A machine learning system has been developed to identify cancer cells based on their light scattering patterns, achieving high accuracy even with cells having similar morphology. The system, which uses dark-field microscopy and machine learning algorithms, has shown promise in distinguishing between different types of cancer cells.
The AI-powered platform, OralScan, uses artificial intelligence and user-captured photographs to identify common dental conditions, enabling longitudinal tracking of oral health findings. The platform targets seven conditions, including dental caries, periodontal disease, and tooth loss, and is available for licensing or collaboration.
Researchers developed V-UNet, a novel model combining global and local information to address noise and redundant information in medical images. The model achieved competitive segmentation performance while maintaining low computational requirements, promising a more efficient and robust AI-assisted medical image analysis.
A scoping review found that only 77 studies validating 52 medical AI products reported demographic data, highlighting the need for more transparent reporting to confirm safe and unbiased performance. The review also showed that performance reporting for demographic subgroups is inadequate, posing a risk to patient care and outcomes.
A new study found that the number of radiologists in practice rose only 12% between 2010 and 2022, despite a 33% growth in radiology residency positions. The radiologist pipeline expanded more slowly than medicine as a whole, highlighting an ongoing workforce strain.
A new study found that laws requiring notification of dense breast tissue led to modest increases in follow-up imaging tests, while insurance mandates had a significant impact, increasing ultrasound use four times. This study examined the influence of notification laws and insurance coverage independently.
Prostate cancer affects 1 in 8 men, with 60% diagnosed in men 65 and older. Clinical trials like Alliance's help improve early detection and treatment, reducing mortality rates by half since 1993. These trials also explore new treatments and survivorship strategies.
Researchers developed a computational model to study aging in 40 types of human tissue, identifying three major aging patterns and underlying molecular changes. Tissues show bimodal structural aging, with accelerated aging from 35-40 and 55-60, coinciding with fertility decline and menopause.
Researchers have found a reliable alternative to bone marrow sampling for children with rhabdomyosarcoma, using PET scans to detect bone marrow involvement with high accuracy. This approach could spare patients from painful procedures and improve diagnostic practices.
USC researchers have developed custom, 3D-printed MRI sensors that provide clearer images of small organs in infants and children. The sensors, which can be customized to individual patients, are made in under 10 minutes and cost around $30.
A new study reveals that fluctuations in estrogen influence brain connections in rats, using a specialized imaging technique called magnetic resonance elastography (MRE). Researchers hope to translate the findings to humans, exploring how hormonal changes affect brain health in women, particularly during menopause.
Researchers from Hebei Medical University review current evidence on prolonged disorders of consciousness (pDOC) management. They propose a brain communication network model to explain how brain injury leads to long-term unconsciousness, identifying key brain electrical signaling activity measures for diagnosis and treatment.
Researchers at the University of Rochester have developed a lower-cost imaging system that overcomes challenges in near-infrared light transmission through deep tissue and dense fog. The AI-enhanced time-gating technique produces clearer images in these environments, improving applications such as cancer detection and LiDAR systems.
Researchers explore how engineered lanthanide carriers can tune the optical and magnetic properties of ions for improved signal contrast and reduced background signals in biomedical imaging. The study highlights opportunities for multimodal imaging, theranostics, and AI-guided probe design.
Recent progress in lanthanide carriers is mapped to improve biomedical imaging and diagnosis, combining optical, X-ray, and magnetic resonance imaging techniques. These carriers enable high-resolution imaging with tunability for different applications, reducing tissue damage and enhancing image clarity.
A new study suggests venture capital is playing a significant role in driving innovation in radiology, particularly in medical devices and artificial intelligence. The study found $11.4 billion was invested across 646 companies between 2000-2023, with funding peaking in 2021.
The AKI-PURIFY survey reveals substantial variability in AKI prevention strategies and KDIGO implementation across ICUs. Despite strong evidence supporting structured preventive approaches, only one-third of respondents reported routine implementation of kidney-protective bundles.
A new study found that socioeconomic challenges hinder lung cancer screening adherence and equity, highlighting the need for targeted interventions to improve patient experience. Lower-income participants reported greater discomfort or anxiety during low-dose chest CT screening.
Researchers have developed a polymer-based microring resonator array with over 40 elements, demonstrating broadband acoustic detection and fine spatial resolution. The system achieved strong correspondence with biological structures, including blood vessel regions, in imaging mouse prostate tissue.
FireANTs, an open-source algorithm, combines AI optimization and geometry to quickly match complex medical images. The new method can accomplish what took weeks in minutes, detecting subtle changes that signal disease or cognitive decline, making it practical for clinical practice.
