PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Image Interpretation, Computer-Assisted”

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 19 recordsLinked to original sources
PubMed2026

A comparative evaluation of multiple enlarged perivascular space segmentation tools.

BACKGROUND: Enlarged perivascular spaces (ePVS) are a marker of cerebral small vessel disease, potentially reflecting reduced waste clearance. Because manual quantification is unfeasible in large datasets, we developed and evaluated an automated tool. METHODS: Detection Of Regions of Enlarged perivascular Spaces (DORES), a 3D nnU-Net-based deep learning algorithm was developed for ePVS segmentation using T1-weighted and fluid-attenuated inversion recovery magnetic resonance imaging (MRI). DORES was developed in two stages: an initial model trained on 35 manually segmented scans and a final model on 1460 pseudo-labeled sessions from the Vanderbilt Memory and Aging Project (VMAP). A subset of VMAP participants with 3 T brain MRI underwent whole-brain manual ePVS tracing (n = 35, 73 ± 9 years, 51% male) and visual rating (n = 388, 71 ± 8 years, 54% male) by a neuroradiologist. DORES was evaluated and compared against three other segmentation tools using Dice and F1 scores, absolute volume and element differences, correlation, and agreement. External validation used an Alzheimer's Disease Neuroimaging Initiative 3 subset with manual tracings (ADNI3, n = 18, 73 ± 9 years, 67% female). RESULTS: DORES achieved Dice scores of 0.61 ± 0.16 (white matter) and 0.72 ± 0.08 (basal ganglia) in VMAP, with strong correlations and agreement for ePVS count and volume. Performances modestly declined in ADNI3 across algorithms. Scanner-stratified analyses showed stronger correlations for Philips versus Siemens images in the basal ganglia, indicating scanner-dependent differences in measurement consistency. CONCLUSIONS: DORES provides a multimodal nnU-Net-based pipeline for ePVS segmentation in older adults. The model demonstrates robust within-cohort performance and reasonable external validity, though scanner-related effects limit application across sites.

Humans↗
PubMedDate not supplied

The diagnosis of dental caries: 3. Rationale and overview of present and possible future techniques.

A number of changes have taken place recently which may influence a practitioners choice of diagnostic methods for detecting and monitoring dental caries. The first two articles described the use of current diagnostic methods at various specific sites. This third paper discusses the rationale behind the use of these techniques and provides an overview of present techniques and those which may be useful in the future.

Dental Caries↗
PubMed2026

Artificial Intelligence and Machine Learning Applications in Fibromuscular Dysplasia: Transforming Diagnosis, Risk Stratification, and Clinical Decision-Making.

Fibromuscular dysplasia (FMD) is a non-atherosclerotic vascular disorder with heterogeneous presentations, making diagnosis and management highly dependent on imaging and clinical expertise. This narrative review examines how artificial intelligence (AI) and machine learning (ML) are transforming FMD care. AI-enhanced imaging, particularly convolutional neural network-based analysis, improves detection of the characteristic "string-of-beads" pattern on CT angiography, magnetic resonance angiography, and ultrasound, although FMD-specific validation remains limited. ML models facilitate risk stratification, prediction of disease progression, and early identification of complications such as aneurysms and stroke by integrating clinical, imaging, and genomic data. AI-driven clinical decision support systems further enable personalized treatment selection through pharmacogenomic insights and robot-assisted interventions. Despite promising real-world applications, challenges persist, including limited large-scale datasets, workflow integration, regulatory barriers, and algorithmic bias affecting underrepresented populations. Future advances in explainable AI, federated learning, and digital health integration may enable a shift toward predictive, patient-centered FMD management.

Humans↗
PubMed2026

Artificial intelligence-assisted histopathological diagnosis of endocervical gastric-type adenocarcinoma: a multicenter model development and validation study.

