Boyi Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Boyi Li has a total of 129 co-authors. Here are the top co-authors by number of shared publications:
- Dean Ta
Fudan University
25 shared publications - Xin Liu
College of Biomedical Engineering Fudan University Shanghai China
14 shared publications - Chengcheng Liu
Fudan University
11 shared publications Dean Ta
10 shared publicationsWenting Gu
Fudan Univ, Acad Engn & Technol, Shanghai 200433, Peoples R China
6 shared publicationsChengcheng Liu
Academy for Engineering and Technology, Fudan University, Shanghai, China
5 shared publications
Latest Publications
- Open Access
- 18 November 2026
- Published by Elsevier in Photoacoustics
Multiparametric photoacoustic imaging for detecting and grading of prostatic inflammation in a rat model of chronic prostatitis/chronic pelvic pain syndrome
- 18 November 2026
- Published by Elsevier in Photoacoustics
Open AccessMultiparametric photoacoustic imaging for detecting and grading of prostatic inflammation in a rat model of chronic prostatitis/chronic pelvic pain syndrome
- 17 August 2026
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 128
A multimodal ultrasound adaptive feature fusion network for prostate cancer grading
- 17 August 2026
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 128
A multimodal ultrasound adaptive feature fusion network for prostate cancer grading
- Open Access
- 5 August 2026
- Published by IOP Publishing in Chinese Physics B
A Multi-Modal Ultrasonic Instrumentation Integrated Attention-Based Y-Shaped Network for Bone Imaging Correction
Medical ultrasonic imaging technique (USI) is widely applied in clinical settings. However, it has not yet been applied for imaging and diagnosis of internal tissues of bones. This study aims to investigate and establish a multi-modal ultrasonic in bone imaging correction (MUBIC) paradigm combining ultrasonic (USI) and photoacoustic (PAI) advantages, and then to develop a USI and PAI bone imaging instrument (UPBII) by training and integrating with an attention-based Y-shaped network (AY-Net) for multi-modal ultrasonic imaging correction. The performance of MUBIC is demonstrated through both simulation and experiments. The results showed that contrast-to-noise ratio (CNR) of the corrected ultrasonic images increased from 3.2 decibels to 7.8 decibels, and the peak signal-to-noise ratio (PSNR) increased from 21.6 decibels to 34.3 decibels; while CNR of the corrected photoacoustic images increased from 2.8 decibels to 7.5 decibels, and PSNR increased from 22.6 decibels to 34.2 decibels. Moreover, the processing time of MUBIC was less than 1.0 second in both USI and PAI modalities. This study developed and applied UPBII to verify the performance of AY-Net, thereby demonstrating that MUBIC is a high-speed and effective paradigm for correcting USI and PAI of bones.
- 5 August 2026
- Published by IOP Publishing in Chinese Physics B
Open AccessA Multi-Modal Ultrasonic Instrumentation Integrated Attention-Based Y-Shaped Network for Bone Imaging Correction
Medical ultrasonic imaging technique (USI) is widely applied in clinical settings. However, it has not yet been applied for imaging and diagnosis of internal tissues of bones. This study aims to investigate and establish a multi-modal ultrasonic in bone imaging correction (MUBIC) paradigm combining ultrasonic (USI) and photoacoustic (PAI) advantages, and then to develop a USI and PAI bone imaging instrument (UPBII) by training and integrating with an attention-based Y-shaped network (AY-Net) for multi-modal ultrasonic imaging correction. The performance of MUBIC is demonstrated through both simulation and experiments. The results showed that contrast-to-noise ratio (CNR) of the corrected ultrasonic images increased from 3.2 decibels to 7.8 decibels, and the peak signal-to-noise ratio (PSNR) increased from 21.6 decibels to 34.3 decibels; while CNR of the corrected photoacoustic images increased from 2.8 decibels to 7.5 decibels, and PSNR increased from 22.6 decibels to 34.2 decibels. Moreover, the processing time of MUBIC was less than 1.0 second in both USI and PAI modalities. This study developed and applied UPBII to verify the performance of AY-Net, thereby demonstrating that MUBIC is a high-speed and effective paradigm for correcting USI and PAI of bones.
