Reduced-Order Methods in Medical Imaging

A.P. Dhawan, H.K. Huang, D.-S. Kim, eds. World Scientific Publishing; 2008, 868 pages, $228.00. At first glance at the cover and size of Principles and Advanced Methods in Medical Imaging and Image Analysis , a passerby might think that it is the latest Harry Potter novel; however, the gadgets and

Introduction: The investigation of the cranial suture morphology and the changes occurring during the oblit­eration is an important precondition for understanding suture maturation and its relation to ageing. The I¼CT-imaging of dry skulls with known age-at-death is a powerful tool in this direction, but this technology is still ex­pensive, difficult to access and entirely inapplicable on living individuals. On the other hand, the diagnostic im­aging of patients accumulates large amount of volumetric images, which could be used for scientific researches. In medical imaging, the generated data is typically artificially blurred for denoising purposes. The standard Gauss­ian blur assigns a weighted average value for the gray-scale intensity of each pixel, based on the initial intensities of the pixel and its neighbors. Unfortunately, together with the noise reduction, such a procedure decreases the contrast on image contours and significantly complicates the study of cranial sutures. Aim: This study aims to apply theoretical and experimental scientific techniques for optimal extraction of useful data for the degree of suture obliteration, from medical tomographic images. Materials: Volumetric images obtained from diagnostic imaging of patients with multislice CT system Toshiba Aquillion64 were used. Methods: We developed a deblurring technique, based on the penalized total variation optimization. We also segmented the output image, using an appropriate threshold in order to separate the suture (region of interest) from the background. Results: After processing the images, the contact between the adjacent cranial bones is clearer and the degree of suture closure could be interpreted more accurately. Acknowledgements: This study was supported by the Bulgarian National Science Fund, grant DN01/15- 20.12.2016.

Mathematics and medicine do not have a long history of close and fruitful cooperation. The application of mathematical methods in medical applications has become viable due to the increasing performance of computers and since more and more digital image data is acquired. While in former times conventional X-ray images — recorded at an analog film — have been used for diagnosis and treatment planning, more and more (digital) 3D datasets, such as Computed Tomography (CT-images) and Magnetic Resonance Imaging (MRI) are acquired for patients with severe diseases. With mathematical methods these 3D datasets may be quantitatively analyzed and visualized such that medical diagnosis and the assessment of therapeutic strategies becomes more reliable and reproducible.

Automatic change detection methods for identifying the changes of serial MR images taken at different times are of great interest to radiologists. The majority of existing change detection methods in medical imaging, and those of brain images in particular, include many preprocessing steps and rely mostly on statistical analysis of magnetic resonance imaging (MRI) scans. Although most methods utilize registration software, tissue classification remains a difficult and overwhelming task. Recently, dictionary learning techniques are being used in many areas of image processing, such as image surveillance, face recognition, remote sensing, and medical imaging. We present an improved version of the EigenBlockCD algorithm, named the EigenBlockCD-2. The EigenBlockCD-2 algorithm performs an initial global registration and identifies the changes between serial MR images of the brain. Blocks of pixels from a baseline scan are used to train local dictionaries to detect changes in the follow-up scan. We use PCA to reduce the dimensionality of the local dictionaries and the redundancy of data. Choosing the appropriate distance measure significantly affects the performance of our algorithm. We examine the differences between [Formula: see text] and [Formula: see text] norms as two possible similarity measures in the improved EigenBlockCD-2 algorithm. We show the advantages of the [Formula: see text] norm over the [Formula: see text] norm both theoretically and numerically. We also demonstrate the performance of the new EigenBlockCD-2 algorithm for detecting changes of MR images and compare our results with those provided in the recent literature. Experimental results with both simulated and real MRI scans show that our improved EigenBlockCD-2 algorithm outperforms the previous methods. It detects clinical changes while ignoring the changes due to the patient's position and other acquisition artifacts.