Endocervical gastric-type adenocarcinoma (GAS) is one of the most aggressive subtypes of cervical cancer and is frequently underdiagnosed due to morphological ambiguity, leading to delayed diagnosis. Despite the availability of molecular and genomic assays, their high cost, complexity, and limited reproducibility restrict clinical use. This study therefore proposes a highly sensitive artificial intelligence (AI)-assisted diagnostic system for GAS based exclusively on H&E-stained histopathological images. We included 309 slides from 96 GAS cases collected at Peking University Third Hospital from January 2018 to January 2025, representing the largest GAS cohort reported to date for AI research. In addition, we incorporated other morphologically analogous diseases, encompassing a total of 1,320 slides sourced from four categories: normal cervical mucosa (NORM), benign endocervical lesion entities (BELE), HPV-associated adenocarcinoma (HPVA), and endometrioid carcinoma with mucinous differentiation (ECMD). We developed GASPath, based on a novel multiple instance learning framework that efficiently captures fine-grained morphological variations from H&E-stained images. Beyond internal validation, GASPath was evaluated across 12 independent retrospective cohorts and further subjected to large-scale real-world validation on more than 7,000 samples from March 2024 to April 2025. Across three stages, GASPath demonstrated high performance. In internal validation (Stage I), it achieved an accuracy of 0.980 (95% CI 0.977-0.983) and an ROC-AUC of 0.995 (95% CI 0.994-0.997). In external validation (Stage II), the sensitivity reached 0.902 and improved to 0.968 with proposed strategies. For biopsy samples, GASPath achieved an ROC-AUC of 0.990 (95% CI 0.984-0.997). In large-scale real-world deployment (Stage III, n = 7,056), GASPath achieved a balanced accuracy of 0.953, with 100% sensitivity for GAS (45/45 cases correctly identified). The heatmaps highlight morphological features of GAS that are easily underestimated, such as irregular, angulated glands, subtle loss of nuclear polarity, and mild cytologic atypia, which show substantial morphological overlap with other diagnostic categories. GASPath enables high-sensitivity detection of GAS in routine H&E-stained slides, obviating the need for extensive auxiliary testing while preventing underdiagnosis and misdiagnosis. This advancement addresses a critical gap by streamlining diagnostic workflows without compromising accuracy. Its implementation could enable cost-effective, scalable AI-assisted diagnostics, potentially transforming the early detection and management of this aggressive cancer subtype.

Female↗
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↗
PubMed2025

Cardiac CT fractal analysis of LV noncompaction and common cardiomyopathies.

BACKGROUND: Left ventricular noncompaction (LVNC), or hypertrabeculation, is a myocardial condition that remains challenging to diagnose and differentiate from other cardiomyopathies. This study evaluated the ability of cardiac CT to differentiate between LVNC, hypertrophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM), and controls using fractal analysis of LV trabeculae. METHODS: Subjects with LVNC, HCM, DCM, as well as controls, who underwent coronary CT angiography were included. LV trabecular structure was quantified using fractal analysis on a stack of 15 short-axis CT images. For each subject, maximum (FDmax) and average (FDglobal) fractal dimensions were reported. A subset of subjects also had clinically acquired cardiac MRI (CMR) exams for comparison. One-way ANOVA, Pearson correlation, and Bland-Altman analysis were used for statistical analysis. RESULTS: The study included 313 subjects (median age: 58.8 [48.1-68.0] years, 153 male) categorized into Control (89), LVNC (46), HCM (106), and DCM (72) cohorts. FDmax was significantly higher in LVNC (1.379 &#x200b;&#xb1; &#x200b;0.047) than in Control (1.305 &#x200b;&#xb1; &#x200b;0.033), HCM (1.321 &#x200b;&#xb1; &#x200b;0.040), and DCM (1.344 &#x200b;&#xb1; &#x200b;0.054) cohorts; all p &#x200b;< &#x200b;0.001. Similarly, FDglobal was significantly higher in LVNC (1.279 &#x200b;&#xb1; &#x200b;0.041) than in the other cohorts; all p &#x200b;< &#x200b;0.05. In a subset of 132 subjects with both CT and CMR exams, fractal dimensions from the two modalities were strongly correlated (r &#x200b;= &#x200b;0.63, p &#x200b;< &#x200b;0.0001), with CT-derived values being higher (1.337 &#x200b;&#xb1; &#x200b;0.049 vs. 1.262 &#x200b;&#xb1; &#x200b;0.045, p &#x200b;< &#x200b;0.0001). CONCLUSIONS: CT-derived fractal dimensions of LV trabecular structure were significantly higher in LVNC compared to control subjects, HCM, and DCM. CT-derived fractal dimensions strongly correlated with, but were higher than, those from cardiac MRI in the same subjects.

Humans↗
PubMed1992

Development and distribution of the white blood cells within various structures of the human menstrual corpus luteum examined using an image analysis system.

