Narrow Results

Image denoising is essential for medical image analysis due to noise introduced by various acquisition methods and efforts to reduce radiation exposure. Noise in medical imaging from equipment, patient variability and environmental factors requires effective denoising to improve image quality and diagnostics. To address these challenges, a Multilevel Convolutional Neural Network with an optimized Visual Attention Network (MCVAN) is developed specifically for image denoising to enhance the Peak Signal-to-Noise Ratio (PSNR). Leopard Seal Optimization (LSO) is fine-tuning the parameters of the network, enhancing denoising performance. The motivation is to address the critical need for effective image denoising in medical imaging. The innovation of this research lies in the development of a MCVAN, developed for image denoising. The LSO to fine-tune parameters further enhances the denoising performance. This architecture effectively adapts to varying noise levels in input images, aiming to significantly reduce noise in medical images for improved diagnostic accuracy and visual clarity. Experimental results show an average PSNR of 43.79dB and a Structural Similarity Index Measure (SSIM) of 0.863 and the MCVAN achieves accuracy (99.9%), precision (99.9%), recall (99.9%) and $F$1-score (99.9%). Overall, the MCVAN demonstrates superior effectiveness in image denoising, surpassing existing techniques in both quality and efficiency.

Medical imaging holds significant importance in disease diagnosis and surgical preparation, necessitating higher standards for the storage and transmission of medical images. We propose a method for compressed sensing reconstruction of medical imaging based on untrained convolutional neural networks. This method does not require any training data, offering vast potential in the field of medical imaging. We employ the generator model of Deep Convolutional Generative Adversarial Networks (DCGAN) as the structure for the untrained convolutional neural network. We utilized the Musculoskeletal Radiographs (MURA) and Open Access Series of Imaging Studies (OASIS) datasets to demonstrate that our method significantly improved the reconstruction effect in medical imaging, regardless of whether the sampling rate was high or low.

The traveling salesman problem (TSP) is a prototypical problem of combinatorial optimization and, as such, it has received considerable attention from neural-network researchers seeking quick, heuristic solutions. An early stage in many computer vision tasks is the extraction of object shape from an image consisting of noisy candidate edge points. Since the desired shape will often be a closed contour, this problem can be viewed as a version of the TSP in which we wish to link only a subset of the points/cities (i.e. the "noisefree" ones). None of the extant neural techniques for solving the TSP can deal directly with this case. In this paper, we present a simple but effective modification to the (analog) elastic net of Durbin and Willshaw which shifts emphasis from global to local behavior during convergence, so allowing the net to ignore some image points. Unlike the original elastic net, this semi-localized version is shown to tolerate considerable amounts of noise. As an example practical application, we describe the extraction of "pseudo-3D" human lung outlines from multiple preprocessed magnetic resonance images of the torso. An effectiveness measure (ideally zero) quantifies the difference between the extracted shape and some idealized shape exemplar. Our method produces average effectiveness scores of 0.06 for lung shapes extracted from initial semi-automatic segmentations which define the noisefree case. This deteriorates to 0.1 when extraction is from a noisy edge-point image obtained fully-automatically using a feedforward neural network.

The automation in the diagnosis of medical images is currently a challenging task. The use of Computer Aided Diagnosis (CAD) systems can be a powerful tool for clinicians, especially in situations when hospitals are overflowed. These tools are usually based on artificial intelligence (AI), a field that has been recently revolutionized by deep learning approaches. These alternatives usually obtain a large performance based on complex solutions, leading to a high computational cost and the need of having large databases. In this work, we propose a classification framework based on sparse coding. Images are first partitioned into different tiles, and a dictionary is built after applying PCA to these tiles. The original signals are then transformed as a linear combination of the elements of the dictionary. Then, they are reconstructed by iteratively deactivating the elements associated with each component. Classification is finally performed employing as features the subsequent reconstruction errors. Performance is evaluated in a real context where distinguishing between four different pathologies: control versus bacterial pneumonia versus viral pneumonia versus COVID-19. Our system differentiates between pneumonia patients and controls with an accuracy of 97.74%, whereas in the 4-class context the accuracy is 86.73%. The excellent results and the pioneering use of sparse coding in this scenario evidence that our proposal can assist clinicians when their workload is high.

