The current status of the Nano-JASMINE project

AbstractNano-JASMINE is a nano-size astrometry satellite that will carry out astrometry measurements of nearby bright stars for more than one year. This will enable us to detect annual parallaxes of stars within 300 pc from the Sun. We expect the satellite to be launched as a piggy-back system as early as in 2009 into a Sun synchronized orbit at the altitude between 500 and 800 km. Being equipped with a beam combiner, the satellite has a capability to observe two different fields simultaneously and will be able to carry out HIPPARCOS-type observations along great circles. A 5 cm all aluminum made reflecting telescope with a aluminum beam combiner is developed. Using the on-board CCD controller, experiments with a real star have been executed. A communication band width is insufficient to transfer all imaging data, hence, we developed an onboard data processing system that extracts stellar image data from vast amount of imaging data. A newly developed 2K × 1K fully-depleted CCD will be used for the mission. It will work in the time delayed integration(TDI) mode. The bus system has been designed with special consideration of the following two points. Those are the thermal stabilization of the telescope and the accuracy of the altitude control. The former is essential to achieve high astrometric accuracies, on the order of 1 mas. Therefore relative angle of the beam combiner must be stable within 1 mas. A 3-axes control of the satellite will be realized by using fiber gyro and triaxial reaction wheel system and careful treatment of various disturbing forces.

Download Full-text

An Ad Hoc Random Initialization Deep Neural Network Architecture for Discriminating Malignant Breast Cancer Lesions in Mammographic Images

Contrast Media & Molecular Imaging ◽

10.1155/2019/5982834 ◽

2019 ◽

Vol 2019 ◽

pp. 1-9 ◽

Cited By ~ 2

Author(s):

Andrea Duggento ◽

Marco Aiello ◽

Carlo Cavaliere ◽

Giuseppe L. Cascella ◽

Davide Cascella ◽

...

Keyword(s):

Breast Cancer ◽

Network Architecture ◽

Ad Hoc ◽

False Negative ◽

Selection Criterion ◽

Current Status ◽

Clinical Image ◽

Imaging Data ◽

Breast Cancer Death ◽

Random Initialization

Breast cancer is one of the most common cancers in women, with more than 1,300,000 cases and 450,000 deaths each year worldwide. In this context, recent studies showed that early breast cancer detection, along with suitable treatment, could significantly reduce breast cancer death rates in the long term. X-ray mammography is still the instrument of choice in breast cancer screening. In this context, the false-positive and false-negative rates commonly achieved by radiologists are extremely arduous to estimate and control although some authors have estimated figures of up to 20% of total diagnoses or more. The introduction of novel artificial intelligence (AI) technologies applied to the diagnosis and, possibly, prognosis of breast cancer could revolutionize the current status of the management of the breast cancer patient by assisting the radiologist in clinical image interpretation. Lately, a breakthrough in the AI field has been brought about by the introduction of deep learning techniques in general and of convolutional neural networks in particular. Such techniques require no a priori feature space definition from the operator and are able to achieve classification performances which can even surpass human experts. In this paper, we design and validate an ad hoc CNN architecture specialized in breast lesion classification from imaging data only. We explore a total of 260 model architectures in a train-validation-test split in order to propose a model selection criterion which can pose the emphasis on reducing false negatives while still retaining acceptable accuracy. We achieve an area under the receiver operatic characteristics curve of 0.785 (accuracy 71.19%) on the test set, demonstrating how an ad hoc random initialization architecture can and should be fine tuned to a specific problem, especially in biomedical applications.

Download Full-text

ESR white paper: blockchain and medical imaging

Insights into Imaging ◽

10.1186/s13244-021-01029-y ◽

2021 ◽

Vol 12 (1) ◽

Author(s):

◽

Elmar Kotter ◽

Luis Marti-Bonmati ◽

Adrian P. Brady ◽

Nandita M. Desouza

Keyword(s):

Medical Imaging ◽

Image Data ◽

Distributed Database ◽

Relevant Information ◽

White Paper ◽

Imaging Data ◽

Data Set ◽

Blockchain Technology ◽

Potential Applications ◽

Basic Technology

AbstractBlockchain can be thought of as a distributed database allowing tracing of the origin of data, and who has manipulated a given data set in the past. Medical applications of blockchain technology are emerging. Blockchain has many potential applications in medical imaging, typically making use of the tracking of radiological or clinical data. Clinical applications of blockchain technology include the documentation of the contribution of different “authors” including AI algorithms to multipart reports, the documentation of the use of AI algorithms towards the diagnosis, the possibility to enhance the accessibility of relevant information in electronic medical records, and a better control of users over their personal health records. Applications of blockchain in research include a better traceability of image data within clinical trials, a better traceability of the contributions of image and annotation data for the training of AI algorithms, thus enhancing privacy and fairness, and potentially make imaging data for AI available in larger quantities. Blockchain also allows for dynamic consenting and has the potential to empower patients and giving them a better control who has accessed their health data. There are also many potential applications of blockchain technology for administrative purposes, like keeping track of learning achievements or the surveillance of medical devices. This article gives a brief introduction in the basic technology and terminology of blockchain technology and concentrates on the potential applications of blockchain in medical imaging.

