Deep learning for biomedical image reconstruction: a survey

Yedder, Hanene Ben; Cardoen, Ben; Hamarneh, Ghassan

doi:10.1007/s10462-020-09861-2

Cited by 68 publications

(37 citation statements)

References 111 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There is a need to layout a lot of equipment and equipment maintenance personnel; the cost is still too high and is very vulnerable to the weather and geographical environment [ 13 ]. Therefore, we still cannot meet the needs of users of the nuclear radiation detection system [ 14 ]. Along with the development of wireless communication technology, in the acquisition of nuclear radiation data, environmental radiation detector, and gamma energy spectrometer in the introduction of wireless communication technology, compared to the use of wired communication system, to a certain extent to reduce the cost, but we still cannot overcome the many shortcomings of the nuclear radiation detection system [ 15 ].…”

Section: Related Workmentioning

confidence: 99%

Estimation of Nuclear Medicine Exposure Measures Based on Intelligent Computer Processing

Wang¹,

Wang²,

Liu³

et al. 2021

Journal of Healthcare Engineering

View full text Add to dashboard Cite

This paper provides an in-depth discussion and analysis of the estimation of nuclear medicine exposure measurements using computerized intelligent processing. The focus is on the study of energy extraction algorithms to obtain a high energy resolution with the lowest possible ADC sampling rate and thus reduce the amount of data. This paper focuses on the direct pulse peak extraction algorithm, polynomial curve fitting algorithm, double exponential function curve fitting algorithm, and pulse area calculation algorithm. The detector output waveforms are obtained with an oscilloscope, and the analysis module is designed in MATLAB. Based on these algorithms, the data obtained from six different lower sampling rates are analyzed and compared with the results of the high sampling rate direct pulse peak extraction algorithm and the pulse area calculation algorithm, respectively. The correctness of the compartment model was checked, and the results were found to be realistic and reliable, which can be used for the analysis of internal exposure data in radiation occupational health management, estimation of internal exposure dose for nuclear emergency groups, and estimation of accidental internal exposure dose. The results of the compartment model of the respiratory tract and the compartment model of the digestive tract can be used to calculate the distribution and retention patterns of radionuclides and their compounds in the body, which can be used to assess the damage of radionuclide internal contamination and guide the implementation of medical treatment.

show abstract

Section: Related Workmentioning

confidence: 99%

Estimation of Nuclear Medicine Exposure Measures Based on Intelligent Computer Processing

Wang¹,

Wang²,

Liu³

et al. 2021

Journal of Healthcare Engineering

View full text Add to dashboard Cite

show abstract

“…Furthermore, their reconstruction quality is highly dependent on the regularization function and related hyper-parameters settings that often require expert parameters tuning. Recently, deep learning has emerged as a powerful alternative for biomedical image reconstruction [9]- [11] and shown potential in improving resolution accuracy and speeding up reconstruction results especially in the presence of noisy and limited view acquisition. The capability of deep neural networks to learn complex functions automatically with fast inference makes their application to this problem ideal.…”

Section: B Current Approaches To Address Dot Inverse Problemmentioning

confidence: 99%

Multitask Deep Learning Reconstruction and Localization of Lesions in Limited Angle Diffuse Optical Tomography

Yedder¹,

Cardoen²,

Hamarneh³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

<div>Diffuse optical tomography (DOT) leverages near-infrared light propagation through tissue to assess its optical properties and identify abnormalities. </div><div>DOT image reconstruction from limited-angle data acquisition is severely ill-posed due to the highly scattered photons in the medium and the relatively small number of collected projections. Reconstructions are thus commonly marred by artifacts and, as a result, it is difficult to obtain accurate reconstruction of target objects, e.g., malignant lesions.</div><div>Reconstruction does not always ensure good localization of small lesions. Furthermore, conventional optimization-based reconstruction methods are computationally expensive, rendering them too slow for real-time imaging applications.</div><div>Our goal is to develop a fast and accurate image reconstruction method using deep learning, where multitask learning ensures accurate lesion localization in addition to improved reconstruction.</div><div>We apply spatial-wise attention and a distance transform based loss function in a novel multitask learning formulation to improve localization and reconstruction compared to single-task optimized methods.</div><div>Given the scarcity of real-world sensor-image pairs required for training supervised deep learning models,</div><div>we leverage physics-based simulation to generate synthetic datasets and use a transfer learning module to align the sensor domain distribution between in silico and real-world data, while taking advantage of cross-domain learning.</div><div>Both quantitative and qualitative results on phantom and real data indicate the superiority of our multitask method in the reconstruction and localization of lesions in tissue compared to state-of-the-art methods.</div><div>The results demonstrate that multitask learning provides sharper and more accurate reconstruction.</div>

show abstract

“…Deep learning, on the other hand, can learn the hierarchical features of pathologies automatically, eliminating the need to manually design the morphological operations of feature extraction and classifier. The deep learning approach excels in several fields, including signal processing [10], pedestrian detection [11], face recognition [12], road crack detection [13], biomedical image analysis [14], and many others. Furthermore, deep learning techniques have produced promising results throughout the agricultural field, helping more farmers and food-producing workers, such as detection of plant disease [15], analysis of weeds [16], discovery of valuable seeds [17], insect detection [18], fruit processing [15], and so on, which has led to dealing with image analysis.…”

Section: Introductionmentioning

confidence: 99%

Automatic Detection of Citrus Fruit and Leaves Diseases Using Deep Neural Network Model

et al. 2021

View full text Add to dashboard Cite

Citrus fruit diseases are the major cause of extreme citrus fruit yield declines. As a result, designing an automated detection system for citrus plant diseases is important. Deep learning methods have recently obtained promising results in a number of artificial intelligence issues, leading us to apply them to the challenge of recognizing citrus fruit and leaf diseases. In this paper, an integrated approach is used to suggest a convolutional neural networks (CNNs) model. The proposed CNN model is intended to differentiate healthy fruits and leaves from fruits/leaves with common citrus diseases such as Black spot, canker, scab, greening, and Melanose. The proposed CNN model extracts complementary discriminative features by integrating multiple layers. The CNN model was checked against many state-of-the-art deep learning models on the Citrus and PlantVillage datasets. The experimental results indicate that the CNN Model outperforms the competitors on a number of measurement metrics. The CNN Model has a test accuracy of 94.55 percent, making it a valuable decision support tool for farmers looking to classify citrus fruit/leaf diseases.

show abstract

Deep learning for biomedical image reconstruction: a survey

Cited by 68 publications

References 111 publications

Estimation of Nuclear Medicine Exposure Measures Based on Intelligent Computer Processing

Estimation of Nuclear Medicine Exposure Measures Based on Intelligent Computer Processing

Multitask Deep Learning Reconstruction and Localization of Lesions in Limited Angle Diffuse Optical Tomography

Automatic Detection of Citrus Fruit and Leaves Diseases Using Deep Neural Network Model

Contact Info

Product

Resources

About