The combination of different sources of information is currently one of the most relevant aspects in the diagnostic process of several diseases. In the field of neurological disorders, different imaging modalities providing structural and functional information are frequently available. Those modalities are usually analyzed separately, although a joint of the features extracted from both sources can improve the classification performance of Computer-Aided Diagnosis (CAD) tools. Previous studies have computed independent models from each individual modality and combined them in a subsequent stage, which is not an optimum solution. In this work, we propose a method based on the principles of siamese neural networks to fuse information from Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET). This framework quantifies the similarities between both modalities and relates them with the diagnostic label during the training process. The resulting latent space at the output of this network is then entered into an attention module in order to evaluate the relevance of each brain region at different stages of the development of Alzheimer’s disease. The excellent results obtained and the high flexibility of the method proposed allow fusing more than two modalities, leading to a scalable methodology that can be used in a wide range of contexts.

The skin is the largest (and the most exposed) organ of the body both in terms of surface area and weight. Its care is of great importance for both aesthetics and health issues. Often, the skin appearance gives us information about the skin health status as well as hints at the biological age. Therefore, the skin surface characterization is of great significance for dermatologists as well as for cosmetic scientists in order to evaluate the effectiveness of medical or cosmetic treatments. So far, no in vivo measurements regarding skin topography characterization could be achieved routinely to evaluate skin aging. This work describes how a portable capacitive device, normally used for fingerprint acquisition, can be utilized to achieve measures of skin aging routinely. The capacitive images give a high resolution (50 μm) representation of skin topography, in terms of wrinkles and cells. In this work, we have addressed the latter: through image segmentation techniques, cells have been localized and identified and a feature related to their area distribution has been generated. Accurate experiments accomplished in vivo show how the feature we conceived is linearly related to skin aging. Besides, since this finding has been achieved using a low cost portable device, this could boost research in this field as well as open doors to an application based on an embedded system.

Quantitative evaluation of the changes in skin topographic structures are of great importance in the dermocosmetic field to assess subjects response to medical or cosmetic treatments. Although many devices and methods are known to measure these changes, they are not suitable for a routine approach and most of them are invasive. Moreover, it has always been difficult to give a measure of the skin health status as well as of the human aging process by simply analyzing the skin surface appearance. This work describes how a portable capacitive device could be utilized to achieve measurements of skin ageing in vivo and routinely. The capacitive images give a high resolution representation of the skin micro-relief, both in terms of skin surface tissue and wrinkles. In a previous work we dealt with the former; here we have addressed the latter. The algorithm we have developed allowed us to extract two original features from wrinkles: the first is based on photometric properties while the second has been achieved through the multiresolution analysis of the wavelet transform. Accurate experiments accomplished on 87 subjects show how the features we conceived are related to skin ageing.

Computed tomography (CT) technologies have been considered for a long time as one of the most effective medical imaging tools for morphological analysis of body parts. Contrast Enhanced CT (CE-CT) also allows emphasising details of tissue structures whose heterogeneity, inspected through visual analysis, conveys crucial information regarding diagnosis and prognosis in several clinical pathologies. Recently, Dynamic CE-CT (DCE-CT) has emerged as a promising technique to perform also functional hemodynamic studies, with wide applications in the oncologic field. DCE-CT is based on repeated scans over time performed after intravenous administration of contrast agent, in order to study the temporal evolution of the tracer in 3D tumour tissue. DCE-CT pushes towards an intensive use of computers to provide automatically quantitative information to be used directly in clinical practice. This requires that visual analysis, representing the gold-standard for CT image interpretation, gains objectivity.

