cabinet x ray system  (KUB Technologies Inc)

Journal: Nature Machine Intelligence

Article Title: Synthetic X‑ray‑driven tracking and control of miniature medical devices

doi: 10.1038/s42256-026-01190-3

Figure Lengend Snippet: a , Real tissue images under X-ray imaging. b , The generation of precisely matched synthetic tissues with randomized and enhanced textures. c , Image generation with mask-guided conditioning and prompts. Randomization of masks and prompts substantially expands the dataset. d , A domain comparison of real and synthetic tissues. Features from 1,140 real and 24,803 synthetic tissue images, extracted using Inception V3, are visualized via PCA. e , An overview of the open-source MMD dataset under X-ray imaging. Dataset 1 (D1) contains real-time recordings and annotations of MMD locomotion. Dataset 2 (D2) includes static MMD images captured under various voltage and current conditions. Dataset 3 (D3) provides synthetic images with corresponding segmentation labels. The suffixes ‘-s’ and ‘-l’ denote the soft and liquid MMDs, respectively.

Article Snippet: As shown in Supplementary Fig. , the soft MMDs were imaged with a clean background under X-ray cabinet imaging (XPERT 80, KUBTEC Scientific) with varying voltages and currents.

Techniques: Imaging, Comparison

Journal: Nature Machine Intelligence

Article Title: Synthetic X‑ray‑driven tracking and control of miniature medical devices

doi: 10.1038/s42256-026-01190-3

Figure Lengend Snippet: a . Impact of X-ray voltage and current on image contrast and noise in static capture mode. b–e . Influence of voltage, current, and frame capture rate ( \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f$$\end{document} f ) on contrast and noise in continuous capture mode. Data are presented as mean values \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\pm$$\end{document} ± standard deviation. Error bars denote standard deviation across five images. f . Quantification of motion-induced blurriness as a function of magnet translation speed ( \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${v}_{{\rm{mag}}}$$\end{document} v mag ) under continuous capture. Blurriness is assessed via the normalized Tenengrad sharpness metric, where lower values indicate increased motion blur. All experiments were performed using the XPERT 80 imaging system (KUBTEC Scientific). In all figures, scale bars represent 5 mm.

Article Snippet: As shown in Supplementary Fig. , the soft MMDs were imaged with a clean background under X-ray cabinet imaging (XPERT 80, KUBTEC Scientific) with varying voltages and currents.

Techniques: Standard Deviation, Imaging