- 28 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75  (00189456) , 1-12
A CNN-Based Measurement Signals Transformation Method for High-Quality Apodized Plane Wave Ultrasound Imaging
Plane wave (PW) transmission enables ultrahigh-frame-rate imaging by insonifying the entire field of view with a full-array excitation. However, because PW transmission lacks transmit focusing, the echo signals suffer from low signal-to-noise ratio, which in turn degrades the overall image quality. In contrast, synthetic transmit aperture (STA) imaging uses single-element transmission, which reduces interference and noise in the measured echoes. Combined with full-aperture reception and dynamic focusing during image reconstruction, STA enables high-quality imaging. However, STA requires as many sequential transmissions as the number of probe elements, resulting in a significantly reduced frame rate due to the prolonged acquisition time. In this context, we propose designing a convolutional neural network (CNN) to map the apodized PW measurement signals to their STA-equivalent counterparts, aiming to preserve the high frame rate of PW imaging while achieving image quality comparable to that of STA. Specifically, based on the superposition principle of linear acoustics, we first establish a mathematical measurement relationship between PW and STA signals. This theoretical model then guides the design of a CNN for measurement signal transformation. Finally, high-resolution B-mode ultrasound images are reconstructed using the delay-and-sum beamforming algorithm. We trained the network via self-supervised learning on a simulated dataset generated by Field II and assessed its performance through both in-vitro and in-vivo experiments. The results demonstrate the feasibility and effectiveness of using a CNN to generate STA measurement signals from apodized PW measurement signals, thereby advancing ultrasound imaging by combining high frame rates with improved image quality. Notably, the proposed method exhibits excellent performance in both lateral and axial (longitudinal) resolution. Overall, this study presents a promising approach for achieving high spatiotemporal resolution in ultrasound imaging based on apodized PW acquisition.
- 28 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75  (00189456) , 1-12
A CNN-Based Measurement Signals Transformation Method for High-Quality Apodized Plane Wave Ultrasound Imaging
Plane wave (PW) transmission enables ultrahigh-frame-rate imaging by insonifying the entire field of view with a full-array excitation. However, because PW transmission lacks transmit focusing, the echo signals suffer from low signal-to-noise ratio, which in turn degrades the overall image quality. In contrast, synthetic transmit aperture (STA) imaging uses single-element transmission, which reduces interference and noise in the measured echoes. Combined with full-aperture reception and dynamic focusing during image reconstruction, STA enables high-quality imaging. However, STA requires as many sequential transmissions as the number of probe elements, resulting in a significantly reduced frame rate due to the prolonged acquisition time. In this context, we propose designing a convolutional neural network (CNN) to map the apodized PW measurement signals to their STA-equivalent counterparts, aiming to preserve the high frame rate of PW imaging while achieving image quality comparable to that of STA. Specifically, based on the superposition principle of linear acoustics, we first establish a mathematical measurement relationship between PW and STA signals. This theoretical model then guides the design of a CNN for measurement signal transformation. Finally, high-resolution B-mode ultrasound images are reconstructed using the delay-and-sum beamforming algorithm. We trained the network via self-supervised learning on a simulated dataset generated by Field II and assessed its performance through both in-vitro and in-vivo experiments. The results demonstrate the feasibility and effectiveness of using a CNN to generate STA measurement signals from apodized PW measurement signals, thereby advancing ultrasound imaging by combining high frame rates with improved image quality. Notably, the proposed method exhibits excellent performance in both lateral and axial (longitudinal) resolution. Overall, this study presents a promising approach for achieving high spatiotemporal resolution in ultrasound imaging based on apodized PW acquisition.
