Articles tagged with Medical Imaging
Pre-training in medical data tackles challenges such as data scarcity and privacy concerns using AI techniques like transfer learning and self-supervised learning. Recent advances and new frontiers of pre-training-based techniques are introduced, including applications to medical images, bio-signal data, EHR data, and multi-modality data.
Cancer cells in brain tumors produce lipids at higher rates than surrounding healthy tissue, offering clues for treatment strategies. The study provides insights into the unique biochemical processes fueling cancer growth in the brain.
A team of researchers developed an acoustic metasurface-based holography technique that uses a deep learning algorithm to generate and iteratively improve a hologram of the Mona Lisa. The technique successfully reconstructed the painting, with even greater detail in her left eye.
Researchers at UC San Diego and Stanford University have developed a roadmap of root chemicals that are critical to plant growth, providing new insights into plant development. The study reveals that key small molecules are clustered in patches across the root, suggesting a purposeful distribution for optimal growth.
A multicenter study found that deep learning-based computer-aided diagnosis significantly improved radiologists' diagnostic performance for breast lesion classification on ultrasound. The use of CAD resulted in a 6.0% upgrade of BI-RADS category 3 assessments to category 4A, with 16.7% being malignant.
A recent study found that COVID-19 infection can lead to increased infarct growth and ongoing tissue consumption even after successful reperfusion through angiography. This suggests a potentially aggressive clinical course for patients with COVID-19 and large-vessel occlusion acute ischemic stroke.
A team of researchers from the University of Oklahoma and Yale University has developed a super-resolution imaging platform technology to visualize nanoparticles within cells. The technique, called expansion microscopy, enables 3-D imaging with resolutions as low as 10 nanometers, allowing for safer and more efficient nanomedicines.
Scientists have developed a new method to deliver genetic information to stem cells using nanoparticles coated with a specific polymer, enabling more efficient control over cellular differentiation. This innovation has the potential to improve the efficiency and effectiveness of regenerative medicine treatments.
Researchers at Caltech have developed a technique that uses quantum entanglement to create biphotons, which can be used to image cells with a resolution twice that of traditional microscopes. By harnessing the properties of quantum entanglement, scientists can now visualize tiny structures within living cells with unprecedented precision.
A diffractive optical network-based multispectral imager offers virtual spectral filter arrays, preserving spatial information and yielding an image cube without reconstruction algorithms. The system achieves ~79% average transmission efficiency across distinct bands.
The City University of Hong Kong has developed a novel electron microscope that combines scanning and transmission electron microscope modes in a compact format. The device can produce high-resolution images in five minutes, enabling the study of atom dynamics and beam-sensitive materials.
Two natural coumarin compounds, 5-MOS and 6-MOS, have been identified in the orange climber plant, exhibiting aggregation-induced emission properties. These compounds can clearly identify mitochondria in live cells without additional processing, making cell imaging easier and faster.
The American Roentgen Ray Society presents its 2023 Honorary Lecture on Advanced High-Resolution CT, dedicated to the late Dr. W. Richard Webb, who transformed thoracic imaging with his foundational work in HRCT. The lecture explores multiple conditions diagnosed via HRCT and the radiologist's role on multidisciplinary teams.
Researchers developed an AI-based method for virtual staining of histopathological tissue samples, producing images that closely resemble actual chemical-stained samples. This reduces chemical burden and manual work while enabling the use of tissues for other purposes.
The study used LED imaging to visualize droplet spreading during dental procedures and found that extra-oral and intra-oral suction devices reduced droplet and aerosol spread by 97.8% and 92.1%, respectively. Understanding the dynamics of aerosols and droplets is crucial for preventing infectious diseases like COVID-19.
A new imaging technique using a novel nuclear medicine tracer has been developed to improve the treatment of intestinal strictures in Crohn's disease patients. The technique allows for precise differentiation between inflammation and fibrosis, enabling targeted therapy.
A printable multi-energy X-ray detector made from perovskite thin films has been developed with enhanced flexibility and sensitivity. The detector can operate in a broad energy range, from 0.1 KeV to tens of KeV, making it suitable for real-time detection and imaging applications such as disease diagnosis and explosives detection.
Researchers have developed a new microfluidic chip that enables the visualization of rare cell types and disease patterns in tissue. This method combines imaging and sequencing techniques, providing spatially resolved information about individual cells and their environment.
Researchers have developed a non-invasive method to track human aging using retinal scans, which are less expensive and more accurate than other aging clocks. The study found that changes in the eye can provide an actionable evaluation of gero-protective therapeutics, offering a new tool for tracking aging.
A national survey of radiologists reveals moral distress is widespread, often stemming from high case volumes and lack of administrative support. The study highlights the need for education of leadership regarding causes of distress among frontline workers to address this issue.