The majority of model-based learned image reconstruction methods in medical imaging have been limited to uniform domains, such as pixelated images. If the underlying model is solved on nonuniform meshes, arising from a finite element method typical for nonlinear inverse problems, interpolation and embeddings are needed. To overcome this, we present a flexible framework to extend model-based learning directly to nonuniform meshes, by interpreting the mesh as a graph and formulating our network architectures using graph convolutional neural networks. This gives rise to the proposed iterative Graph Convolutional Newton-type Method (GCNM), which includes the forward model in the solution of the inverse problem, while all updates are directly computed by the network on the problem specific mesh. We present results for Electrical Impedance Tomography, a severely ill-posed nonlinear inverse problem that is frequently solved via optimization-based methods, where the forward problem is solved by finite element methods. Results for absolute EIT imaging are compared to standard iterative methods as well as a graph residual network. We show that the GCNM has good generalizability to different domain shapes and meshes, out of distribution data as well as experimental data, from purely simulated training data and without transfer training.

Tomographic imaging is an important disease diagnosis method. Specific image workstation is usually needed to process and visualize 3D medical image gathered from tomographic imaging equipments, because this process is computation intensive. Specific image workstation is large and expensive, which limits size and price of medical tomographic imaging system. To break these limits, this paper presents an embedded 3D medical image processing and visualization platform based on common dual-core processor. The proposed system parallelizes 3D medical image processing and visualization computing process to meet system real-time requirements. Main contributions of this paper includes analysis of core 3D medical data processing and visualization algorithm, program parallelization method analysis and implementation, system performance measurement and analysis. Program analysis results show that the proposed system fully utilizes hardware resource, and meets low-end and space-efficient medical imaging system requirements.

ForewordThis volume of Journal of Physics: Conference Series is dedicated to the scientific contributions presented during the 6th International Workshop on New Computational Methods for Inverse Problems, NCMIP 2016 (http://complement.farman.ens-cachan.fr/NCMIP 2016.html). This workshop took place at Ecole Normale Supérieure de Cachan, on May 20, 2016. The prior editions of NCMIP also took place in Cachan, France, firstly within the scope of ValueTools Conference, in May 2011, and secondly at the initiative of Institut Farman, in May 2012, May 2013, May 2014 and May 2015.The New Computational Methods for Inverse Problems (NCMIP) workshop focused on recent advances in the resolution of inverse problems. Indeed, inverse problems appear in numerous scientific areas such as geophysics, biological and medical imaging, material and structure characterization, electrical, mechanical and civil engineering, and finances. The resolution of inverse problems consists in estimating the parameters of the observed system or structure from data collected by an instrumental sensing or imaging device. Its success firstly requires the collection of relevant observation data. It also requires accurate models describing the physical interactions between the instrumental device and the observed system, as well as the intrinsic properties of the solution itself. Finally, it requires the design of robust, accurate and efficient inversion algorithms. Advanced sensor arrays and imaging devices provide high rate and high volume data; in this context, the efficient resolution of the inverse problem requires the joint development of new models and inversion methods, taking computational and implementation aspects into account. During this one- day workshop, researchers had the opportunity to bring to light and share new techniques and results in the field of inverse problems.The topics of the workshop were: algorithms and computational aspects of inversion, Bayesian estimation, Kernel methods, learning methods, convex optimization, free discontinuity problems, metamodels, proper orthogonal decomposition, reduced models for the inversion, non-linear inverse scattering, image reconstruction and restoration, and applications (bio-medical imaging, nondestructive evaluation...).NCMIP 2016 was a one-day workshop held in May 2016 which attracted around seventy attendees. Each of the submitted papers has been reviewed by two reviewers. There have been eleven accepted papers. In addition, three international speakers were invited to present a longer talk.The workshop was supported by Institut Farman (ENS Cachan, CNRS) and endorsed by the following French research networks: GDR ISIS, GDR MIA, GDR MOA, GDR Ondes. The program committee acknowledges the following research laboratories: CMLA, LMT, LURPA and SATIE.Eric Vourc'h and Thomas RodetWorkshop co-chairs:Eric Vourc'h, SATIE laboratory, Ecole Normale Supérieure de Cachan, CNRS, France Thomas Rodet, SATIE laboratory, Ecole Normale Supérieure de Cachan, CNRS, FranceTechnical program committee:Alexandre Baussard, ENSTA Bretagne, Lab-STICC, France Marc Bonnet, ENSTA, ParisTech, FranceLaure Blanc-Féraud, I3S laboratory and INRIA Nice Sophia-Antipolis, FranceAntonin Chambolle, CMAP, Ecole Polytechnique, CNRS, FranceOliver Dorn, School of Mathematics, University of Manchester, UKCécile Durieu, SATIE, ENS Cachan, CNRS, FranceLaurent Fribourg, LSV, ENS Cachan, CNRS, FranceJerôme Idier, IRCCyN Laboratory, Ecole Centrale de Nantes, FrancePierre-Yves Joubert, IEF, Paris-Sud University, CNRS, FranceMarc Lambert, Geeps Laboratory, CNRS, CentraleSupElec, Paris-Sud University, France Giacomo Oliveri, eledia research center/eledia@L2S group, University of Trento, Italy Dominique Lesselier, L2S Laboratory, CNRS, CentraleSupElec, Paris-Sud University, France Matteo Pastorino, DIBE, University of Genoa, ItalyGabriel Peyré, Ceremade laboratory, University of Paris Dauphine, France Anthony Quinn, Trinity College Dublin, Ireland Christian Rey, Safran Tech, FranceMarco Salucci, eledia research center/eledia@L2S group, University of Trento, Italy Jean-Yves Tourneret, IRIT laboratory, ENSEEIHT (Toulouse), FranceLocal chairs:Virginie Pauchont, CMLA Laboratory, ENS Cachan, Paris-Sud University, CNRS, France Sophie Abriet, SATIE Laboratory, ENS Cachan, Paris-Sud University, CNRS, France Béatrice Bacquet, SATIE Laboratory, ENS Cachan, Paris-Sud University, CNRS, France Aurore Gracia, SATIE Laboratory, ENS Cachan, Paris-Sud University, CNRS, FranceInvited speakersLaure Blanc-Féraud, I3S Laboratory, CNRS UMR 6070, Sophia Antipolis, France Michel Defrise, Department of Nuclear Medicine, Vrije Universiteit Brussel, Belgium Houssem Haddar, INRIA, CMAP, Ecole Polytechnique, Palaiseau, France.

