Articles tagged with Image Analysis
Optical convolution computation enables parallel light propagation and multiplexing for faster and more energy-efficient computing systems. The review organizes the field into two paradigms: definition-based and theorem-based, which leverage mathematical principles to implement convolution operations in the physical domain.
Researchers develop a novel framework, LL-Refiner, to enhance high-resolution images in poor lighting conditions, outperforming state-of-the-art techniques. The framework uses a coarse enhancement stage to guide the recovery of fine details, resulting in improved visual quality and performance in downstream computer-vision tasks.
The PolyU research team has developed TRUECAM, an integrated AI framework that ensures data and model trustworthiness in cancer diagnosis. The framework can assess the level of an AI's confidence in its diagnostic outputs and proactively prompts pathologists to review cases when uncertainty is high.
Researchers developed a photospike-based TRNG that harnesses unpredictable light-induced electrical charges to generate true random numbers. The device passed all 15 randomness tests and remained stable over millions of cycles, making it suitable for image authentication and deepfake detection.
A new study revealed that retinal photographs can accurately predict many common risk factors associated with developing Alzheimer's disease. The AI model identified regions of the retina linked to Alzheimer's risk factors, such as arteries and optical nerve, and predicted lifestyle factors like smoking and alcohol use.
A recent study analyzing nearly 800,000 fashion images reveals that while representation has broadened, the typical female model body has remained remarkably stable. Diversity is primarily concentrated in a small number of models at the extremes rather than through a shift in the norm itself.
A new study by Cal Poly faculty found that building density, not urban trees, was the strongest predictor of home loss in Los Angeles firestorms. The study examined 15,082 structures and 52,893 tree canopies within the Eaton and Palisades fire scars.
A new study published in JCOM finds that images of GMOs tend to reinforce pre-existing attitudes, amplifying polarization rather than changing minds. The study used a representative sample of the US population and found that even positive images could make people who already supported GMOs more positive, while making skeptics and uncer...
Three new neural network-based tools enable precise tracking of neurons in worms and jellyfish, overcoming visibility challenges. The tools have been applied to the study of brain activity in C. hemisphaerica jellyfish, allowing researchers to extract neural activity data from videos of the animals.
A team of scientists from Tokyo Metropolitan University discovered how fertilized rice seeds begin to divide and establish their body axis. They found that the process involves radical steps different from Arabidopsis, with cells acting collectively to allow axis development despite apparent randomness.
A machine-learning-based algorithm developed by Tokyo Metropolitan University researchers can accurately count sister chromatid exchanges (SCEs) in chromosomes, giving a more objective measurement. The accuracy rate is 84%, which could help diagnose disorders like Bloom syndrome with greater consistency.
Researchers have developed a system called CytoDiffusion that uses generative AI to study the shape and structure of blood cells. The system can accurately identify normal blood cell appearances and spot unusual or rare cells that may indicate disease, outperforming existing systems in tests.
A study has identified over 240 scientific publications with potentially problematic images, casting doubt on the validity of these studies. The findings highlight the need for journals and publishers to investigate image-related issues carefully to ensure the trustworthiness of animal-based research in this field.
Researchers developed DeMemSeg, an AI-driven pipeline that accurately segments overlapping membrane structures with accuracy comparable to expert manual analysis. The approach enables large-scale, objective, and quantitative analysis of morphological data, providing a foundational technology for advancing disease mechanisms.
Researchers developed a parallel microdevice that combines high-throughput intracellular delivery with automated single-cell image cytometry using AI. The device can deliver gene-silencing RNA and plasmid DNA across multiple cell types, enabling broad utility for cell engineering and personalized therapies.
BiaPy breaks down the barrier to AI-based image analysis in biomedicine, allowing more scientists and healthcare professionals to harness its potential. The tool offers various types of analysis, including cell identification, element counting, and improving image quality, on both two-dimensional and three-dimensional images.
A new robotic system utilizes optically-induced dielectrophoresis (ODEP) for the classification and analysis of patient-derived endometrial stromal cells. The system enables precise micromanipulation and measurement of single cells, providing insights into cell properties and responses to nonuniform electric fields.
Researchers at TUM developed an AI-powered algorithm to predict kidney damage in prostate cancer patients undergoing lutetium-177 PSMA therapy. Early detection could enable personalized treatment adjustments to prevent organ damage.
Researchers at JAIST have developed a low-dose imaging technique that maps the three-dimensional atomic structure of titanium oxyhydroxide nanoparticles without damaging them. This breakthrough enables safer analysis and opens possibilities for designing materials with enhanced functionality.
