Deep Learning in Radiology: Does One Size Fit All?

Erickson, Bradley J.; Korfiatis, Panagiotis; Kline, Timothy L.; Akkus, Zeynettin; Philbrick, Kenneth A.; Weston, Alexander D.

doi:10.1016/j.jacr.2017.12.027

Cited by 100 publications

(51 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sci. 2019, 9, 2921; doi:10.3390/app9142921 www.mdpi.com/journal/applsci only can reduce cost-effectiveness and mortality, but also provides good predictions that facilitate precise treatment [1][2][3]. Cardiac Disease (CD) monitoring and detection, in particular, is a difficult task which requires identifying the patterns and interaction among variables using various techniques [2].…”

Section: Introductionmentioning

confidence: 99%

“…It can facilitate the exploration of novel factors in score systems or add hidden risk factors to existing models [11], classify novel genotypes and phenotypes from heterogeneous cardiac diseases [2], detect lymph node metastases from breast cancer [12], detect cardiomyopathy [13], and use a risk factor prediction of bleeding and stroke to provide the optimal dose and anticoagulant therapy duration and to identify additional stroke risk factors [14]. In the diagnosis of cardiac disease, the implementation of DL produces good results [1,5,10]. The algorithms provide a very in-depth analysis for an artificial real-time cardiac imaging with better spatial and temporal resolution.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Automated ECG Beat Classification System Using Deep Neural Networks with an Unsupervised Feature Extraction Technique

et al. 2019

View full text Add to dashboard Cite

An automated classification system based on a Deep Learning (DL) technique for Cardiac Disease (CD) monitoring and detection is proposed in this paper. The proposed DL architecture is divided into Deep Auto-Encoders (DAEs) as an unsupervised form of feature learning and Deep Neural Networks (DNNs) as a classifier. The objective of this study is to improve on the previous machine learning technique that consists of several data processing steps such as feature extraction and feature selection or feature reduction. It is also noticed that the previously used machine learning technique required human interference and expertise in determining robust features, yet was time-consuming in the labeling and data processing steps. In contrast, DL enables an embedded feature extraction and feature selection in DAEs pre-training and DNNs fine-tuning process directly from raw data. Hence, DAEs is able to extract high-level of features not only from the training data but also from unseen data. The proposed model uses 10 classes of imbalanced data from ECG signals. Since it is related to the cardiac region, abnormality is usually considered for an early diagnosis of CD. In order to validate the result, the proposed model is compared with the shallow models and DL approaches. Results found that the proposed method achieved a promising performance with 99.73% accuracy, 91.20% sensitivity, 93.60% precision, 99.80% specificity, and a 91.80% F1-Score. Moreover, both the Receiver Operating Characteristic (ROC) curve and the Precision-Recall (PR) curve from the confusion matrix showed that the developed model is a good classifier. The developed model based on unsupervised feature extraction and deep neural network is ready to be used on a large population before its installation for clinical usage.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

An Automated ECG Beat Classification System Using Deep Neural Networks with an Unsupervised Feature Extraction Technique

et al. 2019

View full text Add to dashboard Cite

show abstract

“…These issues limit the dataset that can be composed for training and validation purposes, and create a risk of "overfitting" the data and loss of generalisability, even if there are different techniques to identify and reduce overfitting. 44,45 In the light of significant differences in disease prevalence, imaging protocols with different imaging characteristics, choice of reference standard and equipment both within and across different countries, the scope of application of an AI software will need to be critically evaluated. 5,6 Upscaling AI-based techniques will require a much clearer understanding of the clinical need (or use case) and the business case (if commercialised), product regulation, verification, and monitoring.…”

