medical-imaging

BackgroundEndometriosis is a common gynecologic condition in which pelvic MRI plays an important role in diagnosis and preoperative assessment. AI-enabled automated uterus segmentation on pelvic MRI could support endometriosis care by enabling standardized volumetric measurements and quantitative imaging analyses. However, developing robust AI models for this task is challenging because endometri…

IntroductionMulti-modal image registration leverages complementary information from diverse imaging sources to achieve precise spatial alignment. However, aligning visible (VIS), near-infrared (NIR), and thermal (TH) modalities remains challenging due to appearance differences and limited annotated datasets.MethodsThis study proposes a ResU-Net-inspired framework combining heatmap prediction and …

High‑quality 3D medical imaging data is the foundation of modern radiology AI, but access to it is often constrained by data scarcity, privacy restrictions,...

Meningiomas are the most common brain cancers arising from the protective soft tissue cover of the brain called the meninges.

IntroductionThe process of precise structural evaluation for paranasal sinuses based on CT scan data establishes a foundation for medical professionals to assess human anatomical variations, supporting the diagnosis and treatment of ear, nose, and throat (ENT) conditions. Existing deep learning methods face difficulties in analyzing complex sinus structures due to limited annotated datasets and l…

Breast cancer is the second most common cancer among women in the United States today. Now, advanced technologies originally developed at the DOE's Thomas Jefferson National Accelerator Facility (Jefferson Lab) for studying the tiniest particles inside matter have been adapted to aid doctors in diagnosis and treatment of breast cancer patients.

Today we schedule a medical imaging scan in a clinic, doctor’s office, or at a hospital. What if we could self-monitor with a wearable unit? Researchers are working towards enabling practical wearable units and have taken two key steps to making this a reality. The post Medical imaging in a wearable form: monitoring health on the go first appeared on Acoustics.org .

Although the porous skull severely impacts the quality of photoacoustic brain images, the deep learning model U-Net can be used for effective image post-processing. This talk explores U-Net’s robustness by training on images with 2D wave effects, then later testing on images with 2D+3D wave effects. The post Using Deep Learning to Enhance Photoacoustic Brain Images first appeared on Acoustics.org…

Nature Biomedical Engineering, Published online: 24 April 2026; doi:10.1038/s41551-026-01662-2 The OCTCube models integrate optical coherence tomography and other imaging modalities to achieve state-of-the-art performance in predicting retinal diseases with strong generalizability across cohorts, devices and modalities.

Accurate detection of pediatric fractures in radiographs remains challenging due to subtle visual cues and the high prevalence of false-positive detections produced by automated systems. To address this limitation, we propose a lightweight region-of-interest (Region of Interest) adjudication framework that operates as a second-stage verification module to refine detector-generated candidates. The…

Medical image segmentation—the process of partitioning digital images into distinct anatomical or pathological regions—is a critical step in clinical workflows, including tumor volume estimation, treatment planning, and disease progression monitoring. Manual segmentation by radiologists is time consuming, prone to inter-observer variability, and increasingly impractical given the rising volume of…

Journal of Computer Science, Published online: 18 April 2026; doi:10.3844/jcssp.2026.1254.1278 Image denoising is a vital process in medical imaging that involves removing noise or distortions introduced during image acquisition. Random noise can degrade image quality and reduce contrast, makin...

BackgroundBias in medical image segmentation can lead to unequal performance across demographic subgroups, raising concerns about fairness and reliability in clinical AI systems. While deep learning models have achieved high segmentation accuracy, ensuring equitable performance across race and gender remains a significant challenge, particularly in privacy-sensitive healthcare environments.Method…

AI Diagnoses Knees, Drugs, Configs, and License‑Buying Agents Artificial intelligence is moving beyond research labs into everyday tools. From medical imaging to drug pipelines, from config automation to autonomous agents, the week shows how AI is reshaping both health and development workflows. The momentum spans clinical trials, financing rounds, and open‑source tooling, signaling a broader shi…

Nature Biomedical Engineering, Published online: 26 March 2026; doi:10.1038/s41551-026-01638-2 A video-based deep-learning system was trained to understand the spectrum of human cardiovascular disease by the self-supervised method of contrastive learning, using pairs of cardiac MRI scans and their corresponding text reports that are generated as part of routine clinical practice.

Nature Biomedical Engineering, Published online: 07 April 2026; doi:10.1038/s41551-026-01639-1 BUSGen is a foundation generative model designed for analysing breast ultrasound images that supports diverse tasks and improves breast cancer screening, diagnosis and prognosis.

A new hybrid imaging system brings optical contrast into ultrasound to visualize the human body in 3D.