Researchers used AI-driven methods to analyze thousands of digital images of melanoma tumor tissue, identifying key immune cell structures that boost immunotherapy effects. The presence of these structures was linked to significantly better overall survival for patients with advanced melanoma.
Researchers mapped a lung tumor's cellular neighborhoods in 3D using single-cell spatial technologies, identifying 18 cell types and potential targets for personalized cancer therapy. The study reveals new insights into how tumor cells interact with their surroundings and how to reverse immune suppression mechanisms.
Schmid's contributions have helped computers recognize complex objects, understand video analysis, and process realistic settings. Her leadership has built active research communities, mentoring and supervising peers across the field of computer vision.
A team of researchers developed Lp-Convolution, a novel method that uses multivariate p-generalized normal distribution to reshape CNN filters dynamically. This breakthrough improves the accuracy and efficiency of image recognition systems while reducing computational burden.
Researchers developed a new method that uses simple grayscale eye photos to predict anemia in children. The technique analyzes patterns and textures in the conjunctiva of the eye, avoiding problems caused by different light conditions or camera models.
Researchers at Tokyo Metropolitan University have created a novel technique using phase-contrast microscopy to track and analyze the motion of unlabeled cells. This allows for the accurate differentiation of cancerous cells with up to 94% accuracy, opening new avenues for diagnosis and research on cell motility related functions.
A new study by Texas A&M University researchers has revealed insights into Mars' geological history and potential for ancient life. The team analyzed diverse volcanic rocks in the Jezero Crater, providing a window into the planet's distant past and signs of altered olivine.
A new study finds that a millisecond magnetar could have triggered the flashes of GRB 230307A, an extremely bright GRB detected in March 2023. The observation suggests that the magnetar model is consistent with the features of the prompt emission and the long-lasting X-ray plateau.
MIT researchers have found that a computational model of the ventral stream, which processes object recognition, also performs well on spatial tasks such as determining an object's location and orientation. This challenges the dominant perspective that the ventral stream is optimized for object recognition.
A new microscopy technique, SIMIP, combines structured illumination with mid-infrared photothermal detection to achieve high-speed chemical imaging with superior resolution. The method outperforms conventional methods in terms of spatial resolution and chemical contrast.
A new study reveals that 81% of O’ahu's coastline could experience erosion by 2100, with a further 40% loss happening by 2030. The research used computer models incorporating satellite imagery to predict the seasonal movement of sand, resulting in more severe erosion projections than previous studies.
A new system for early detection of Autism Spectrum Disorder (ASD) has been developed using virtual reality and artificial intelligence. The system, which uses biomarkers related to behavior, motor activity, and gaze direction, achieves an accuracy of over 85%, surpassing traditional methods.
A new AI tool called MindGlide can extract key information from brain images to measure damaged areas and highlight subtle changes in MS patients. The tool performed better than other AI tools and expert clinical analysis, enabling researchers to unlock valuable insights into multiple sclerosis.
A new case report presents a highly unusual combination of two benign ovarian tumors: serous cystadenofibroma and collision lesions. Accurate preoperative diagnosis is critical for effective treatment planning, and this case emphasizes the need for personalized evaluation of each ovarian mass.
The study found that solar wind radiation plays a dominant role in space weathering on the lunar farside, differing from the nearside. The Chang'e-6 samples showed less melt drops and no nanophase metallic iron particles, indicating variations in the space environment.
Researchers found that medical imaging experts can solve common optical illusions, including judging the size of objects. Training to improve visual perception can also make experts less susceptible to these illusions. This study has implications for training medical image analysts.
A machine learning algorithm has been developed to diagnose coeliac disease with high accuracy, outperforming human pathologists in over 97 cases. The AI tool has the potential to speed up diagnosis and reduce delays in receiving an accurate diagnosis for patients suffering from this autoimmune disease.
Researchers developed a new geospatial intelligence methodology to accurately delineate areas of natural vegetation and agricultural production by crop type. The results showed 95% accuracy in mapping, providing support for public policies aimed at agricultural production and environmental conservation.
A study using NHANES data found that inflammation, rather than diet and exercise, has the strongest association with telomere shortening. Managing chronic inflammation may be key to preserving telomere length and promoting healthy aging.
Researchers developed a method that uses high-resolution satellite imagery and deep machine learning to double the mapped infrastructure of Alaska, more accurately projecting economic risks associated with permafrost thaw. The new model nearly doubles the amount of information available for Alaska on OpenStreetMap.
Full Waveform Inversion (FWI) technology provides unprecedented precision in seismic imaging, breaking resolution limitations of traditional methods. It characterizes complex structures within the Earth's interior and offers higher-resolution subsurface models.