PROBLEM: Emerging evidences suggest that immunoendocrine interactions play definitive roles during development and regression of the human menstrual corpus luteum (hmCL). We have studied the distribution of immune cells within individual structures of hmCL during various stages of its development. METHOD: Immunoperoxidase-stained ultra-thin frozen sections were evaluated using light microscopy fitted with an image analysis system. RESULTS: The results suggest that monocytes/macrophages and MHC class II positive cells are the most prominent immune cells within the hmCL throughout its whole lifespan. Both cell types are concentrated within the trabeculae. In addition, MHC class II positive cells are abundant also within the granulosa-luteal layer. T helper/inductor (Th/i) and T cytotoxic/suppressor (Tc/s) cells were detected only in minor amounts within the thecal trabeculae of mature tissue. CONCLUSIONS: Possible links between the occurrence and functional roles of the immune cells studied are discussed.

Antigens, CD↗
PubMed1992

[Technique for measuring the atheroma volume in men].

The evaluation of the impact of therapy on the evolution of atherosclerotic lesions or restenosis after angioplasty requires the use of techniques of vascular imaging. The reference invasive method is digital angiography although it does not provide data on the arterial wall thickness. This parameter can be approached however by intravascular ultrasound imaging, a technique which has a number of important practical limitations. Of the non-invasive techniques available, Doppler ultrasonography is the only one that can be used in clinical trials. Nuclear magnetic resonance imaging is the object of much research and is without doubt the technique of the future. The choice of model of atherosclerosis influences that of the imaging technique: cineangiography for coronary arteries, digital angiography or Doppler ultra sonography for lower limb arteries and Doppler ultrasonography for the carotid arteries. Interpretation of angiography is now performed quantitatively by videodensitometry. Interpretation of other techniques should be performed by a second independent observer and "blinded" with respect to the order in which the investigations were performed and to the treatment administered. The criteria of judgment may be qualitative (progression, stabilisation, regression) or quantitative, the latter having a number of advantages over the former. Of the quantitative criteria, the percentage stenosis, though widely used, does not fully answer the question posed, and neither does the diameter of the stenosis. The volume of the arterial lumen calculated from videodensitometric data would seem to be the best, by its sensitivity and additivity, current angiographic parameter.(ABSTRACT TRUNCATED AT 250 WORDS)

Angiography, Digital Subtraction↗
PubMed2026

Integrating histology and spatial transcriptomics via multimodal transformers and contrastive representation learning for accurate gene expression prediction.

Predicting spatial gene expression from Histological images is a fundamental task in understanding tissue organization and molecular phenotypes. However, existing methods often rely on single-model representations or lack effective alignment between image and transcriptomic features. To address these limitations, we propose a unified multimodal learning framework that integrates histological imaging and spatial transcriptomics through a shared latent representation space. Specifically, histological H&E images are encoded by a ResNet50-based convolutional stem and a MobileViT Transformer backbone to extract hierarchical visual representations. Both modalities are projected into a shared latent space via linear-GELU-dropout transformation blocks, enabling cross-modal alignment through a contrastive learning objective that maximizes agreement between the corresponding image and the spot embeddings. Experimental results on the 10x Genomics Visium dataset of human liver tissue demonstrate that MViTGene achieves significantly higher prediction accuracy than existing methods across multiple gene subsets, with improvements of 20%, 33%, and 12% in predicting marker genes, highly expressed genes, and highly variable genes, respectively. The significant improvement in relevance indicates that the model can more accurately capture the true correspondence between tissue morphology and gene expression, therefore enabling more reliable biological interpretation. It provides a computational tool for high-throughput spatial gene expression prediction that balances performance and interpretability.

Humans↗
PubMed1976

[Theory of image discernment and resolution of classification problems using medico-psychological methods].

The report deals with questions pertaining to the use of the theory of image discernment for solving psychodiagnostical problems in clinical practice. The traditional understanding of a psychological test as a measurement instrument is being opposed to its new interpretation as an instrument of classification. On this basis the authors described the successional stages of the general procedure in the elaboration of tests. Special attention is being drawn to the selection of primary signs with the aid of informational measures and the formation of scales as discriminators. The suggested approach is being illustrated on the model of elaborating scales of a personality questionnaire and intellectual techniques.

Diagnosis, Computer-Assisted↗
PubMed2020

Chemical Imaging of Retinal Pigment Epithelium in Frozen Sections of Zebrafish Larvae Using ToF-SIMS.

