PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Computer vision”

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.

At least 37 records · Page 2Linked to original sources
PubMed2026

AI echo INSIGHT study: A prospective blinded randomized trial of artificial intelligence echocardiogram interpretation.

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.

Humans↗
PubMed1988

Integration of depth modules: stereo and shading.

We studied the integration of image disparities, edge information, and shading in the three-dimensional perception of complex yet well-controlled images generated with a computer-graphics system. The images showed end-on views of flat- and smooth-shaded ellipsoids, i.e., images with and without intensity discontinuities (edges). A map of perceived depth was measured by adjusting a small stereo depth probe interactively to the perceived surface. Our data show that disparate shading (even in the absence of disparate edges) yields a vivid stereoscopic depth perception. The perceived depth is significantly reduced if the disparities are completely removed (shape-from-shading). If edge information is available, it overrides both shape-from-shading and disparate shading. Degradations of depth perception corresponded to a reduced depth rather than to an increased scatter in the depth measurement. The results are compared with computer-vision algorithms for both single cues and their integration for three-dimensional vision.

Computer Graphics↗
PubMed1988

Computational studies of the extraction of visual spatial information from binocular and motion cues.

This paper reviews some of the contributions that work in computational vision has made to the study of biological vision systems. We concentrate on two areas where there has been strong interaction between computational and experimental studies: the use of binocular stereo to recover the distances to surfaces in space, and the recovery of the three-dimensional shape of objects from relative motion in the image. With regard to stereo, we consider models proposed for solving the stereo correspondence problem, focussing on the way in which physical properties of the world constrain possible methods of solution. We also show how critical observations regarding human stereo vision have helped to shape these models. With regard to the recovery of structure from motion, we focus on how the constraint of object rigidity has been used in computational models of this process.

Humans↗
PubMed1992

Artificial vision approach to the understanding of heart motion.

To overcome the major drawbacks of conventional descriptive methods, we have developed a computer vision approach to aid understanding of heart motion from a series of sequential X-ray images. The computation is addressed of local descriptors of the heart pumping function from ventricular contours. Physical constraints are exploited such as spatial smoothness of the displacement field and shape correspondence between ventricular boundaries during the beat. A computational method is proposed for the estimation of the displacement field of the left ventricular boundary. Moreover, the spatial arrangement of the estimated motion field is rendered explicit so that it may be utilized in the medical clinic (or for high-level symbolic processing). This is achieved using a grouping criterion which allows the clustering of contiguous points of the left ventricular outline into curve segments which have homogeneous motion properties.

Algorithms↗
PubMed1991

Reliability of cephalometric analysis using manual and interactive computer methods.

This study compares the results of cephalometric analyses using manual and interactive computer graphics methods. Results are statistically in favour of the interactive computer system. This study provides a basis for ongoing research into alternative methods of cephalometric analyses, such as digitization and automatic landmark identification using sophisticated computer vision systems.

Cephalometry↗
PubMed1991

[Assessment of vision using a computer].

1. The authors elaborated and tested a computer method for assessment of vision. This method is compatible with the basic procedure for assessment of vision and its variability. 2. The authors recommend to supplement values of vision by the parameter of line steepness. This parameter is independent on the value of vision and gives the number of normalized lines on which the number of correctly assessed optotypes declines from 95 to 5%.

Adult↗
PubMed1990

Architecture and design of a computerised stereogram generator for vision test.

A new microcomputer-based stereogram generator was designed and implemented to generate various visual stimuli that are used for testing the binocular vision system. The system is capable of generating static and dynamic stereoscopic stereograms that can be varied in size, shape, speed and disparity. It can also be used to generate a luminous stimulus on a dark background which, except for the depth parameters, can be varied in a similar way to the stereoscopic stimulus. A 16/32-bit microprocessor has been employed for the overall control of the stereogram parameters, which provides flexibility, versatility, compactness and speed at reduced cost. We have applied this system to the measurement of eye movement and computer vision.