Download Full-text

Identification of Tumor-Specific MRI Biomarkers Using Machine Learning (ML)

Diagnostics ◽

10.3390/diagnostics11050742 ◽

2021 ◽

Vol 11 (5) ◽

pp. 742

Author(s):

Rima Hajjo ◽

Dima A. Sabbah ◽

Sanaa K. Bardaweel ◽

Alexander Tropsha

Keyword(s):

Machine Learning ◽

High Specificity ◽

Cancer Biomarkers ◽

Cancer Prognosis ◽

Current Status ◽

Imaging Data ◽

Non Invasive ◽

Machine Learning Methods ◽

Cancer Types ◽

Magnetic Resonance Imaging Mri

The identification of reliable and non-invasive oncology biomarkers remains a main priority in healthcare. There are only a few biomarkers that have been approved as diagnostic for cancer. The most frequently used cancer biomarkers are derived from either biological materials or imaging data. Most cancer biomarkers suffer from a lack of high specificity. However, the latest advancements in machine learning (ML) and artificial intelligence (AI) have enabled the identification of highly predictive, disease-specific biomarkers. Such biomarkers can be used to diagnose cancer patients, to predict cancer prognosis, or even to predict treatment efficacy. Herein, we provide a summary of the current status of developing and applying Magnetic resonance imaging (MRI) biomarkers in cancer care. We focus on all aspects of MRI biomarkers, starting from MRI data collection, preprocessing and machine learning methods, and ending with summarizing the types of existing biomarkers and their clinical applications in different cancer types.

Download Full-text

An efficient approach to converting three-dimensional image data into highly accurate computational models

Philosophical Transactions of The Royal Society A Mathematical Physical and Engineering Sciences ◽

10.1098/rsta.2008.0090 ◽

2008 ◽

Vol 366 (1878) ◽

pp. 3155-3173 ◽

Cited By ~ 139

Author(s):

P.G Young ◽

T.B.H Beresford-West ◽

S.R.L Coward ◽

B Notarberardino ◽

B Walker ◽

...

Keyword(s):

Mesh Generation ◽

Computational Models ◽

Three Dimensional ◽

Image Data ◽

Model Material ◽

Imaging Data ◽

Material Inhomogeneity ◽

Dimensional Image ◽

Wide Range ◽

Impact Modelling

Image-based meshing is opening up exciting new possibilities for the application of computational continuum mechanics methods (finite-element and computational fluid dynamics) to a wide range of biomechanical and biomedical problems that were previously intractable owing to the difficulty in obtaining suitably realistic models. Innovative surface and volume mesh generation techniques have recently been developed, which convert three-dimensional imaging data, as obtained from magnetic resonance imaging, computed tomography, micro-CT and ultrasound, for example, directly into meshes suitable for use in physics-based simulations. These techniques have several key advantages, including the ability to robustly generate meshes for topologies of arbitrary complexity (such as bioscaffolds or composite micro-architectures) and with any number of constituent materials (multi-part modelling), providing meshes in which the geometric accuracy of mesh domains is only dependent on the image accuracy (image-based accuracy) and the ability for certain problems to model material inhomogeneity by assigning the properties based on image signal strength. Commonly used mesh generation techniques will be compared with the proposed enhanced volumetric marching cubes (EVoMaCs) approach and some issues specific to simulations based on three-dimensional image data will be discussed. A number of case studies will be presented to illustrate how these techniques can be used effectively across a wide range of problems from characterization of micro-scaffolds through to head impact modelling.

Download Full-text

Design and implementation of remote image processing system for space launch site

E3S Web of Conferences ◽

10.1051/e3sconf/202123304026 ◽

2021 ◽

Vol 233 ◽

pp. 04026

Author(s):

Ma Li ◽

Wang Bai Yan ◽

Liu Tao ◽

WangYu Chao ◽

Xiang Yu ◽

...

Keyword(s):

Image Processing ◽

Auxiliary Information ◽

Processing System ◽

Image Data ◽

Image Transmission ◽

Engineering Practice ◽

Image Processing System ◽

Rocket Flight ◽

Launch Site ◽

Space Launch

Telemetry image has the characteristics of intuitive image in the process of rocket flight. Through real-time acquisition of rocket flight video image, it can provide the working status of key nodes in the process of rocket flight, and provide intuitive decision-marking auxiliary information for commanders. This paper analyzes the design content of the image processing system of the space launch site from the aspects of image transmission mechanism, information flow, image data processing and image decoding, so as to provide technical basis for the image receiving, transmission and decoding process in the engineering practice of the image processing system.