Variants of the SLC24A5 gene, which encodes a putative potassium-dependent sodium-calcium exchanger (NCKX5) that most likely resides in the melanosome or its precursor, affect pigmentation in both humans and zebrafish (Danio rerio). This finding suggests that genetic variations influencing human skin pigmentation alter melanosome biogenesis via ionic changes. Gaining an understanding of how changes in the ionic environment of organelles impact melanosome morphogenesis and pigmentation will require a spatially resolved way to characterize the chemical environment of melanosomes in pigmented tissue such as retinal pigment epithelium (RPE). The imaging mass spectrometry technique most suited for this type of cell and tissue analysis is time-of-flight secondary ion mass spectrometry (ToF-SIMS) because it is able to detect many biochemical species with high sensitivity and with submicron spatial resolution. Here, we describe chemical imaging of the RPE in frozen-hydrated sections of larval zebrafish using cryo-ToF-SIMS. To facilitate the data interpretation, positive and negative polarity ToF-SIMS image data were transformed into a single hyperspectral data set and analyzed using principal component analysis. The combination of a novel protocol and the use of multivariate data analysis allowed us to discover new marker ions that are attributable to leucodopachrome, a metabolite specific to the biosynthesis of eumelanin. The described methodology may be adapted for the investigation of other classes of molecules in frozen tissues from zebrafish and other organisms.

Animals↗
PubMed2024

Digital pathology and spatial omics in steatohepatitis: Clinical applications and discovery potentials.

Steatohepatitis with diverse etiologies is the most common histological manifestation in patients with liver disease. However, there are currently no specific histopathological features pathognomonic for metabolic dysfunction-associated steatotic liver disease, alcohol-associated liver disease, or metabolic dysfunction-associated steatotic liver disease with increased alcohol intake. Digitizing traditional pathology slides has created an emerging field of digital pathology, allowing for easier access, storage, sharing, and analysis of whole-slide images. Artificial intelligence (AI) algorithms have been developed for whole-slide images to enhance the accuracy and speed of the histological interpretation of steatohepatitis and are currently employed in biomarker development. Spatial biology is a novel field that enables investigators to map gene and protein expression within a specific region of interest on liver histological sections, examine disease heterogeneity within tissues, and understand the relationship between molecular changes and distinct tissue morphology. Here, we review the utility of digital pathology (using linear and nonlinear microscopy) augmented with AI analysis to improve the accuracy of histological interpretation. We will also discuss the spatial omics landscape with special emphasis on the strengths and limitations of established spatial transcriptomics and proteomics technologies and their application in steatohepatitis. We then highlight the power of multimodal integration of digital pathology augmented by machine learning (ML)algorithms with spatial biology. The review concludes with a discussion of the current gaps in knowledge, the limitations and premises of these tools and technologies, and the areas of future research.

Humans↗
PubMed1978

Dual radionuclide study of acute myocardial infarction: comparison of thallium-201 and technetium-99m stannous prophosphate imaging in man.

We evaluated dual imaging with thalium-201 (201TI) and technetium-99m (99mTc) pyrophosphate in 80 patients with documented acute myocardial infarction (55 transmural, 25 nontransmural infarction). Color-coded isocount display of 201TI images was essential for interpretation in 16 patients. Combined 201 TI and 99mTc-pyrophosphate imaging for infarct detection was 100% sensitive; however, either was falsely negative in 12 of 80 patients. False-negative individual 201TI or 99mTc-pyrophosphate infarct images were most common in patients with small infacts or left ventricular hypertrophy. Thallium-201 images correctly localized the site of acute transmural infarction in all 51 patients with a positive image, while 99mTc-pyrophosphate localized the site of infarction in 49 of 53 with an abnormal image. Comparison of the size of the imaged infarct region revealed size discordance in 25 of 49 patinets, with 99mTc-pyrophosphate larger in 21 of 49 and 201TI larger in only four of 49. Thus dual radionuclide imaging provides definition of the presence and location of acute myocardial infarction.

Acute Disease↗
PubMed1976

Methods for measurement of cerebral blood flow in man.