- 15 August 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 106, 103762
Pixel-responsive optimization beamforming method for ultrasound transcranial imaging
- 15 August 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 106, 103762
Pixel-responsive optimization beamforming method for ultrasound transcranial imaging
- Open Access
- 1 December 2024
- Published by AIP Publishing in AIP Advances
- Vol. 14  (12)
A deep learning-based method of acoustic holographic lens generation for transcranial focused ultrasound
Acoustic holographic lenses provide the potential for transcranial focusing because they enable one to accurately and economically overcome distortion caused by the skull on ultrasonic waves. However, challenges remain in the design of acoustic holographic lenses for transcranial focusing. The time inversion method, which is the standard method for generating acoustic holographic lenses for transcranial focusing, is laborious. To overcome this limitation, we propose a U-Net-based transcranial focusing method that can effectively produce acoustic holographic lenses. The simulation results demonstrate that compared to traditional time-reversal methods, the proposed U-net-based approach can produce acoustic holographic lenses for transcranial focusing hundreds of times faster with comparable reconstruction quality. The experimental results show that the performance of the acoustic holograms produced by this methodology is comparable to that of the holograms produced by the traditional time-reversal method. However, the holograms are generated at a pace that is faster by a factor of more than 4000. The results demonstrate that the technique can quickly and accurately produce acoustic holographic lenses for transcranial focusing, opening the door to potential real-time transcranial focusing applications based on these lenses.
- 1 December 2024
- Published by AIP Publishing in AIP Advances
- Vol. 14  (12)
Open AccessA deep learning-based method of acoustic holographic lens generation for transcranial focused ultrasound
Acoustic holographic lenses provide the potential for transcranial focusing because they enable one to accurately and economically overcome distortion caused by the skull on ultrasonic waves. However, challenges remain in the design of acoustic holographic lenses for transcranial focusing. The time inversion method, which is the standard method for generating acoustic holographic lenses for transcranial focusing, is laborious. To overcome this limitation, we propose a U-Net-based transcranial focusing method that can effectively produce acoustic holographic lenses. The simulation results demonstrate that compared to traditional time-reversal methods, the proposed U-net-based approach can produce acoustic holographic lenses for transcranial focusing hundreds of times faster with comparable reconstruction quality. The experimental results show that the performance of the acoustic holograms produced by this methodology is comparable to that of the holograms produced by the traditional time-reversal method. However, the holograms are generated at a pace that is faster by a factor of more than 4000. The results demonstrate that the technique can quickly and accurately produce acoustic holographic lenses for transcranial focusing, opening the door to potential real-time transcranial focusing applications based on these lenses.
- 4 November 2024
- Published by AIP Publishing in Applied Physics Letters
- Vol. 125  (19)
Acoustic holographic lenses for transcranial focusing in an ex vivo human skull
The skull has long been an obstacle for transcranial ultrasound therapy and neuromodulation, leading to waveform distortion and focal points misalignment. In this work, we propose a dual-focus transcranial ultrasound focusing method for ex vivo human skulls to bridge this barrier, which is demonstrated by using a 2 MHz single-element transducer placed outside the occipital bone, coupled to a 3D-printed acoustic holographic lens. The experimental results show that the proposed method allows for achieving high-precision single- and dual-focus ultrasound focusing through human skulls with the maximum target registration error less than one wavelength in water. The in-target ratio spans a range from 72% to 92%, with an average of 80.7%, while the out-of-target ratio varies from 0% to 43.9%, with an average of 18.3%. Additionally, the focal spots are substantially free of the skull-induced distortion, closely matching the ideal ones, and maintain a clearly identifiable circular focus shape. The method opens up a route toward the subsequent applications in clinical research and may enable to implement transcranial ultrasound therapy and neuromodulation.