Physicists at Delft University of Technology have developed a new technology on a microchip combining optical trapping and frequency combs to measure distances with high precision in opaque materials. The technology uses sound vibrations instead of light, offering a simple and low-power solution for applications such as monitoring the ...
A novel lensless radiomicroscope has been developed for nuclear medicine imaging, offering high-resolution images of cells with a field of view over 500 times larger than current technology. The instrument is made from off-the-shelf parts and costs less than $100, making it accessible to more labs.
A study of US newspaper articles found that lung cancer screening was generally positively covered between 2010 and 2022. However, crucial aspects such as enrollment criteria and cost issues were frequently omitted, highlighting a need for radiologists to take an active role in media coverage.
Researchers from Dartmouth and University Medicine Essen found that strong links between brain measures and traits can be obtained when machine learning algorithms are utilized. This approach allows for high-powered results from moderate sample sizes, opening up studies of many traits and clinical conditions previously inaccessible.
Researchers found that increasing melanin levels in human skin reduces Cherenkov emission intensity, while blood concentration affects different color channels. The study suggests using multispectral signatures to correct attenuated signals based on patient's blood volume or skin color.
Researchers at the University of Sheffield have developed a new ultrasound method to measure tissue tension, which can indicate abnormal tissue, scarring, and cancer. This breakthrough technique uses sound waves to detect forces within soft tissue, enabling earlier disease diagnosis and potentially improving healthcare outcomes.
A team of POSTECH researchers has developed a deep learning-based multimodal fusion network for segmentation and classification of breast cancers using B-mode and strain elastography ultrasound images. The method achieved high accuracy in distinguishing benign from malignant lesions, outperforming conventional methods.
A new method for image reconstruction in electrical impedance tomography has been developed by Prof. DU Jiangfeng's team, enabling high-quality images without training data. This innovation provides crucial theoretical support for the application of electrical impedance tomography in various diseases.
Automated AI body composition tools achieved high technical adequacy rates (97.8-99.1%) in a large sample of external abdominal CT scans. The results support the generalizability and potential for widespread adoption of these tools, which can help build trust among radiologists and physicians.
Researchers created adaptive optical phantoms by combining multiple pigments to mimic target tissue's optical properties, successfully validating them in extensive experiments. The new platform enables broader band spectra for emerging hybrid modalities and novel instruments.
A new project aims to expand the network of European bioimage analysts and make cutting-edge techniques accessible globally. The initiative will support career development for young researchers and enhance collaboration among imaging and life science communities.
Researchers have developed a novel approach for imaging red blood cells and oxygenation using color TSFG microscopy, enabling label-free visualization of RBCs. The technique provides chemically specific contrast and can measure oxygenation dynamics in vivo, with potential applications in medical technology and biological studies.
Researchers at the Queensland Brain Institute developed a new ultrafast fMRI technique to study brain activity. The technique revealed that brain activity follows preferential circuits depending on cell type organisation, and blood signals can reflect network organisation and cell type distribution.
Researchers from the University of Tsukuba have found distinct changes in brain activity patterns that can differentiate between Alzheimer's disease and Lewy body dementia. These changes were present before symptoms emerged and could be detected using conventional MRI.
The Chan Zuckerberg Initiative grant will support advanced imaging training for over 2,000 scientists through a combination of hands-on events and online resources. Array tomography, an imaging technique that offers higher resolution and detailed molecular labeling, will be a key focus of the project.
Scientists at Göttingen University have created a novel approach to produce X-ray images in color from a single exposure, eliminating the need for focusing and scanning. This breakthrough method uses an X-ray color camera and a specially structured plate to capture the intensity pattern of fluorescing atoms in a sample.
Researchers developed a novel, calibration-free, and reconstruction-free imaging technique that directly obtains a clear image from a single shot of speckle images. The method uses real-time video imaging to build a three-dimensional image, taking slices through the speckles.
Researchers at Beckman Institute for Advanced Science and Technology will use a unique combination of MRI and mass spectrometry imaging techniques to study Alzheimer's disease on an unprecedented scale. By capturing detailed chemical information, they aim to understand the molecular-level changes that occur in aging brains and develop ...
A study found that nearly half of Medicare patients with non-small cell lung cancer do not receive the recommended FDG PET/CT imaging prior to radiation therapy, leading to decreased overall and cancer-specific survival rates. This lack of adherence is linked to greater mortality rates.
Researchers analyzed MRI images of Olympic athletes' injuries during the 2020 Tokyo Games and found that those with tendon tears were more likely to drop out. The study aims to improve injury management and prevention by understanding different types of muscle injuries and their outcomes.