A survey of the currently available methods for the measurement of cerebral blood flow in man is given. Many of the clinically important brain diseases such as tumors, stroke, brain trauma or epilepsy entail focal or regional flow alterations. Therefore a special emphasis is placed on methods allowing measurements of regional cerebral flow, rCBF. The intra-arterial 133Xenon injection method is now widely used as a standard method for rCBF measurement. It affords a good two-dimensional resolution when using a suitable dynamic gamma camera which allows a high counting rate to be recorded. But, due to the superposition of tissues the three-dimensional resolution is limited. This, in particular, means that smaller areas of ischemia (low flow) tend to be overlooked whereas local hyperemia is readily discerned. The 133Xenon inhalation method is less accurate, contaminated by extra-cerebral uptake, and insensitive both for detecting regional ischemia and regional hyperemia. The spatial resolution is also much more limited. For these reasons great caution must be exercised in interpreting the results. Methods yielding three-dimensional rCBF data will be needed in order to gain more precise information both on spatial localization and, especially, on ischemic areas. The most promising is computer-assisted axial tomography with freely diffusible radioactive isotopes or with x-rays using an intra-arterial injection of contrast. But, the available techniques are still too slow: in order to measure blood flow one "exposure" must be taken every second. Only a few methods give quantitative information of the blood flow in the human brain. This is mainly due to the inaccessibility of the brain within the skull and to the complexity of the cerebral arterial and venous systems. Before reviewing the various methods used in man, it should be mentioned, that much of the fundamental knowledge has been gained by methods only applicable to animals. Measurements of the diameter of the small arteries on the surface of the brain antedates even the classical studies of Roy and Sherrington (1890). This technique continues to be useful, modern technical improvements consisting of the use of micropipettes and a stereo microscope in combination with an image splitter and a television camera which allows the accurate assessment of diameter variations of a few percent [22]. Autoradiography of brain slices using diffusible indicators is the best quantitative method for measuring local blood flow in a great many parts of the brain [7, 45]. Microspheres are also being used, but it is still not quite clear that this technique gives reliable quantitative data in small masses of tissue [34, 41, 50].

Animals↗
PubMed2026

Analysis of deep learning techniques in computer-aided diagnosis for meniscus injuries: a systematic literature review.

Meniscus informatics is a growing subject of study in the healthcare industry. One of the major hindrances to the healthcare system's transformation is obtaining knowledge and meaningful information from complicated, high-dimensional and diverse sources. Modern biomedical research, for instance, has seen an increase in the use of complex, dissimilar, poorly documented, and generally unstructured electronic health records, imaging, sensor data and text, even after many current techniques have been used to extract more robust and useful elements from the data for analysis. New efficient standards for building end-to-end learning models from complex data are therefore needed. Therefore, the current study aims to examine the most recent research on the use of deep learning techniques for diagnosing meniscus tears and recommend creating comprehensive and meaningful interpretable structures that might benefit the healthcare industry. We also draw attention to shortcomings and the need for better technique development, and we provide new perspectives about this exciting new development in the field.

Humans↗
PubMedDate not supplied

X-ray solution scattering studies on vinblastine-induced polymers of microtubule protein: structural characterisation and effects of temperature.

We report here on X-ray solution scattering and electron microscopy studies of microtubule protein in the presence of the antimitotic drug, vinblastine. In buffer conditions used for microtubule assembly, vinblastine caused the formation of coil-like structures. The coils appeared to be made up of two protofilaments. Details of the structure and behaviour of coils in solution were obtained from interpretation of their solution scattering patterns. Upon increasing temperature from 4 to 37 degrees C the pitch of the coils increased from 25.92 to 26.96 nm. However, little change was observed in their mean diameters (38.46 and 38.45 nm, respectively). Increasing the temperature also favoured increased formation and/or elongation of the coils. The effect of temperature on the pitch was fully reversible. Vinblastine-induced assembly of pure tubulin also showed the formation of coils. However, these coils appeared to consist of only one protofilament. Their mean diameters (38.35 nm) were similar to those of the coils formed from microtubule protein.

Animals↗
PubMedDate not supplied

The Staphylococcus aureus alpha-toxin channel complex and the effect of Ca2+ ions on its interaction with lipid layers.

Using the techniques of two-dimensional crystallization on supported lipid bilayers together with computer image processing, two distinct two-dimensional crystal types of staphylococcal alpha-toxin complex are formed depending on the presence or absence of Ca2+ ions. Without Ca2+, these are hexagonally packed (in A, a = b = 89.5 +/- 2.5 A; theta = 119.7 degrees) With Ca2+ present, rectangular crystal packing is seen (in A, a = 114.8 +/- 1.6 A, b = 140.2 +/- 0.7 A; theta = 89.1 degrees). A third, banded crystal type is also seen which is interpreted as a side-to-side packing of regular tubules. We use these tubular crystals for cross-correlation searches with top and side-on views of the complex from single particle reconstructions, and with the repeating units from the two-dimensional crystal types. The results lead us to propose a model in which the different two-dimensional crystal types are formed as a result of alpha-toxin hexamers packing in different orientations. In the hexagonal crystals the hexamers lie end-on with a 6-fold axis in projection. On the addition of Ca2+, the hexamers reorient to lie tilted with respect to the support, thus giving rise to a rectangular projection.

Bacterial Toxins↗