This work presents the first automatic approach to quantify and classify the lung tumour heterogeneities based on DCE-CT image sequences, so as it is performed through visual analysis by experts. The approach developed relies on the spatio-temporal indices we devised, which also allow exploiting temporal data that enrich the knowledge of the tissue heterogeneity by providing information regarding the lesion status.

Over recent years, non-rigid registration has become a major issue in medical imaging. It consists in recovering a dense point-to-point correspondence field between two images, and usually takes a long time. This is in contrast to the needs of a clinical environment, where usability and speed are major constraints, leading to the necessity of reducing the computation time from slightly less than an hour to just a few minutes. As financial pressure makes it hard for healthcare organizations to invest in expensive high-performance computing (HPC) solutions, cluster computing proves to be a convenient solution to our computation needs, offering a large processing power at a low cost. Among the fast and efficient non-rigid registration methods, we chose the demons algorithm for its simplicity and good performances. The parallel implementation decomposes the correspondence field into spatial blocks, each block being assigned to a node of the cluster. We obtained an acceleration of 11 by using 15 2GHz PC's connected through a 1GB/s Ethernet network and reduced the computation time from 40min to 3min30. In order to further optimize the costs and the maintenance load, we investigate in the second part the transparent use of shared computing resources, either through a graphic client or a Web one.

This paper describes the implementation of a parallel image processing algorithm, the aim of which is to give good contrast enhancement in real time, especially on the boundaries of an object of interest defined by a grey homogeneity (for example, an object of medical interest having a functional or morphologic homogeneity, like a bone or tumor). The implementation of a neural network algorithm which does this contrast enhancement has been done on a SIMD massively parallel machine (a MasPar of 8192 processors) and the communication between its processors has been optimized.

Graphene oxide colloid has been widely used in the synthesis of various graphene-based materials. Graphene oxide sheets, with a low bending rigidity, can be folded when assembled in aqueous phase. A simple but industrial scalable way, aerosol processing, can be used to fabricate folded graphene-based materials. These folded materials can carry various cargo materials and be used in different applications such as time-controlled drug release, medical imaging enhancement, catalyst support and energy related areas. The aerosol synthesis of folded graphene-based materials can also be easily extended to fabricate hybrid nanomaterials without any complicated chemistries.

Periodicity (in time or space) is a part and parcel of every living being: one can see, hear and feel it. Everyday examples are locomotion, respiration and heart beat. The reinforced N-dimensional periodicity over two or more crystalline solids results in the so-called phononic band gap crystals. These can have dramatic consequences on the propagation of phonons, vibrations and sound. The fundamental physics of cleverly fabricated phononic crystals can offer a systematic route to realize the Anderson localization of sound and vibrations. As to the applications, the phononic crystals are envisaged to find ways in the architecture, acoustic waveguides, designing transducers, elastic/acoustic filters, noise control, ultrasonics, medical imaging and acoustic cloaking, to mention a few. This review focuses on the brief sketch of the progress made in the field that seems to have prospered even more than was originally imagined in the early nineties.

This paper proposes a method for extracting the human hippocampus based on multiscale structure matching scheme. Focusing on the feature that an overextraction occurs on anatomically specific place, the method detects the redundancy by comparing with given desired models. Since each of the desired models has information about locations of their redundant segments, the place of corresponding redundancy can be specified on the overextracted object. Then, subtle intensity difference around their connecting place is investigated to separate the hippocampus and redundancy. The matching process can proceed in parallel for various types of redundancy and individual variances. Qualitative evaluation of a physician shows that our method can detect the redundancies and extract hippocampus correctly.

In this paper, we describe a method of automatic 3D–2D projective registration between the 3D (i.e. polygonal face surface derived from CT or MRI data) and the 2D faces of the same individual in the photographs. Our task is to make a realistic 3D model face for post-surgical simulation by pasting color textures accurately on the face surface. We utilize edge features such as external edge (facial outline) and internal edges like eye, nose and mouth edges from both the 3D face and photographs for matching. We define 3D edge as a set of 3D surface points which is 2D edge on the projected space. We choose 3D edges within the specific regions alone by automatically categorizing the 3D face into eye, nose, mouth and ear regions using a knowledge-based technique. Experimentally we have shown that for human face matching selected region-based edge yields better matching accuracy than that of the usual edge. Moreover, we average the root mean square (RMS) measures of the selected facial regions rather than computing a single RMS measure to obtain matching uniformity over the entire region.