Section: Professional Responsibilitiesmentioning

confidence: 99%

Governance of automated image analysis and artificial intelligence analytics in healthcare

et al. 2019

View full text Add to dashboard Cite

The hype over artificial intelligence (AI) has spawned claims that clinicians (particularly radiologists) will become redundant. It is still moot as to whether AI will replace radiologists in day-today clinical practice, but more AI applications are expected to be incorporated into the workflows in the foreseeable future. These applications could produce significant ethical and legal issues in healthcare if they cause abrupt disruptions to its contextual integrity and relational dynamics. Sustaining trust and trustworthiness is a key goal of governance, which is necessary to promote collaboration among all stakeholders and to ensure the responsible development and implementation of AI in radiology and other areas of clinical work. In this paper, the nature of AI governance in biomedicine is discussed along with its limitations. It is argued that radiologists must assume a more active role in propelling medicine into the digital age. In this respect, professional responsibilities include inquiring into the clinical and social value of AI, alleviating deficiencies in technical knowledge in order to facilitate ethical evaluation, supporting the recognition, and removal of biases, engaging the "black box" obstacle, and brokering a new social contract on informational use and security. In essence, a much closer integration of ethics, laws, and good practices is needed to ensure that AI governance achieves its normative goals.

show abstract

“…In the present work, we show results on a broad panel of 12 different MS‐mimicking diseases, collected on four different MRI scanners. CVSnet, our approach to automatically assess the CVS, is based on a 3D convolutional neural network architecture . Because it uses deep learning, it is an end‐to‐end learning approach that does not require handcrafted discriminative features or filters.…”

Section: Introductionmentioning

confidence: 99%

CVSnet: A machine learning approach for automated central vein sign assessment in multiple sclerosis

et al. 2020

View full text Add to dashboard Cite

The central vein sign (CVS) is an efficient imaging biomarker for multiple sclerosis (MS) diagnosis, but its application in clinical routine is limited by inter-rater variability and the expenditure of time associated with manual assessment. We describe a deep learning-based prototype for automated assessment of the CVS in white matter MS lesions using data from three different imaging centers. We retrospectively analyzed data from 3 T magnetic resonance images acquired on four scanners from two different vendors, including adults with MS (n = 42), MS mimics (n = 33, encompassing 12 distinct neurological diseases mimicking MS) and uncertain diagnosis (n = 5). Brain white matter lesions were manually segmented on FLAIR* images. Perivenular assessment was performed according to consensus guidelines and used as ground truth, yielding 539 CVS-positive (CVS + ) and 448 CVS-negative (CVS − ) lesions. A 3D convolutional neural network ("CVSnet") was designed and trained on 47 datasets, keeping 33 for testing. FLAIR* lesion patches of CVS + /CVS − lesions were used for training and validation (n = 375/298) and for testing (n = 164/150). Performance was evaluated lesion-wise and subject-wise and compared with a state-of-the-art vesselness filtering approach through McNemar's test. The proposed CVSnet approached human performance, with lesion-wise median balanced accuracy of 81%, and subject-wise balanced accuracy of 89% on the validation set, and 91% on the test set. The process of CVS assessment, in previously manually segmented lesions, was~600-fold faster using the proposed CVSnet compared with human visual assessment (test set: 4 seconds vs. 40 minutes). On the validation and test sets, the lesion-wise performance outperformed the vesselness filter method (P < 0.001). The proposed deep learning prototype shows promising performance in differentiating MS from its mimics. Our approach was evaluated using data from different hospitals, enabling larger multicenter trials to evaluate the benefit of introducing the CVS marker into MS diagnostic criteria.Abbreviations used: AUC, area under the curve; CVS, central vein sign; FLAIR, 3D T2-weighted fluid-attenuated inversion recovery; GPU, graphics processing unit; MRI, magnetic resonance imaging; MS, multiple sclerosis; NIR, no information rate; ReLU, rectified linear unit; ROC, receiver operating characteristic; T2*-EPI, T2*-weighted segmented echo-planar imaging; 3D, threedimensional; WM, white matter.Pietro Maggi and Mário João Fartaria contributed equally to this work.

show abstract

Deep Learning in Radiology: Does One Size Fit All?

Cited by 100 publications

References 23 publications

An Automated ECG Beat Classification System Using Deep Neural Networks with an Unsupervised Feature Extraction Technique

An Automated ECG Beat Classification System Using Deep Neural Networks with an Unsupervised Feature Extraction Technique

Governance of automated image analysis and artificial intelligence analytics in healthcare

CVSnet: A machine learning approach for automated central vein sign assessment in multiple sclerosis

Contact Info

Product

Resources

About