A new AI model developed by Tokyo University of Science's researchers predicts dendritic growth in thin films, offering a powerful pathway for optimizing thin-film fabrication. The model analyzes morphology using persistent homology and machine learning with energy analysis, revealing conditions that drive branching behavior.
A team led by Dr. Marcus Botacin is creating a large language model (LLM) to automatically identify malware and write rules to defend against it. The LLM will use signatures to complement human analysts' skills, identifying malware faster and more accurately.
Researchers developed a simple, inexpensive tool using robotics and artificial intelligence to analyze dried salt solutions from images. The method increases the accuracy of chemical analysis in scenarios where large samples are difficult to obtain, making it valuable for space exploration, law enforcement, and hospital use.
A research team at Kumamoto University developed a deep learning-based method for analyzing the cytoskeleton more accurately and efficiently than ever before. This technique enabled more reliable measurements of cytoskeleton density, which is critical for understanding cellular structure and function.
A new study by the University of Plymouth investigated the effect of changing global climate conditions on Christmas Island's red crab embryos. The researchers found that lower salinity levels did not delay embryonic development, but emphasized the need for further research to understand the species' response to environmental stressors.
Researchers analyzed satellite data from the Sentinel-2 satellites to predict variations in tree traits and map functional diversity. They found significant differences in forest function across continents, with American tropical forests showing greater functional richness than African and Asian forests.
Researchers used terahertz spectroscopy to study agave plants' ability to retain water in dry environments. They found that agaves store water in a specialized leaf structure and fructans act like molecular sponges to retain moisture. This discovery could lead to better farming practices and drought-resistant crops
CHARCHA, developed by Carnegie Mellon University and MIT researchers, uses real-time physical interactions to verify user identity in generative video content. The system prevents unauthorized deepfakes and gives users greater control over their likeness.
The AI-GS network achieves high accuracy with 93.52% sensitivity at 95% specificity, excelling in detecting early-stage glaucoma. The system is portable, requiring minimal computational power and delivering results in under a second.
A new AI-based system called LILAC can accurately detect changes and predict outcomes in various medical applications. The system has been demonstrated on diverse longitudinal image series covering IVF embryos, healing tissue after wounds, and aging brains.
Researchers developed a deep-learning framework, STAIG, to automatically map distinct genetic activity to tissue regions without manual alignment. The study demonstrates superior performance across various conditions, showcasing its potential for cancer research and understanding complex biological systems.
This study utilized deep learning models to diagnose and predict the likelihood of malignant transformation in oral potentially malignant disorders. AI-driven approaches offer noninvasive, cost-effective, and objective means to enhance early detection and improve patient outcomes.
An international research team developed a user-friendly software method called Segment Anything for Microscopy, which can precisely segment images of tissues, cells, and similar structures. The new model improved performance for cell segmentation, enabling researchers to automate tasks that previously took weeks of manual effort.
A study of Viking skulls using CT scans reveals a range of diseases including sinus and ear infections, osteoarthritis, and dental diseases. The results provide greater understanding of the health and wellbeing of the Viking population.
Researchers have created a new imaging technique that uses the nanostructures found on butterfly wings to analyze cancerous tissues, providing a simpler and more accessible tool for cancer diagnosis. The method has shown comparable results to conventional staining methods and advanced imaging techniques, offering a stain-free alternative.
Researchers from Politecnico di Milano and Georgia Tech analyzed NASA's DART mission to asteroid Dimorphos, revealing a viable mechanism for ejecta evolution and understanding the impact of an asteroid's shape on deflection. The studies suggest that sending multiple smaller impactors can increase the asteroid push while reducing costs.
A new method combines traditional histopathology with spatial transcriptomics data to improve understanding of chronic kidney disease lesions at the cellular and molecular levels. This approach has the potential to identify new biomarkers and therapeutic strategies for patients.
Researchers used Synthetic Aperture Radar satellites to quantify off-fault damage and surface displacement caused by the two 2023 earthquakes. The study suggests that off-fault damage can reach up to five kilometers from the fault, contradicting previous estimates.
Researchers developed a synergetic strategy combining millimeter-wave-terahertz-infrared photo-monitoring and computer-vision three-dimensional modeling for ubiquitous non-destructive inspections. The approach allows for material composition identifications and structural reconstructions of composite multi-layered objects.
Researchers developed a five-minute quality test for sustainable cement industry materials, reducing testing time from seven days to just five minutes. The test uses colorimetry and camera technology for real-time quality control of calcined clays, which can partially replace ordinary Portland cement.