A neural network-based machine learning model accelerates diffuse optical tomography by over a million-fold, enabling real-time diagnosis. The model accurately reproduces signals even for unseen parameter combinations, with each inference taking approximately 2 milliseconds.
A new set of consensus standards for classification, annotation, and quality control is being implemented for artificial intelligence applications in dry eye imaging. These standards aim to enhance consistency and multicenter collaboration in AI-assisted diagnosis.
Researchers developed a novel diagnostic support framework using visual question answering to generate interpretable findings from chest CT images. The system demonstrated strong agreement with reference descriptions and provided clinically meaningful outputs.
SourceMeijo University·JournalInternational Journal of Computer Assisted Radiology and Surgery·TypeComputational simulation/modeling·DateMay 11, 2026
After a mild head injury, CSF motion increased in some regions and decreased in others. The technique may help clarify the link between post-traumatic brain conditions and changes in cognitive function.
Researchers developed multiplexed MRI technology, enabling simultaneous imaging of signals from multiple molecules in the brain. The technology provides a comprehensive view of the brain's structure, physiology, and molecular processes, allowing for more precise diagnosis and individualized treatment planning.
A new AI system, SPARK, enables autonomous scientific discovery in cancer pathology by generating biological hypotheses and translating them into analytical tools. The system identifies clinically relevant tissue markers linked to disease course, treatment response, and tumour progression.
New technologies are shifting healthcare towards proactive care, using AI for early heart disease detection and miniaturized CRISPR-based diagnostics. These innovations enable seamless integration of existing medical screenings and provide laboratory-grade accuracy in a portable format.
The research team developed a novel AI pathology analysis system named PRET that can accurately recognize multiple types of cancer using only a minimal number of samples. PRET outperformed existing methods in 20 tasks, achieving high diagnostic accuracy rates and stable generalizability across different populations and regions.
The center offers state-of-the-art facilities for practical skills acquisition, digital technologies integration, and innovation in medical procedures. Future physicians will benefit from cadaver training, real-time imaging, and collaborative research opportunities.
A new study found that A&E patients with visible blood in their urine who receive a scan within 48 hours are 2.5% less likely to die within three months. The study also showed that patients with cancer are diagnosed significantly faster when they receive prompt investigation.
A new imaging test, PSMA PET/CT scan, has been shown to safely reduce the number of biopsies needed for suspected prostate cancer, with no harm to patients. The PRIMARY2 trial found that the scan could identify low-risk patients who did not need a biopsy, while targeting suspicious areas for those who did.
Recent advances in photonic nanomaterials and healthcare devices have led to the development of wearable and implantable medical devices. These devices utilize light for precise manipulation of cells and tissues, offering new possibilities for early disease detection, light-based therapies, and personalized precision medicine.
Research presented at EAU26 confirms that prostate cancer screening leads to a reduction in mortality, but also detects cancers that would otherwise remain undetected. The study's findings suggest that using MRI in prostate cancer screening can reduce overdiagnosis and unnecessary biopsies.
Artificial intelligence can complement human expertise to improve radiology workflows and patient care. The JACR Focus Issue on AI in Workflow Optimization explores the integration of AI technology to enhance efficiency and deliver better care.
The study reveals marked variation in radiation dose to patients from diagnostic testing, particularly affecting low- and middle-income countries. Standardized protocols and updated equipment are urgently needed to reduce global radiation exposure and improve the quality of CAD diagnosis.
A new study highlights the clinical relevance of brain functional connectome uniqueness in identifying biomarkers for major depressive disorder. Researchers found that patients with MDD displayed reduced FC uniqueness, especially in frontoparietal and sensorimotor networks.
A new JACR focus issue explores alternative approaches to resident education, radiology residency applications, and investing in pre-clinical medical education. The study aims to spark meaningful dialogue around how radiology education is valued, supported, and delivered amidst economic challenges.
Steatotic liver disease can be precisely assessed using three-dimensional ultrafast vascular ultrasound. The technology visualizes subtle microvascular changes, enabling real-time monitoring of disease progression and therapeutic response.
A study published in Biological Psychiatry identifies a distinct immuno-inflammatory biomarker in the brain linked to immune system dysfunction and poorer response to standard treatments. The findings provide potential value for clinical prediction and precision therapies.
A new study led by Concordia researchers found that individuals with coronary artery disease (CAD) have widespread structural changes in their white matter compared to healthy controls. The changes were particularly noticeable in regions critical for cognitive and motor functions.