Download Full-text

L2MXception: An Improved Xception Network for Classification of Peach diseases

10.21203/rs.3.rs-31469/v2 ◽

2020 ◽

Author(s):

Na Yao ◽

Fuchuan Ni ◽

Ziyan Wang ◽

Jun Luo ◽

Wing-Kin Sung ◽

...

Keyword(s):

Deep Learning ◽

Early Identification ◽

Image Data ◽

Yield Reduction ◽

Disease Prediction ◽

Imaging Data ◽

Regularization Term ◽

Detection And Identification ◽

L2 Norm

Abstract Background: Peach diseases can cause severe yield reduction and decreased quality for peach production. Rapid and accurate detection and identification of peach diseases is of great importance. Deep learning has been applied to detect peach diseases using imaging data. However, peach disease image data is difficult to collect and samples are imbalance. The popular deep networks perform poor for this issue.Results: This paper proposed an improved Xception network named as L2MXception which ensembles regularization term of L2-norm and mean. With the peach disease image dataset collected, results on seven mainstream deep learning models were compared in details and an improved loss function was integrated with regularization term L2-norm and mean (L2M Loss). Experiments showed that the Xception model with L2M Loss outperformed the current best method for peach disease prediction. Compared to the original Xception model, the validation accuracy of L2MXception was up to 93.85%, increased by 28.48%. Conclusions: The proposed L2MXception network may have great potential in early identification of peach diseases.

Download Full-text

The deployable telescope: a cutting-edge solution for high spatial and temporal resolved Earth observation

Advanced Optical Technologies ◽

10.1515/aot-2018-0043 ◽

2018 ◽

Vol 7 (6) ◽

pp. 365-376 ◽

Cited By ~ 1

Author(s):

Dennis Dolkens ◽

Hans Kuiper ◽

Victor Villalba Corbacho

Keyword(s):

Temporal Resolution ◽

Systems Engineering ◽

Systems Design ◽

Image Data ◽

Earth Observation ◽

Cutting Edge ◽

High Mass ◽

Current Status ◽

Active Optics ◽

The Status

Abstract The increase of spatial and temporal resolution for Earth observation (EO) is the ultimate driver for science and societal applications. However, the state-of-the-art EO missions like DigitalGlobe’s Worldview-3, are very costly. Moreover, this system has a high mass of 2800 kg and limited swath width of about 15 km which limits the temporal resolution. In this article, we present the status of the deployable space telescope (DST) project, which has been running for 6 years now at the Delft University of Technology, as a cutting-edge solution to solve this issue. Deployable optics have the potential of revolutionising the field of high resolution EO. By splitting up the primary mirror (M1) of a telescope into deployable segments and placing the secondary mirror (M2) on a deployable boom, the launch volume of a telescope can be reduced by a factor of 4 or more, allowing for much lower launch costs. This allows for larger EO constellations, providing image data with a much better revisit time than existing solutions. The DST specification baseline, based on Wordview-3, aims to provide images with a ground resolution of 25 cm (panchromatic 450–650 nm) from an orbital altitude of 500 km. In this paper, the current status of the optical, thermo-mechanical, and active optics systems design are described. The concurrent design approach combined with a strict bottom-up and top-down compliant systems engineering approach show that the DST is a healthy system concept.

Download Full-text

Clinical Applications of Patient-Specific 3D Printed Models in Cardiovascular Disease: Current Status and Future Directions

Biomolecules ◽

10.3390/biom10111577 ◽

2020 ◽

Vol 10 (11) ◽

pp. 1577

Author(s):

Zhonghua Sun

Keyword(s):

Cardiovascular Disease ◽

3D Printing ◽

Current Status ◽

Patient Specific ◽

Future Research ◽

Junior Doctors ◽

Imaging Data ◽

Training Tool ◽

Complex Anatomy ◽

3D Printed

Three-dimensional (3D) printing has been increasingly used in medicine with applications in many different fields ranging from orthopaedics and tumours to cardiovascular disease. Realistic 3D models can be printed with different materials to replicate anatomical structures and pathologies with high accuracy. 3D printed models generated from medical imaging data acquired with computed tomography, magnetic resonance imaging or ultrasound augment the understanding of complex anatomy and pathology, assist preoperative planning and simulate surgical or interventional procedures to achieve precision medicine for improvement of treatment outcomes, train young or junior doctors to gain their confidence in patient management and provide medical education to medical students or healthcare professionals as an effective training tool. This article provides an overview of patient-specific 3D printed models with a focus on the applications in cardiovascular disease including: 3D printed models in congenital heart disease, coronary artery disease, pulmonary embolism, aortic aneurysm and aortic dissection, and aortic valvular disease. Clinical value of the patient-specific 3D printed models in these areas is presented based on the current literature, while limitations and future research in 3D printing including bioprinting of cardiovascular disease are highlighted.

Download Full-text