SEARCH · PubMed Health
Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
The field of computer vision is devoted to discovering algorithms, data representations, and computer architectures that embody the principles underlying visual capabilities. This article describes how the field of computer (and robot) vision has evolved, particularly over the past 20 years, and introduces its central methodological paradigms.
Computer vision (CV), a computerized method to analyze digital images (e.g., CT scans), and computer graphics (CG), a set of computer programs for displaying two, three- or four-dimensional data, are recent computer techniques which are appropriate to assist functional stereotactic surgery. CV and CG are useful for the evaluation of spatial relations between anatomical structures and the sites of spontaneous neuronal noise or of electrophysiological stimulation data gathered during stereotactic interventions for movement disorders. CV and CG can be employed for stereotactic operation planning, during the operation for target point evaluation and for further postoperative data analysis.
As the number of available three dimensional coordinates of proteins increases, it is now recognized that proteins from different families and topologies are constructed from independent motifs. Detection of specific structural motifs within proteins aids in understanding their role and the mechanism of their operation. To aid in identification and use of these motifs it has become necessary to develop efficient methods for systematic scanning of structural databases. To date, methods of structural protein comparison suffer from at least one of the following limitations: (1) are not fully automated (require human intervention), (2) are limited to relatively similar structures, (3) are constrained to linear alignments of the structures, (4) are sensitive to insertions, deletions or gaps in the sequences or (5) are very time consuming. We present a method to overcome the above limitations. The method discovers and ranks every piece of structural similarity between the structures compared, thus allowing the simultaneous detection of real 3-D motifs in different domains, between domains, in active sites, surfaces etc. The method uses the Geometric Hashing Paradigm which is an efficient technique originally developed for Computer Vision. The algorithm exploits the geometrical constraints of rigid objects, it is especially geared towards recognition of partial structures in rigid objects belonging to large data bases and is straightforwardly parallelizable. Computer Vision techniques are for the first time applied to molecular structure comparison, resulting in an efficient, fully automated tool. The method has been tested in a number of cases, including comparisons of the haemoglobins, immunoglobulins, serine proteinases, calcium binding proteins, DNA binding proteins and others. In all examples our results were equivalent to the published results from previous methods and in some cases additional structural information was obtained by our method.
Macromolecules carrying biological information often consist of independent modules containing recurring structural motifs. Detection of a specific structural motif within a protein (or DNA) aids in elucidating the role played by the protein (DNA element) and the mechanism of its operation. The number of crystallographically known structures at high resolution is increasing very rapidly. Yet, comparison of three-dimensional structures is a laborious time-consuming procedure that typically requires a manual phase. To date, there is no fast automated procedure for structural comparisons. We present an efficient O(n3) worst case time complexity algorithm for achieving such a goal (where n is the number of atoms in the examined structure). The method is truly three-dimensional, sequence-order-independent, and thus insensitive to gaps, insertions, or deletions. This algorithm is based on the geometric hashing paradigm, which was originally developed for object recognition problems in computer vision. It introduces an indexing approach based on transformation invariant representations and is especially geared toward efficient recognition of partial structures in rigid objects belonging to large data bases. This algorithm is suitable for quick scanning of structural data bases and will detect a recurring structural motif that is a priori unknown. The algorithm uses protein (or DNA) structures, atomic labels, and their three-dimensional coordinates. Additional information pertaining to the structure speeds the comparisons. The algorithm is straightforwardly parallelizable, and several versions of it for computer vision applications have been implemented on the massively parallel connection machine. A prototype version of the algorithm has been implemented and applied to the detection of substructures in proteins.
This paper reviews some basic computer vision techniques and speculates about their possible relevance to the modeling of human visual processes. Special emphasis is given to image segmentation techniques and how they relate to processes of perceptual organization, such as those embodied in the Gestalt "laws."
Descriptions of physical properties of visible surfaces, such as their distance and the presence of edges, must be recovered from the primary image data. Computational vision aims to understand how such descriptions can be obtained from inherently ambiguous and noisy data. A recent development in this field sees early vision as a set of ill-posed problems, which can be solved by the use of regularization methods. These lead to algorithms and parallel analog circuits that can solve 'ill-posed problems' and which are suggestive of neural equivalents in the brain.
Automatic detection of several features characteristic of basal cell epitheliomas is described. The features selected for this feasibility study are semitranslucency, telangiectasia, ulcer, crust, and tumor border. Image processing methods used in this study include frequency analysis of the Fourier transform of the image, the Sun-Wee texture analysis algorithm, and several other image analysis techniques suitable for skin photographs. This image analysis software is designed for use with AI/DERM, an expert system that models diagnosis of skin tumors by dermatologists.