Researchers developed a revolutionary AI platform to assess the fairness and accuracy of commercial algorithms in detecting diabetic eye disease. The platform was tested on 1.2 million images from diverse ethnic groups, achieving high accuracy rates compared to human grading.
A study at Technical University of Munich found that AI-simplified CT reports reduced reading time to two minutes and improved patient comprehension. Patients rated the simplified texts as more helpful and informative, with 82% finding them easier to read and understand.
A new model called DAC enables medical image segmentation with limited labelled data, achieving consistent generalization across unseen domains. The approach uses feature-level supervision and asymmetric co-training to reduce errors, especially in low-contrast structures.
Artificial intelligence is being used to improve diagnostic accuracy in molecular pathology, with studies showing 93% accuracy for cancer diagnosis. Researchers are also using AI to analyze chromosomal changes in blood cancer patients and provide personalized treatment plans.
A deep learning system developed by Incheon National University integrates images and clinical details to improve early skin cancer diagnosis. The model achieved 94.5% accuracy, outperforming popular image-only models.
The new Molecular Imaging and Theranostics Centre at NUH and NUS Medicine enables faster, safer and more precise diagnoses, while researchers can observe real-time tracer movement throughout the body. This opens possibilities for validating next-generation diagnostics and therapies, advancing theranostics.
A new study has uncovered 42 genetic locations associated with hypertrophy of the left ventricle, a major risk factor for sudden death. The research, conducted using three-dimensional MRI images and genome-wide analysis, could lead to earlier identification of individuals at greater risk.
Researchers at the University of Arizona are developing a new optical technology that can image deep into biological tissues without invasive procedures. This approach aims to overcome current challenges in skin cancer imaging, allowing for earlier detection, precise evaluation, and real-time monitoring of treatment response.
A new study shows that a multi-step effort to reduce unneeded pre-operative testing, using a tailored program for each hospital, can lead to significant reductions in wasteful tests. The findings have already spurred an expansion of the program to 16 more Michigan hospitals.
The PRISM trial will examine the impact of AI on mammogram accuracy and patient experience, with hundreds of thousands of mammograms analyzed. The study aims to understand whether AI enhances cancer detection by radiologists or leads to more false alarms.
Researchers developed an AI model that diagnoses achalasia using plain chest X-rays, outperforming physicians' reviews. The diagnostic performance of the AI model demonstrated higher sensitivity and specificity, enabling early diagnosis and potentially improving treatment efficacy.
A new machine learning model predicts heart disease risk in women by analyzing mammograms, offering a 'two-for-one' screening approach that combines breast and cardiovascular screenings. The model performs comparable accuracy to traditional risk calculators without requiring extensive clinical data.
Researchers have demonstrated a portable, noninvasive technology that can detect metabolic changes linked to Alzheimer's disease by measuring cytochrome c oxidase activity. The study found that including oxCCO measures improved the ability of the brain-monitoring tool to capture clinically relevant brain changes.
The International Osteoporosis Foundation identifies critical global barriers to osteoporosis care, including limited DXA scanning and outdated treatment criteria. The IOF advocates for a paradigm shift in bone health management worldwide, recognizing high fracture risk as a valid criterion for treatment and reimbursement.
Scientists at Saarland University have developed a pioneering robot-assisted procedure for joint replacement surgery, eliminating the need for bone pins and external infrared cameras. The new approach uses the robot's built-in sensors to precisely scan the surgical field and create an accurate 3D model.
A new diagnostic method confirms sepsis infections in as little as two hours, cutting critical time for treatment. The technique uses centrifugation and artificial intelligence to detect bacteria in blood samples, enabling prompt antibiotic treatment and reducing survival rates by 8% per hour of delayed treatment.
A new diagnostic schema for COPD improves patient identification by incorporating CT lung imaging and respiratory symptoms. This approach identified additional individuals at risk of poor respiratory outcomes and excluded those with airflow obstruction without respiratory symptoms.
A new AI tool can learn to read medical images with far less data, cutting down the amount of required data by up to 20 times. The tool improves upon medical image segmentation, a labor-intensive task often performed by experts, and boosts model performance in settings with limited annotated data.
Researchers develop compact, noninvasive imaging system combining LC-OCT and Raman microspectroscopy to examine skin cancer structures and chemical composition. The AI model achieves high classification accuracy for basal cell carcinoma and other types, offering new insights into cancer development and behavior.
The 250,000-square-foot facility provides greater availability to high-quality integrated care, addressing the region's increasing demand for innovative diagnostic and treatment services. The pavilion brings together a broad range of nationally ranked clinical programs, including cancer clinics, neurology, and surgery.