Diagnosis, Computer-Assisted↗
PubMed2025

The role of chromatin state in intron retention: A case study in leveraging large scale deep learning models.

Complex deep learning models trained on very large datasets have become key enabling tools for current research in natural language processing and computer vision. By providing pre-trained models that can be fine-tuned for specific applications, they enable researchers to create accurate models with minimal effort and computational resources. Large scale genomics deep learning models come in two flavors: the first are large language models of DNA sequences trained in a self-supervised fashion, similar to the corresponding natural language models; the second are supervised learning models that leverage large scale genomics datasets from ENCODE and other sources. We argue that these models are the equivalent of foundation models in natural language processing in their utility, as they encode within them chromatin state in its different aspects, providing useful representations that allow quick deployment of accurate models of gene regulation. We demonstrate this premise by leveraging the recently created Sei model to develop simple, interpretable models of intron retention, and demonstrate their advantage over models based on the DNA language model DNABERT-2. Our work also demonstrates the impact of chromatin state on the regulation of intron retention. Using representations learned by Sei, our model is able to discover the involvement of transcription factors and chromatin marks in regulating intron retention, providing better accuracy than a recently published custom model developed for this purpose.

Deep Learning↗
PubMed1988

Deficits in stereoscopic depth perception by mildly mentally retarded adults.

The ability of mildly mentally retarded adults to perceive specific perceptual phenomena attendant to global stereopsis produced by random element stereograms was investigated. From the standpoint of computational vision, these phenomena are difficult to process, yet nonretarded persons perceive them effortlessly and without error. Retarded subjects in this study, however, exhibited large qualitative deficits not attributable to an absence of stereopsis or a failure to comprehend. These results suggest that the computational requirements of the stimuli exceeded resources and imply the presence of a substantial structural deficit in an automatic preattentive perceptual stage quite distant from the domain of cognition.

Adult↗
PubMed1979

Cerebral color blindness: an acquired defect in hue discrimination.

In contrast to the traditional view that striate visual cortex (area 17) is surrounded by two homogeneous cortical areas (areas 18 and 19), recent studies have shown that mammalian extrastriate visual cortex contains several anatomically and functionally distinct subregions. One such region, the V-4 complex of the rhesus monkey, is highly specialized for the analysis of color information, suggesting that a lesion in a homologous region might produce a defect in color vision while sparing other visual functions. We have studied a patient whose clinical syndrome supports this suggestion: a 44-year-old man with normal color vision suffered two cerebral infarctions that produced first a right and then a left superior homonymous quadrantanopia and also caused prosopagnosia, topographical disorientation, and severely impaired color vision. Computed tomography demonstrated extensive lesions in both inferior occipital lobes in the territories of the lateral branches of the posterior cerebral arteries, involving the lingual and medial occipitotemporal gyri bilaterally; these gyri contain the inferior portion of striate cortex and segments of extrastriate visual cortex. The patient had no difficulty in giving the correct color names associated with common objects presented either verbally or in outline drawings. Standardized testing with the Farnsworth-Munsell 100-hue test, the Nagel anomaloscope, and a method that tests for just-noticeable differences between monochromatic stimuli all showed that the patient's ability to distinguish one color from another was markedly imparied but not totally absent. In contrast, visual acuity, reading, visually guided eye movements, and stereopsis were normal. Cells in the V-4 complex of monkey extrastriate cortex are highly specialized for distinguishing one color from another; the hue discrimination deficit that was demonstrated in this patient with cerebral color blindness indicates that a region or regions with similar function has been damaged.

Adult↗
PubMed2026

The Role of Artificial Intelligence Combined With Digital Cholangioscopy for Indeterminant and Malignant Biliary Strictures: A Systematic Review and Meta-analysis.