The inverse sonification problem is investigated in this article in order to detect hardly capturing details in a medical image. The direct problem consists in converting the image data into sound signals by a transformation which involves three steps - data, acoustics parameters and sound representations. The inverse problem is reversing back the sound signals into image data. By using the known sonification operator, the inverse approach does not bring any gain in the sonified medical imaging. The replication of the image already known does not help the diagnosis and surgical operation. In order to bring gains in the medical imaging, a new sonification operator is advanced in this paper, by using the Burgers equation of sound propagation. The sonified medical imaging is useful in interpreting the medical imaging that, however powerful they may be, are never good enough to aid tumour surgery. The inverse approach is exercised on several medical images used to surgical operations.

This volume of Journal of Physics: Conference Series is dedicated to the scientific contributions presented during the 3rd International Workshop on New Computational Methods for Inverse Problems, NCMIP 2013 (http://www.farman.ens-cachan.fr/NCMIP_2013.html). This workshop took place at Ecole Normale Supérieure de Cachan, in Cachan, France, on 22 May 2013, at the initiative of Institut Farman. The prior editions of NCMIP also took place in Cachan, France, firstly within the scope of the ValueTools Conference, in May 2011 (http://www.ncmip.org/2011/), and secondly at the initiative of Institut Farman, in May 2012 (http://www.farman.ens-cachan.fr/NCMIP_2012.html). The NCMIP Workshop focused on recent advances in the resolution of inverse problems. Indeed inverse problems appear in numerous scientific areas such as geophysics, biological and medical imaging, material and structure characterization, electrical, mechanical and civil engineering, and finances. The resolution of inverse problems consists of estimating the parameters of the observed system or structure from data collected by an instrumental sensing or imaging device. Its success firstly requires the collection of relevant observation data. It also requires accurate models describing the physical interactions between the instrumental device and the observed system, as well as the intrinsic properties of the solution itself. Finally, it requires the design of robust, accurate and efficient inversion algorithms. Advanced sensor arrays and imaging devices provide high rate and high volume data; in this context, the efficient resolution of the inverse problem requires the joint development of new models and inversion methods, taking computational and implementation aspects into account. During this one-day workshop, researchers had the opportunity to bring to light and share new techniques and results in the field of inverse problems.The topics of the workshop were: algorithms and computational aspects of inversion, Bayesian estimation, kernel methods, learning methods, convex optimization, free discontinuity problems, metamodels, proper orthogonal decomposition, reduced models for the inversion, non-linear inverse scattering, image reconstruction and restoration, and applications (bio-medical imaging, non-destructive evaluation...).NCMIP 2013 was a one-day workshop held in May 2013 which attracted around 60 attendees. Each of the submitted papers has been reviewed by three reviewers. Among the accepted papers, there are seven oral presentations, five posters and one invited poster (On a deconvolution challenge presented by C Vonesch from EPFL, Switzerland). In addition, three