A recent study from West Virginia University found that states permitting nurse practitioners to exercise full practice authority (FPA) tend to have a more diverse and racially aligned NP workforce. In FPA states, Black NPs serve 2.8% more Black Medicare beneficiaries than in non-FPA states.
Researchers found that individuals who were resistant to anaesthesia had fundamental differences in the fronto-parietal regions of the brain, which could be predicted prior to sedation. These brain differences may require higher doses of propofol for individuals to become nonresponsive.
KAUST researchers have designed and built novel organic scintillator materials for detecting X-rays at low doses, overcoming stability issues with existing ceramic or perovskite materials. The new approach uses heavy atoms to improve X-ray absorption capability and exciton utilization efficiency.
Researchers found that MMTV-NeuT/ATTAC mice treated with anti-PD-1 therapy developed increased tumor-associated macrophages, EMT, fibroblast proliferation, and enhanced extracellular matrix. These findings suggest potential therapeutic avenues to enhance PD-1 immune checkpoint sensitivity.
Researchers developed an innovative deep learning-based approach to classify epileptic seizures in near real-time, overcoming clinical limitations. The system uses action recognition with intelligent 3D cropping to distinguish between frontal and temporal lobes seizures or non-epileptic events.
The procedure requires thorough pre-procedural evaluation and assessment of anatomical and hemodynamic data. Long-term results indicate that TPVR can effectively restore RVOT function, while improving survival rates and reducing the need for reintervention across age groups. Vigilant testing is recommended to avoid rare but serious com...
Researchers from Nagoya University found that 5G network telemedicine is effective in providing telerehabilitation and teletreatment in a remote community. The study showed that 5G transmitted high-definition images with low latency, reducing the need for physical therapist visits and improving patient outcomes.
A new mesoscopic oblique plane microscopy method captures up to three times more resolvable image points than other similar systems, enabling whole-body volumetric recordings of neuronal activity and blood flow dynamics. The technique allows for single-cell tracking within the complete 3D circulation system for the first time.
A study published in the American Journal of Roentgenology found that reconstruction using BI64 kernel and 0.4-mm slice thickness yielded improved bronchial division identification and pulmonary fissure sharpness without loss in pulmonary vessel sharpness or pathology conspicuity.
Researchers developed a novel optical technique to directly monitor cerebral oxygen consumption and blood flow in real-time. The technique uses phosphorescent probes to track the brain's oxygen gradient, providing valuable insights into brain activity and metabolism.
A new deep learning model using conditional generative adversarial networks (cGANs) has been developed to diagnose bowel ischemia. The model achieved an accuracy of over 93% in segmenting ischemic intestine images, surpassing current subjective methods.
Researchers found variable voltages in breast cancer cell membranes, which may indicate an electrical communication network between cells. This discovery could lead to new treatments by disrupting this network, potentially making cancer cells easier to treat.
Researchers at Duke University have developed a new method called 3D Tracking and Imaging Microscopy (3D-TrIm) to capture real-time 3D footage of viruses approaching cells. This technique combines two microscopes in one, allowing scientists to study the early moments before infection begins and how viruses navigate complex barriers.
Researchers develop hybrid brightfield-darkfield transport of intensity approach, expanding accessible sample spatial frequencies and achieving 5-fold resolution increase. This method enables precise detection and quantitative analysis of subcellular features in large-scale cell studies.
Dual-energy CT (DECT) metrics provide insight into physiologic changes after lung cancer surgery, beyond PFTs and conventional CT. Lung perfusion ratio increases were greater after lobectomy than limited resection.
A new project aims to create more race-inclusive AI in medicine by developing a distributed, inclusive data collection and learning framework that relies on smartphone apps. The framework uses federated learning, which allows models to be trained on device data while protecting user privacy.
A deep learning algorithm trained on adult populations shows improved sensitivity for diagnosing thyroid nodules in children, with higher accuracy than traditional ACR TI-RADS. The study suggests the potential benefits of this new approach for evaluating thyroid nodules in younger patients.
A study published in the American Journal of Roentgenology found that large specialized referral centers for complex IVC filter retrievals achieved a 100% success rate and low adverse event rates. Prolonged filter dwell time was identified as an independent predictor of adverse events.
Researchers from Xi'an Jiaotong-Liverpool University found that brain stimulation combined with a nose spray containing nanoparticles can improve recovery after ischemic stroke. The treatment increased cognitive and motor functions, and weighed more quickly than those treated with TMS alone.
Researchers used advanced imaging techniques to understand the structure of bacterial propellers, which are made of a single protein. The study reveals that bacteria push themselves forward by coiling these appendages into corkscrew shapes, and that similar structures have evolved independently in archaea.