BACKGROUND: Current endoscopic retrograde cholangiopancreatography (ERCP) and cholangioscopic-based diagnostic sampling for indeterminant biliary strictures remain suboptimal. Artificial intelligence (AI)-based algorithms by means of computer vision in machine learning have been applied to cholangioscopy in an effort to improve diagnostic yield. The aim of this study was to perform a systematic review and meta-analysis to evaluate the diagnostic performance of AI-based diagnostic performance of AI-associated cholangioscopic diagnosis of indeterminant or malignant biliary strictures. METHODS: Individualized searches were developed in accordance with PRISMA and MOOSE guidelines, and meta-analysis according to Cochrane Diagnostic Test Accuracy working group methodology. A bivariate model was used to compute pooled sensitivity and specificity, likelihood ratio, diagnostic odds ratio, and summary receiver operating characteristics curve (SROC). RESULTS: Five studies (n=675 lesions; a total of 2,685,674 cholangioscopic images) were included. All but one study analyzed a deep learning AI-based system using a convoluted neural network (CNN) with an average image processing speed of 30 to 60 frames per second. The pooled sensitivity and specificity were 95% (95% CI: 85-98) and 88% (95% CI: 76-94), with a diagnostic accuracy (SROC) of 97% (95% CI: 95-98). Sensitivity analysis of CNN studies (4 studies, 538 patients) demonstrated a pooled sensitivity, specificity, and accuracy (SROC) of 95% (95% CI: 82-99), 88% (95% CI: 72-95), and 97% (95% CI: 95-98), respectively. CONCLUSIONS: Artificial intelligence-based machine learning of cholangioscopy images appears to be a promising modality for the diagnosis of indeterminant and malignant biliary strictures.

Humans↗
PubMedDate not supplied

Image feature analysis and computer-aided diagnosis in digital radiography. 3. Automated detection of nodules in peripheral lung fields.

We are investigating the characteristic features of lung nodules and the surrounding normal anatomic background in order to develop an algorithm of computer vision for use as an aid in the detection of nodules in digital chest radiographs. Our technique involves an attempt to eliminate the background anatomic structures in the lung fields by means of a difference image approach. Then, feature-extraction techniques, such as tests for circularity, size, and their variation with threshold level, are applied so that suspected nodules can be isolated. Preliminary results of this automated detection scheme yielded high true-positive rates and low false-positive rates in the peripheral lung regions of the chest. This detection scheme, which can assist the final diagnosis by the clinician, has the potential to improve the early detection of lung carcinomas.

Diagnostic Errors↗
PubMed1987

Computer automated cell size and shape analysis in cryomicroscopy.

Computer vision techniques have been developed for quantitative analysis of size and shape changes in cells frozen on a cryomicroscope. The analysis is based on implementation of standard serial edge detection algorithms in conjunction with a shape transform to isolate individual cells in complex scenes which may include adjacent and overlying ice crystals. In the present study the sensitivity of the automated analysis procedure is evaluated for images obtained by various microscope optical systems for progressive degrees of subject blurring by defocusing. Size measurements in calibration trials for freezing latex spheres with extracellular ice in the field of view were least sensitive for bright field images, although the most consistent data was obtained by differential interference contrast microscopy. In all cases phase contrast images produced the least accurate data. An example analysis is presented for the freezing of pancreas beta-cells.

Algorithms↗
PubMedDate not supplied

Improvement in specificity of ultrasonography for diagnosis of breast tumors by means of artificial intelligence.

A set of ultrasonograms of lesions from 200 patients between the ages of 14 and 93 years who underwent mammography followed by ultrasonographic examination and excisional biopsy has been studied with computer vision techniques to improve the ultrasonographic specificity of the diagnosis. Selected features representing the texture of the lesion were calculated and then classified by an artificial neural network. This network was biased toward correctly classifying all the malignant cases at the expense of some misclassification of the benign cases. The network diagnosed the malignant cases with 100% sensitivity and 40% specificity (compared with 0% specificity for the radiologists diagnosing the same set of cases in the breast imaging setting), and tests performed with a leave-one-out technique indicate that the network will generalize well to new cases. This suggests that methods based on neural network classification of texture features show promise for potentially decreasing the number of unnecessary biopsies by a significant amount in patients with sonographically identifiable lesions.