A novel method to obtain point correspondence in pairs of images is presented. Our approach is based on automatically establishing correspondence between linear structures which appear in images using robust features such as orientation, width and curvature extracted from those structures. The extracted points can be used to register sets of images. The potential of the developed approach is demonstrated on mammographic images.

Extracting the human brain from magnetic resonance head scans is difficult because of its highly convoluted and nonuniform geometry. A technique based on Non-Uniform Rational B-Splines (NURBS) surfaces and energy minimizing deformable models to extract and visualize the brain surface patterns accurately from magnetic resonance head scans is presented. The weighting parameter that comes with the NURBS definition is explored to attract the surface into regions showing high curvature. The weight at each control point is adjusted automatically according to the curvature properties of the evolving surface. This process facilitates a deformable model with increased local flexibility that adapts to complex geometrical features of the brain surface. The results show that the proposed model is capable of capturing the correct brain surface with a higher accuracy than the existing techniques.

Artificial Intelligence (AI) has promoted countless contributions in the field of healthcare and medical imaging. In this paper, we thoroughly analyze peer-reviewed research findings/articles on AI-guided tools for Covid-19 analysis/screening using chest X-ray images in the year 2020. We discuss on how far deep learning algorithms help in decision-making. We identify/address data collections, methodical contributions, promising methods, and challenges. However, a fair comparison is not trivial as dataset sizes vary over time, throughout the year 2020. Even though their unprecedented efforts in building AI-guided tools to detect, localize, and segment Covid-19 cases are limited to education and training, we elaborate on their strengths and possible weaknesses when we consider the need of cross-population train/test models. In total, with search keywords: (Covid-19 OR Coronavirus) AND chest x-ray AND deep learning AND artificial intelligence AND medical imaging in both PubMed Central Repository and Web of Science, we systematically reviewed 58 research articles and performed meta-analysis.

Expert systems and image understanding have traditionally been considered as two separate application fields of artificial intelligence (AI). In this paper it is shown, however, that the idea of building an expert system for image understanding may be fruitful. Although this paper may serve as a framework for situating existing works on knowledge-based vision, it is not a review paper. The interested reader will therefore be referred to some recommended survey papers in the literature.

This paper presents a load-balanced parallelization of the well known Marching-Cubes algorithm, that aims at constructing an iso-surface in a 3D image. We first derive a modelization for the computation time as a function of the generated surface complexity. The workload associated to each slice of the input data is evaluated by counting the number of vertices that will be generated on that slice. The slices are then locally redistributed to ensure a balanced workload. We give an upper bound on the number of polygons of the triangulation, and present a family of surfaces whose number of triangles tends to this bound. This analysis allows us to foresee (and thus to allocate) the memory size needed for the data structures and to assign to each vertex a unique global reference.

Experiments done on an Intel Paragon machine are given both for synthetic and medical images. They show the usefulness of our dynamic data redistribution scheme.

Magnetic resonance imaging (MRI) has become a widely used research and clinical tool in the study of the human brain. The ability to robustly and accurately quantify repeatable morphological measures from such data is aided by the ability to accurately segment the MRI data set into homogeneous regions such as gray matter, white matter, and cerebro spinal fluid. The large amount of data associated with typical MRI scans makes completely manual segmentation prohibitive on a large scale. In this paper an efficient approach to the segmentation of such MR imagery is presented. The approach uses an estimation-theoretic interpretation of the segmentation problem to develop a computationally efficient, statistically-based recursive technique for its solution. Being statistically based, the method also provides associated measures of uncertainty of the resulting estimates, which are extremely important both for evaluation of the estimates as well as their combination with other sources of information.