristics of machine learning (ML) in the era of IoT are explained, followed by the introduction to neural networks. Second, the optimization process of the neural network is described. Since it has no memory, the traditional deep neural network (DNN) needs to be improved. Then, the Generative Adversarial Network (GAN) is utilized to expand the sample data. Moreover, the confusion matrix is used in the indicator evaluation. The backpropagation neural network (BPNN) is optimized by the genetic algorithm (GA) to improve the network configuration and model performance. In addition, the connotation and development status of IMS are analysed, and the feasibility of applying deep learning to the evaluation of IMS is discussed. Finally, the evaluation model of IMS is designed and built via a neural network, which is ultimately verified in the simulation experiment. The experimental results indicate that the BPNN‐GA algorithm has advantages in predicting fitting accuracy, taking up less memory, and shortening training time; it can achieve generalization and accuracy in the forefront of optimal Pareto value, and the fitting accuracy attains 95.3%. The horizontal evaluation of IMS proves the effectiveness and feasibility of the evaluation model. This shows that the BPNN‐GA algorithm has an excellent evaluation effect on IMS. And the IMS evaluation model is applied to assess the IMS of sample enterprises. The evaluation results demonstrate that there are problems in the current manufacturing industry. For example, the transactions are challenging to be completed smoothly, information technology is updated slowly, and IMS is dependent on collecting resources.

show abstract

Deep learning-based automated detection of pulmonary embolism on CT pulmonary angiograms: No significant effects on report communication times and patient turnaround in the emergency department nine months after technical implementation

Cited by 23 publications

References 33 publications

Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning

Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning

Automation in Contemporary Clinical Information Systems: a Survey of AI in Healthcare Settings

The evaluation of enterprise supply chain intelligent manufacturing system for agricultural interconnection data based on machine learning

Contact Info

Product

Resources

About