- 4 November 2024
- Published by AIP Publishing in Applied Physics Letters
- Vol. 125  (19)
Acoustic holographic lenses for transcranial focusing in an ex vivo human skull
The skull has long been an obstacle for transcranial ultrasound therapy and neuromodulation, leading to waveform distortion and focal points misalignment. In this work, we propose a dual-focus transcranial ultrasound focusing method for ex vivo human skulls to bridge this barrier, which is demonstrated by using a 2 MHz single-element transducer placed outside the occipital bone, coupled to a 3D-printed acoustic holographic lens. The experimental results show that the proposed method allows for achieving high-precision single- and dual-focus ultrasound focusing through human skulls with the maximum target registration error less than one wavelength in water. The in-target ratio spans a range from 72% to 92%, with an average of 80.7%, while the out-of-target ratio varies from 0% to 43.9%, with an average of 18.3%. Additionally, the focal spots are substantially free of the skull-induced distortion, closely matching the ideal ones, and maintain a clearly identifiable circular focus shape. The method opens up a route toward the subsequent applications in clinical research and may enable to implement transcranial ultrasound therapy and neuromodulation.
- 1 November 2024
- Published by Optica Publishing Group in Biomedical Optics Express
- Vol. 15  (11) , 6638-6653
Super resolution reconstruction of fluorescence microscopy images by a convolutional network with physical priors
Super-solution fluorescence microscopy, such as single-molecule localization microscopy (SMLM), is effective in observing subcellular structures and achieving excellent enhancement in spatial resolution in contrast to traditional fluorescence microscopy. Recently, deep learning has demonstrated excellent performance in SMLM in solving the trade-offs between spatiotemporal resolution, phototoxicity, and signal intensity. However, most of these researches rely on sufficient and high-quality datasets. Here, we propose a physical priors-based convolutional super-resolution network (PCSR), which incorporates a physical-based loss term and an initial optimization process based on the Wiener filter to create excellent super-resolution images directly using low-resolution images. The experimental results demonstrate that PCSR enables the achievement of a fast reconstruction time of 100 ms and a high spatial resolution of 10 nm by training on a limited dataset, allowing subcellular research with high spatiotemporal resolution, low cell phototoxic illumination, and high accessibility. In addition, the generalizability of PCSR to different live cell structures makes it a practical instrument for diverse cell research.
- 1 November 2024
- Published by Optica Publishing Group in Biomedical Optics Express
- Vol. 15  (11) , 6638-6653
Super resolution reconstruction of fluorescence microscopy images by a convolutional network with physical priors
Super-solution fluorescence microscopy, such as single-molecule localization microscopy (SMLM), is effective in observing subcellular structures and achieving excellent enhancement in spatial resolution in contrast to traditional fluorescence microscopy. Recently, deep learning has demonstrated excellent performance in SMLM in solving the trade-offs between spatiotemporal resolution, phototoxicity, and signal intensity. However, most of these researches rely on sufficient and high-quality datasets. Here, we propose a physical priors-based convolutional super-resolution network (PCSR), which incorporates a physical-based loss term and an initial optimization process based on the Wiener filter to create excellent super-resolution images directly using low-resolution images. The experimental results demonstrate that PCSR enables the achievement of a fast reconstruction time of 100 ms and a high spatial resolution of 10 nm by training on a limited dataset, allowing subcellular research with high spatiotemporal resolution, low cell phototoxic illumination, and high accessibility. In addition, the generalizability of PCSR to different live cell structures makes it a practical instrument for diverse cell research.
- 15 July 2024
- Published by Elsevier in Ultrasound in Medicine & Biology
- Vol. 50  (10) , 1530-1543
Removing Artifacts in Transcranial Photoacoustic Imaging With Polarized Self-Attention Dense-UNet
- 15 July 2024
- Published by Elsevier in Ultrasound in Medicine & Biology
- Vol. 50  (10) , 1530-1543
Removing Artifacts in Transcranial Photoacoustic Imaging With Polarized Self-Attention Dense-UNet
- 6 July 2024
- Published by Elsevier in Ultrasound in Medicine & Biology
- Vol. 50  (9) , 1459-1471
Reconstructing Cancellous Bone From Down-Sampled Optical-Resolution Photoacoustic Microscopy Images With Deep Learning
- 6 July 2024
- Published by Elsevier in Ultrasound in Medicine & Biology
- Vol. 50  (9) , 1459-1471
Reconstructing Cancellous Bone From Down-Sampled Optical-Resolution Photoacoustic Microscopy Images With Deep Learning