international speakers were invited to present a longer talk. The workshop was supported by Institut Farman (ENS Cachan, CNRS) and endorsed by the following French research networks (GDR ISIS, GDR Ondes, GDR MOA, GDR MSPC). The program committee acknowledges the following research laboratories CMLA, LMT, LSV, LURPA, SATIE. Laure Blanc-Féraud and Pierre-Yves Joubert Workshop co-chair Laure Blanc-Féraud, I3S laboratory and INRIA Nice Sophia-Antipolis, France Pierre-Yves Joubert, IEF, Paris-Sud University, CNRS, France Technical program committee Gilles Aubert, J-A Dieudonné Laboratory, CNRS and University of Nice-Sophia Antipolis, France Nabil Anwer, LURPA, ENS Cachan, France Alexandre Baussard, ENSTA Bretagne, Lab-STICC, France Marc Bonnet, ENSTA, ParisTech, France Antonin Chambolle, CMAP, Ecole Polytechnique, CNRS, France Oliver Dorn, School of Mathematics, University of Manchester, UK Cécile Durieu, SATIE, ENS Cachan, CNRS, France Gérard Favier, I3S Laboratory, University of Nice Sophia-Antipolis, France Mário Figueiredo, Instituto Superior Técnico, Lisbon, Portugal Laurent Fribourg, LSV, ENS Cachan, CNRS, France Marc Lambert, L2S Laboratory, CNRS, SupElec, Paris-Sud University, France Dominique Lesselier, L2S Laboratory, CNRS, SupElec, Paris-Sud University, France Matteo Pastorino, DIBE, University of Genoa, Italy Christian Rey, LMT, ENS Cachan, CNRS, France Simon Setzer, Saarland University, Germany Cedric Vonesch, EPFL, SwitzerlandLocal chair Sophie Abriet, SATIE Laboratory, ENS Cachan, France Béatrice Bacquet, SATIE Laboratory, ENS Cachan, France Lydia Matijevic, LMT Laboratory, ENS Cachan FranceInvited speakers Jérôme Idier, IRCCyN (UMR CNRS 6597), Ecole Centrale de Nantes, France Massimo Fornasier, Faculty of Mathematics, Technical University of Munich, Germany Matthias Fink, Institut Langevin, ESPCI, Université Paris Diderot, France

Rapid localization of ROI (Region of Interest) for tomographic medical images (TMIs) is an important foundation for efficient image reading, computer-aided education, and well-informed rights of patients. However, due to the multimodality of clinical TMIs, the complexity of anatomy, and the deformation of organs caused by diseases, it is difficult to have a universal and low-cost method for ROI organ localization. This article focuses on actual concerns of TMIs from medical students, engineers, interdisciplinary researchers, and patients, exploring a universal registration method between the clinical CT/MRI dataset and CVH (Chinese Visible Human) to locate the organ ROI in a low-cost and lightweight way. The proposed method is called Two-step Progressive Registration (TSPR), where the first registration adopts "eye-nose triangle" features to determine the spatial orientation, and the second registration adopts the circular contour to determine the spatial scale, ultimately achieving CVH anatomical knowledge automated mapping. Through experimentation with representative clinical TMIs, the registration results are capable of labeling the ROI in the images well and can adapt to the deformation problem of ROI, as well as local extremum problems that are prone to occur in inter-subject registration. Unlike the ideal requirements for TMIs' data quality in laboratory research, TSPR has good adaptability to incomplete and non-thin-layer quality in real clinical data in a low-cost and lightweight way. This helps medical students, engineers, and interdisciplinary researchers independently browse images, receive computer-aided education, and provide patients with better access to well-informed services, highlighting the potential of digital public health and medical education.