Adolescent↗
PubMed1989

Epithelium-lined cyst of the pretectal region: case report and electron microscope study.

Ultrastructural findings of an epithelium-lined cyst in the left pretectal region are reported. A 38-year-old woman developed a sensory disturbance on her right side and blurring of vision. Computed tomography scans and magnetic resonance images disclosed a round cystic lesion in the left pretectum. Light microscopically, the cyst was found to be lined by a single layer of cuboidal epithelial cells. Electron microscope examination revealed that the epithelial cells of the cyst possessed clear nuclei, abundant tonofilaments, and glycogen granules, featured as well as the usual organellae. The free surface of the epithelial cells had numerous finger-like microvilli with coating materials. These cells were interconnected by well-developed desmosomes and interdigitations, and also possessed basal lamina materials on the basal surface. These cytological features suggest a heterogenous origin of the cells rather than a neuroepithelial one.

Adult↗
PubMed1992

Cranial fasciitis of childhood: a case report.

Cranial fasciitis of childhood is very rare, only 17 cases having been reported in the literature. We report an additional case of this rare disease. The patient was a 5-year-old boy who complained of left exophthalmos and double vision. Computed tomography (CT) and magnetic resonance imaging (MRI) revealed a large epidural mass in the left frontal region that had invaded into the underlying anterior skull base. The tumor showed homogeneous, low density with nonhomogeneous contrast enhancement on the CT scans, and low intensity on the T1-weighted and high intensity on the T2-weighted MRI images. A whitish-pink, elastic, hard tumor was revealed in the epidural space in the left anterior cranial fossa, which was totally excised with curettage of the affected anterior skull base. The origin of the tumor was suspected to be the fibrous connective tissue of the sphenofrontal suture. The histological diagnosis was that of cranial fasciitis. There was no evidence of recurrence 1 year postoperatively.

Child, Preschool↗
PubMed2026

Automated CEAP Classification of Venous Duplex Reports Using Multimodal Artificial Intelligence.

OBJECTIVE: To develop and internally validate a prototype multimodal artificial intelligence system for automated CEAP (Clinical, Etiological, Anatomical and Pathophysiological) classification of venous duplex ultrasound (VDUS) reports, integrating natural language processing of free-text components with computer vision analysis of hand-drawn anatomical diagrams. METHODS: Single centre retrospective observational study using routinely collected clinical data. One thousand consecutive venous duplex ultrasound reports from Cambridge University Hospitals NHS Foundation Trust, UK (July 2024 - May 2025) were labelled according to the CEAP classification, excluding the Etiological component, which could not be reliably determined from duplex reports alone. Transfer learning was applied using ClinicalBERT for text and MobileNetV3 for diagrammatic data. Clinical classes were predicted from request line text. Text- and image-based pathophysiological models were developed for four anatomical territories (Great Saphenous Vein, Small Saphenous Vein, Deep system, Perforators), combined using late fusion with probability averaging. RESULTS: The clinical CEAP model achieved accuracy of 0.91, macro-F1 of 0.82, and macro-AUC of 0.98. Pathophysiological prediction varied, with text models broadly outperforming image models. Fusion yielded heterogeneous benefits, improving SSV performance but reducing Deep system accuracy. The performance of the final pathophysiological CEAP fusion models varied across anatomical territories: accuracy ranged from 0.70-0.92 and macro-AUC from 0.80-0.92. CONCLUSION: This study demonstrates the feasibility of automated CEAP classification from VDUS reports. Despite class imbalance affecting minority class predictions, the strong discriminatory performance validates this multimodal ML model for extracting clinically meaningful information from real-world data. This approach offers potential, pending external validation, to streamline vascular services through automated triage and guideline-compliant decision making.

Artificial intelligence↗