A Review of Deep Learning in Medical Imaging: Imaging Traits, Technology Trends, Case Studies With Progress Highlights, and Future Promises

Zhou, Shenglong; Greenspan, Hayit; Davatzikos, Christos; Duncan, James S.; Ginneken, Bram van; Madabhushi, Anant; Prince, Jerry L.; Rueckert, Daniel; Summers, Ronald M.

doi:10.1109/jproc.2021.3054390

Cited by 470 publications

(282 citation statements)

References 207 publications

Supporting

Mentioning

277

Contrasting

Unclassified

Order By: Relevance

“…Conventional modes of regularization include adding some weights to the loss function. However, CNN uses a different regularization strategy; they utilize the hierarchical pattern in data and assemble more complex patterns using smaller and simpler patterns, reducing the connectedness and complexity [49], [54]- [56]. CNN learns the filters which were hand-engineered in traditional algorithms.…”

Section: A Deep Neural Network Architectures 1) Convolution Neural Network (Cnns)mentioning

confidence: 99%

“…Typically, the encoder takes an input sequence and transforms it into a fixed-shaped state (hidden representation), and the decoder maps the encoded state to an output sequence. Auto-encoders are a specific case of encoder-decoder models in which both input and output are identical [44], [63]- [67].…”

Section: ) Encoder-decoder Network Modelsmentioning

confidence: 99%

“…This review is limited to efficient models capable of dealing with partially annotated datasets. The authors in [42], and [49] have reviewed most of the medical image segmentation architectures. They have provided the state-of-the-art performance of various medical modalities; nonetheless, they have not compared various architectures' performance and limitations for a given dataset.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Deep Neural Architectures for Medical Image Semantic Segmentation: Review

et al. 2021

View full text Add to dashboard Cite

Deep learning has an enormous impact on medical image analysis. Many computer-aided diagnostic systems equipped with deep networks are rapidly reducing human intervention in healthcare. Among several applications, medical image semantic segmentation is one of the core areas of active research to delineate the anatomical structures and other regions of interest. It has a significant contribution to healthcare and provides guided interventions, radiotherapy, and improved radiological diagnostics. The underlying article provides a brief overview of deep convolutional neural architecture, the platforms and applications of deep neural networks, metrics used for empirical evaluation, state-of-the-art semantic segmentation architectures based on a foundational convolution concept, and a review of publicly available medical image datasets highlighting four distinct regions of interest. The article also analyzes the existing work and provides open-ended potential research directions in deep medical image semantic segmentation.

show abstract

Section: A Deep Neural Network Architectures 1) Convolution Neural Network (Cnns)mentioning

confidence: 99%

Section: ) Encoder-decoder Network Modelsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep Neural Architectures for Medical Image Semantic Segmentation: Review

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Due to the high spatial resolution and the unique relationship between CT density and lung air content [7]- [10], CT is widely preferred to recognize and segment the typical signs of COVID-19 infection [11]. Furthermore, segmentation of COVID-19 lesions provides crucial information for quantitative measurement and follow-up assessment [12].…”

Section: Introductionmentioning

confidence: 99%

Label-Free Segmentation of COVID-19 Lesions in Lung CT

Yao

Xiao

Liu

et al. 2021

IEEE Trans. Med. Imaging

Self Cite

103

View full text Add to dashboard Cite

Scarcity of annotated images hampers the building of automated solution for reliable COVID-19 diagnosis and evaluation from CT. To alleviate the burden of data annotation, we herein present a label-free approach for segmenting COVID-19 lesions in CT via voxel-level anomaly modeling that mines out the relevant knowledge from normal CT lung scans. Our modeling is inspired by the observation that the parts of tracheae and vessels, which lay in the high-intensity range where lesions belong to, exhibit strong patterns. To facilitate the learning of such patterns at a voxel level, we synthesize 'lesions' using a set of simple operations and insert the synthesized 'lesions' into normal CT lung scans to form training pairs, from which we learn a normalcy-recognizing network (NormNet) that recognizes normal tissues and separate them from possible COVID-19 lesions. Our experiments on three different public datasets validate the effectiveness of NormNet, which conspicuously outperforms a variety of unsupervised anomaly detection (UAD) methods.

show abstract

“…Medical imagery exhibits multi-modal long-tailed distributions with significant heterogeneity in the presentation of similar disease patterns due to equipment, scan technique, and patient variability [57]. This issue is demonstrated by most large CXR archives as they are limited to a single clinical site which reduces the captured variability of the factors of heterogeneity [57]. The resulting sparsity of training data has obvious implications for making the development of robust diagnostic models more challenging.…”

Section: Introductionmentioning

confidence: 99%

Evaluating the Clinical Realism of Synthetic Chest X-Rays Generated Using Progressively Growing GANs

Segal

Rubin

et al. 2021

SN COMPUT. SCI.

View full text Add to dashboard Cite

Chest X-rays are a vital diagnostic tool in the workup of many patients. Similar to most medical imaging modalities, they are profoundly multi-modal and are capable of visualising a variety of combinations of conditions. There is an ever pressing need for greater quantities of labelled images to drive forward the development of diagnostic tools; however, this is in direct opposition to concerns regarding patient confidentiality which constrains access through permission requests and ethics approvals. Previous work has sought to address these concerns by creating class-specific generative adversarial networks (GANs) that synthesise images to augment training data. These approaches cannot be scaled as they introduce computational trade offs between model size and class number which places fixed limits on the quality that such generates can achieve. We address this concern by introducing latent class optimisation which enables efficient, multi-modal sampling from a GAN and with which we synthesise a large archive of labelled generates. We apply a Progressive Growing GAN (PGGAN) to the task of unsupervised X-ray synthesis and have radiologists evaluate the clinical realism of the resultant samples. We provide an in depth review of the properties of varying pathologies seen on generates as well as an overview of the extent of disease diversity captured by the model. We validate the application of the Fréchet Inception Distance (FID) to measure the quality of X-ray generates and find that they are similar to other high-resolution tasks. We quantify X-ray clinical realism by asking radiologists to distinguish between real and fake scans and find that generates are more likely to be classed as real than by chance, but there is still progress required to achieve true realism. We confirm these findings by evaluating synthetic classification model performance on real scans. We conclude by discussing the limitations of PGGAN generates and how to achieve controllable, realistic generates going forward. We release our source code, model weights, and an archive of labelled generates.

show abstract

A Review of Deep Learning in Medical Imaging: Imaging Traits, Technology Trends, Case Studies With Progress Highlights, and Future Promises

Cited by 470 publications

References 207 publications

Deep Neural Architectures for Medical Image Semantic Segmentation: Review

Deep Neural Architectures for Medical Image Semantic Segmentation: Review

Label-Free Segmentation of COVID-19 Lesions in Lung CT

Evaluating the Clinical Realism of Synthetic Chest X-Rays Generated Using Progressively Growing GANs

Contact Info

Product

Resources

About