Academic Radiology, Год журнала: 2024, Номер unknown
Опубликована: Ноя. 1, 2024
Язык: Английский
Academic Radiology, Год журнала: 2024, Номер unknown
Опубликована: Ноя. 1, 2024
Язык: Английский
Journal of Robotic Surgery, Год журнала: 2025, Номер 19(1)
Опубликована: Апрель 17, 2025
Язык: Английский
Процитировано
0Engineering materials, Год журнала: 2025, Номер unknown, С. 297 - 329
Опубликована: Янв. 1, 2025
Язык: Английский
Процитировано
0Ultrasonics, Год журнала: 2025, Номер 155, С. 107709 - 107709
Опубликована: Май 27, 2025
Язык: Английский
Процитировано
0Biomedical Optics Express, Год журнала: 2024, Номер 15(8), С. 4689 - 4689
Опубликована: Июль 10, 2024
Accurate prediction of breast cancer (BC) is essential for effective treatment planning and improving patient outcomes. This study proposes a novel deep learning (DL) approach using photoacoustic (PA) imaging to enhance BC accuracy. We enrolled 334 patients with lesions from Shenzhen People's Hospital between January 2022 2024. Our method employs ResNet50-based model combined attention mechanisms analyze ultrasound (PA-US) images. Experiments demonstrated that the PAUS-ResAM50 achieved superior performance, an AUC 0.917 (95% CI: 0.884 -0.951), sensitivity 0.750, accuracy 0.854, specificity 0.920 in training set. In testing set, maintained high performance 0.870 0.778-0.962), 0.786, 0.872, 0.836. significantly outperformed other models, including PAUS-ResNet50, BMUS-ResAM50, BMUS-ResNet50, as validated by DeLong test (p < 0.05 all comparisons). Additionally, improved radiologists' diagnostic without reducing sensitivity, highlighting its potential clinical application. conclusion, demonstrates substantial promise optimizing diagnosis aiding radiologists early detection BC.
Язык: Английский
Процитировано
2Advanced Optical Materials, Год журнала: 2024, Номер unknown
Опубликована: Ноя. 7, 2024
Abstract Serendipitously discovered, carbon dots (CDs) have attracted significant attention as a potential contrast agent for photoacoustic imaging (PAI) in the biomedical sector. CDs play an essential role PAI, contributing significantly to early detection of diseases and monitoring treatment progress, particularly tumor imaging. This review emphasizes domain highlighting their characteristics like biocompatibility, enhanced spatial resolution, optical absorption NIR region, facile surface functionalization tumor‐ targeted The study explores use enhancing resolution PAI improved visualization tumors organs such breast, cervical, liver, gastrointestinal, skin, cardiovascular system, nervous others. Challenges associated with optimizing signal‐to‐noise ratio ensuring stability under physiological conditions, also been discussed.
Язык: Английский
Процитировано
2Advanced Functional Materials, Год журнала: 2024, Номер 34(44)
Опубликована: Май 29, 2024
Abstract Current diagnostic technique in direct identification of multi‐site plaques and simultaneous assessment plaque vulnerability remains a challenge, which is crucial for indicating the risk atherosclerotic cardiovascular diseases (ASCVD). Herein, an osteopontin (OPN)‐specific nanoprobe (OPN Ab‐Au/FeNiPO 4 @ICG) with both multiple spectra optoacoustic tomography (MSOT) computed (CT) imaging, constructed successfully realizing systemic screening vulnerable plaque. OPN @ICG specifically targeted OPN‐overexpressed foam cells recognized at molecular level. In AS mice, CT imaging exhibits that effectively avoid interference from calcification accurately visualized MSOT functional results reveals after injection nanoprobe, carotid exhibited much higher signal than aortic arch ( P = 0.0291). Further pathological analysis displays possessed score 0.0247), agreement signals. More importantly, linear regression confirms high correlation between signals R 0.7095 0.0216), demonstrating potential proposed systematic evaluation vulnerability. This work employs dual‐model strategy localization assessment, greatly advancing accurate diagnosis ASCVD.
Язык: Английский
Процитировано
1Applied Sciences, Год журнала: 2024, Номер 14(12), С. 5331 - 5331
Опубликована: Июнь 20, 2024
Photoacoustic imaging (PAI) is an emerging technique that offers real-time, non-invasive, and radiation-free measurements of optical tissue properties. However, image quality degradation due to factors such as non-ideal signal detection hampers its clinical applicability. To address this challenge, paper proposes algorithm for super-resolution reconstruction segmentation based on deep learning. The proposed enhanced minimalistic network (EDSR-M) not only mitigates the shortcomings original regarding computational complexity parameter count but also employs residual learning attention mechanisms extract features enhance details, thereby achieving high-quality PAI. DeepLabV3+ used segment images before after verify performance. experimental results demonstrate average improvements 19.76% in peak-signal-to-noise ratio (PSNR) 4.80% structural similarity index (SSIM) reconstructed compared those their pre-reconstructed counterparts. Additionally, mean accuracy, intersection union (IoU), boundary F1 score (BFScore) showed enhancements 8.27%, 6.20%, 6.28%, respectively. enhances effect texture PAI makes overall structure restoration more complete.
Язык: Английский
Процитировано
1Nondestructive Testing And Evaluation, Год журнала: 2024, Номер unknown, С. 1 - 29
Опубликована: Сен. 17, 2024
Язык: Английский
Процитировано
1Advanced imaging., Год журнала: 2024, Номер 1(3), С. 032002 - 032002
Опубликована: Янв. 1, 2024
Язык: Английский
Процитировано
1Applied Sciences, Год журнала: 2024, Номер 14(12), С. 5161 - 5161
Опубликована: Июнь 13, 2024
Photoacoustic imaging integrates the strengths of optics and ultrasound, offering high resolution, depth penetration, multimodal capabilities. Practical considerations with instrumentation geometry limit number available acoustic sensors their “view” target, which result in image reconstruction artifacts degrading quality. To address this problem, YOLOv8-Pix2Pix is proposed as a hybrid artifact-removal algorithm, advantageous comprehensively eliminating various types effectively restoring details compared to existing algorithms. The algorithm demonstrates superior performance artifact removal segmentation photoacoustic images brain tumors. For purpose further expanding its application fields aligning actual clinical needs, an experimental system for detection designed paper be verified. results show that processed are better than pre-processed terms metrics PSNR SSIM, also significantly improved, provides effective solution development technology.
Язык: Английский
Процитировано
0