Developments in finite-difference time-domain (FD-TD) computational modeling of Maxwell's equations, super-computer technology, and computed tomography (CT) imagery open the possibility of accurate numerical simulation of electromagnetic (EM) wave interactions with specific, complex, biological tissue structures. One application of this technology is in the area of treatment planning for EM hyperthermia. In this paper, we report the first highly automated CT image segmentation and interpolation scheme applied to model patient-specific EM hyperthermia. This novel system is based on sophisticated tools from the artificial intelligence, computer vision, and computer graphics disciplines. It permits CT-based patient-specific hyperthermia models to be constructed without tedious manual contouring on digitizing pads or CRT screens. The system permits in principle near real-time assistance in hyperthermia treatment planning. We apply this system to interpret actual patient CT data, reconstructing a 3-D model of the human thigh from a collection of 29 serial CT images at 10 mm intervals. Then, using FD-TD, we obtain 2-D and 3-D models of EM hyperthermia of this thigh due to a waveguide applicator. We find that different results are obtained from the 2-D and 3-D models, and conclude that full 3-D tissue models are required for future clinical usage.
This study describes a computer vision approach for sensing the environment with the intent of helping people with a visual impairment. The principal goal in applying computer vision is to exploit, in an optimal fashion, the information acquired by the camera(s) to yield useful descriptions of the viewed environment. The objective is to seek efficient and reliable guidance cues in order to improve the mobility needs of individuals with a visual impairment. In this research direction, the following problems are identified and addressed: 1) the vision system design; 2) establishment of the mapping principles between the two-dimensional (2-D) camera images and the three-dimensional (3-D) real world; 3) development of appropriate imaging techniques for the interpretation of the 2-D images; and, 4) establishment of a communication link between the vision system and the user. The soundness of this research direction is assessed by means of a theoretical framework and experimental evaluations.
A large number of computer vision algorithms for finding intensity edges, computing motion, depth, and color, and recovering the three-dimensional shape of objects have been developed within the framework of minimizing an associated "energy" or "cost" functional. Particularly successful has been the introduction of binary variables coding for discontinuities in intensity, optical flow field, depth, and other variables, allowing image segmentation to occur in these modalities. The associated nonconvex variational functionals can be mapped onto analog, resistive networks, such that the stationary voltage distribution in the network corresponds to a minimum of the functional. The performance of an experimental analog very-large-scale integration (VLSI) circuit implementing the nonlinear resistive network for the problem of two-dimensional surface interpolation in the presence of discontinuities is demonstrated; this circuit is implemented in complementary metal oxide semiconductor technology.
The sensation of depth that is obtained with human binocular vision results from the differences in the projection of the world onto the two retinae. The process entails solving the problem of stereo correspondence, which involves choosing the correct matches between left and right image features. Many computational models of stereo vision assume a uniqueness constraint on stereo matching-that is, each feature identified in one image should eventually be matched with only one feature in the other image. This constraint would seem to be justified, as allowing non-unique matches would be tantamount to supposing that the scene entities to which matches relate are in two places at once. The value of the uniqueness constraint for eliminating false matches has been demonstrated in a variety of stereo algorithms. Yet on the basis of psychophysical results Weinshall concluded that it was not used by humans in dealing with certain types of ambiguous random-dot stereograms. We have now tested how Weinshall's stereograms are dealt with by PMF, a stereo algorithm which uses a unique-matches selection procedure in conjunction with a purely local similar-disparity support scheme. We found that PMF produces results that are closely analogous to the psychophysical results. This suggests that Weinshall's experiments should not be interpreted as evidence that the human stereo mechanism establishes non-unique matches.
Figure-ground discrimination and pattern discrimination represent one of the most important problems in computer vision. Based on computational neuronal network model proposed by Reichardt et al. for figure-ground discrimination in the visual system of the fly, entire systematic computer simulations were carried out under nonstationary conditions. Results show that the model network and simulations have predictive power for behavioural experimental results. This paper proposes that motion perception and elementary pattern discrimination in human visual system may be mediated by some kind of figure-ground system with movement detectors as input layer.
Computer vision techniques implemented on an IBM PC/AT have been applied to the study of microvascular permeability and interstitial diffusion in dorsal skin flap chamber preparations of hamsters. Experimental data was obtained for the leakage of fluorescent labelled dextran (70,000 daltons) after a precisely controlled mild degree of localized thermal trauma and compared with control data acquired prior to burn injury. Computer vision analysis techniques were applied to convert the fluorescent images into two-dimensional concentration maps. Interstitial diffusion coefficient values were computed from measured extravascular concentration profiles around a vessel of interest, assuming cylindrical or rectangular geometry, and optimally fitting a diffusion model to the data. An increase in the apparent diffusivity after mild thermal trauma was observed. Novel techniques were applied to solve hardware problems related to data acquisition and analysis, and a new library of software was developed to handle specific image processing requirements.