Organ-on-a-chip (OOC) is a new type of biochip technology. Various types of OOC systems have been developed rapidly in the past decade and found important applications in drug screening and precision medicine. However, due to the complexity in the structure of both the chip-body itself and the engineered-tissue inside, the imaging and analysis of OOC have still been a big challenge for biomedical researchers. Considering that medical imaging is moving towards higher spatial and temporal resolution and has more applications in tissue engineering, this paper aims to review medical imaging methods, including CT, micro-CT, MRI, small animal MRI, and OCT, and introduces the application of 3D printing in tissue engineering and OOC in which medical imaging plays an important role. The achievements of medical imaging assisted tissue engineering are reviewed, and the potential applications of medical imaging in organoids and OOC are discussed. Moreover, artificial intelligence - especially deep learning - has demonstrated its excellence in the analysis of medical imaging; we will also present the application of artificial intelligence in the image analysis of 3D tissues, especially for organoids developed in novel OOC systems.

The diagnosis of patients is significantly influenced by medical imaging. These images are frequently transmitted over the network for timely and accurate diagnosis. On the external networks, these medical images are prone to attack. Therefore, medical images are encrypted before transmission; this single layer of protection is not sufficient. Recently, the idea of embedding patient personal information, such as patient identity and medical records, in medical images via reversible data hiding (RDH) in the encrypted image has been proposed. Still, these schemes suffer from low embedding rates. In this paper, a RDH mechanism based on an interpolation scheme and histogram shifting is highlighted. In the proposed method, the achievable embedding capacity (EC) is 1 bit per pixel (bpp). The image encryption is achieved using blockchain technology based on 3 keys. The proposed blockchain mechanism is safe from external attacks. The results are compared with the state-of-the-art methods for both digital and medical images to validate the usefulness of the proposed method.

Summary Introduction : There has been a rapid development of deep learning (DL) models for medical imaging. However, DL requires a large labeled dataset for training the models. Getting large-scale labeled data remains a challenge, and multi-center datasets suffer from heterogeneity due to patient diversity and varying imaging protocols. Domain adaptation (DA) has been developed to transfer the knowledge from a labeled data domain to a related but unlabeled domain in either image space or feature space. DA is a type of transfer learning (TL) that can improve the performance of models when applied to multiple different datasets. Objective : In this survey, we review the state-of-the-art DL-based DA methods for medical imaging. We aim to summarize recent advances, highlighting the motivation, challenges, and opportunities, and to discuss promising directions for future work in DA for medical imaging. Methods : We surveyed peer-reviewed publications from leading biomedical journals and conferences between 2017-2020, that reported the use of DA in medical imaging applications, grouping them by methodology, image modality, and learning scenarios. Results : We mainly focused on pathology and radiology as application areas. Among various DA approaches, we discussed domain transformation (DT) and latent feature-space transformation (LFST). We highlighted the role of unsupervised DA in image segmentation and described opportunities for future development. Conclusion : DA has emerged as a promising solution to deal with the lack of annotated training data. Using adversarial techniques, unsupervised DA has achieved good performance, especially for segmentation tasks. Opportunities include domain transferability, multi-modal DA, and applications that benefit from synthetic data.

Memetic improved cuckoo search algorithm for automatic B-spline border approximation of cutaneous melanoma from macroscopic medical images

Pre-training deep learning models on ImageNet and transferring the models to medical image applications facilitate to improve the medical image analysis and reduce the need for labeled medical image data. However, some images from ImageNet may be fundamentally different from medical images in feature representation and lead to the negative transfer effects. To deal with this issue, we propose a novel strategy based on evidence theory to select reliable instances from ImageNet for medical image domain adaptation. Specifically, we formulate an evidential mass function to measure the ignorance and reliability of the images from ImageNet with respect to the classification tasks of medical images. Through selecting reliable instances with low ignorance degree from ImageNet, we can enhance the transfer performances of deep neural networks in medical image domain adaptation. Moreover, the proposed data selection strategy is independent of specific learning algorithm and can be viewed as a common preprocessing technique. Numerical experiments on tomography images, X-Ray images, and ultrasound images are given to comprehensively demonstrate the effectiveness of the selection strategy.