RATIONALE AND OBJECTIVE: To alert radiologists to possible nodule locations and subsequently to reduce the number of false-negative diagnoses, the authors are developing a computer-aided diagnostic (CAD) scheme for the detection of lung nodules in digital chest images. METHODS: A computer-vision scheme was applied to photofluorographic films obtained in a mass survey for detection of asymptomatic lung cancer in Japan. Ninety-five patients with abnormal test results who had primary and metastatic lung cancers and 103 patients with normal test results were included. RESULTS: The sensitivity of the computer output was comparable with that of physicians in this mass survey (62%). The computer detected approximately 40% of all nodules missed in the mass survey, but missed 17 true-positive results identified in the mass survey. The CAD scheme produced an average of 15 false-positive findings per image. CONCLUSION: If the number of false-positive results can be significantly reduced, computer-vision schemes such as this may have a role in lung cancer screening programs.
Although gradient schemes for detecting the motion of images and measuring their velocities are commonly used in computer vision, and although there is increasing evidence to support the existence of such schemes in biological vision, little attention has been directed to suggesting how such computations might be realized by neural hardware. This paper proposes two simple models, consisting of physiologically realistic networks of neurons, that approximate the gradient scheme. Computer simulations demonstrate that the models measure the speed of an object or pattern independently of its structural properties.
Computer algorithms have been developed for several early vision processes, such as edge detection, stereopsis, motion, texture, and color, that give separate cues to the distance from the viewer of three-dimensional surfaces, their shape, and their material properties. Not surprisingly, biological vision systems still greatly outperform computer vision programs. One of the keys to the reliability, flexibility, and robustness of biological vision systems is their ability to integrate several visual cues. A computational technique for integrating different visual cues has now been developed and implemented with encouraging results on a parallel supercomputer.
Modern livestock breeding has mastered genotyping. Genome-wide association studies, genomic selection, and SNP arrays enable genetic merit prediction at lower cost. However, phenotyping remains the bottleneck, as manual measurement is slow, expensive, subjective, and unable to capture spatial or temporal trait organization. Digital phenotyping via artificial intelligence could resolve this, but deep learning requires thousands of labelled examples, impractical when phenotyping cost itself limits datasets to hundreds of individuals. This creates a paradox: AI could accelerate phenotyping but requires large numbers of samples to train the models. Here, we demonstrate that integrating computer vision with machine learning offers sample-efficient digital phenotyping using eggshell colour as a model system. Rather than learning features from scratch (deep learning), we engineer physically motivated features via Wavelet transforms that decompose images into multi-scale spatial components. Wavelet features captured 14.2 percentage points more variance (R2 = 0.976 vs. 0.834, p < 0.001) than standard colorimetry, with 50% better sample efficiency (achieving at n = 60 what colorimetry required n = 120). Variance decomposition revealed 77% of discriminative capacity derives from spatial patterns (bands, spots, gradients) invisible to scalar averages. Additionally, we identified "cryptic phenotypes" (3.3%) where spatial patterns contradicted average colour, cases where colorimeters failed but Wavelets succeeded. The underlying principle-that spatial decomposition can recover organizational information lost by scalar averaging-may be applicable to other traits with spatial or temporal structure, such as marbling, dermatitis, or pigmentation rhythms, although whether comparable performance gains would be observed remains to be tested empirically. Hence, for breeding programs implementing genomic selection, computer vision-based digital phenotyping captures complex trait variation without massive training datasets, addressing the bottleneck that increasingly limits genetic progress as genotyping becomes trivial.
BACKGROUND: Transthoracic echocardiography (TTE) is the most commonly performed cardiac imaging modality with over 30 million studies annually. Demand for timely expert interpretation continues to outpace capacity, creating diagnostic delays and inter-observer variability that impact patient care. Recent research has suggested computer vision artificial intelligence (AI) models can generate accurate preliminary comprehensive TTE reports, however, prospective evaluation is needed to determine whether AI-assisted TTE interpretation can improve clinician efficiency while preserving diagnostic accuracy. METHODS: AI ECHO INSIGHT is a prospective randomized blinded clinical trial conducted at Kaiser Permanente Northern California that will evaluate 1200 historical TTE studies (1000 consecutive unselected studies plus 200 with moderate or greater valvular disease) interpreted using three workflows: (1) AI-generated preliminary report finalized by a blinded cardiologist (AI-assisted); (2) cardiologist-generated preliminary report finalized by a blinded cardiologist (cardiologist-assisted); and (3) sonographer-generated preliminary report finalized by a blinded cardiologist (sonographer-assisted). The primary outcome is the rate of substantial change between preliminary and final reports, comparing the AI-assisted workflow to the pooled cardiologist-assisted and sonographer-assisted workflows. Secondary outcomes include cardiologist interpretation time for report finalization, superiority testing for diagnostic accuracy, and reporting consistency. CONCLUSION: AI ECHO INSIGHT is a prospective randomized blinded clinical trial evaluating the clinical impact of AI-assisted TTE interpretation on diagnostic accuracy, cardiologist efficiency, and reporting consistency in real-world echocardiography workflows. TRIAL REGISTRATION: ClinicalTrials.gov registration number NCT07229300.