Conference Proceedings and Presentations

Permanent URI for this collection

Browse

Search Collection

Now showing 1 - 5 of 13753
  • Presentation
    Dual illumination handheld photoacoustic imaging system with light emitting diode and pulsed laser diode
    (International society for optics and photonics (SPIE), 2025-03-20) Periyasamy, Vijitha ; Das, Avishek ; Pramanik, Manojit ; Department of Electrical and Computer Engineering
    Photoacoustic (PA) imaging is made more affordable and safer using light emitting diodes (LED). LEDs of 690 nm and 850 nm were integrated with handheld array-based transducer in the AcousticX, system (CYBERDYNE INC, Tsukuba, Japan). Performance of the system has been demonstrated for various applications such as vascular imaging of human finger and foot, tumor imaging and so on. Even with high pulse repetition rate (⁓4 kHz) illumination, the low energy per pulse of the LED (2.7 μJ/cm2 at 690 nm) limit the use of the system for laboratory or pre-clinical applications. In this work, we use a dual illumination (pulsed laser diode (PLD) and LED) setup, which extends the usability and functionality of the commercial system. The PLD (QD-Q1924-ILO-W, Quantel) which is operated at 813 nm has a max pulse energy of 7.6 mJ (@ 2k pulse repetition rate). The higher pulse energy leads to higher contrast in PA images. The wavelength of PLD is different from the wavelength of the LED of the commercial system. This enables multiwavelength imaging. In addition, the effective frame rate of the imaging system is doubled (8 kHz) when the LED and PLD light illumination pulse is appropriately delayed temporally. The contrast enhancement of LED based system with PLD is demonstrated using a graphite phantom. Imaging of phantoms of indocyanine green and methylene blue is done to demonstrate the capability of multiwavelength imaging. The doubled frame rate was demonstrated for flow imaging. The PA signal strength increases 100 -folds when PLD is used with LED. Future work is to use the PLD integrated commercial linear array system for pre-clinical imaging.
  • Presentation
    Nanosecond-pulsed light emitting diode (LED)-based photoacoustic computed tomography
    (International society for optics and photonics (SPIE), 2025-03-20) Das, Avishek ; Pramanik, Manojit ; Department of Electrical and Computer Engineering
    Light-emitting diodes (LEDs) are emerging as a promising alternative to traditional laser sources in photoacoustic computed tomography (PACT) systems. LEDs offer several advantages including a wide range of available wavelengths, cost-effectiveness, and flexibility in system design due to low form factor. Moreover, LEDs present a lower safety hazard, eliminating the requirement for the regulatory permits usually necessary to operate high-energy class IV lasers or laser diodes. A key feature of LEDs is their variable frequency option, which allows for high contrast imaging through the averaging of multiple signals obtained from pulsed illumination. This study introduces a novel nanosecond pulsed LED array-based PACT system (LED-PACT) powered by a custom-built Nanosecond Pulsed Current Source (NSPCS). The NSPCS drives an array of 37 high-speed IR LEDs (850 nm, 1350 mW/sr), achieving rapid pulsing and simultaneous activation for enhanced illumination and improved image quality. A series of comprehensive tests were conducted using various phantoms to validate the system’s performance. The reconstructed PACT image results using a simple delay and sum (DAS) algorithm demonstrated the system’s capability to produce detailed and accurate representations of the phantoms, highlighting its potential for various biomedical applications. In conclusion, the designed LED-PACT, with its innovative design and superior performance, holds great promise for advancing the field of photoacoustic imaging. Future work will focus on optimizing the system for in vivo studies and exploring its potential for clinical applications.
  • Presentation
    Wireless ultracompact handheld dual-mode ultrasound and photoacoustic imaging
    (International society for optics and photonics (SIPE), 2025-03-20) Periyasamy, Vijitha ; Das, Avishek ; Pramanik, Manojit ; Department of Electrical and Computer Engineering
    Photoacoustic (PA) imaging is a hybrid imaging modality that combines optical contrast and ultrasound (US) resolution. Dual mode ultrasound + photoacoustic imaging system based on food and drug administration (FDA) approved clinical ultrasound platform can expedite clinical translation of photoacoustic imaging. However, developing dual model US+PA imaging system is challenging, as it requires access to raw radio frequency (RF) data from the ultrasound system. On top, making the US+PA system compact and inexpensive is even more challenging. Here, we report for the first time a wireless, ultra-compact US+PA imaging system based on FDA approved clinical ultrasound platform. The developed system can be completely controlled from iOS or android devices. We have used a pulsed laser diode (PLD) for the PA imaging part. The clinical ultrasound platform is from Clarius, which is routinely used in hospitals by the clinician to image breast, musculoskeletal, nerve, thyroid and so on. The high frequency linear array transducer L15 HD3 operates at the frequency of 5–15 MHz. The PLD (QD-Q1924-ILO-W, Quantel) is operated at the wavelength of 813 nm. The transducer and the PLD is synchronized using the trigger from the transducer. We characterize the performance of the system using hair and pencil lead phantoms. The lateral resolution is approximately 300 μm. In future we would integrate the wireless transducer with light emitting diodes (LEDs) which would reduce the weight and the cost of the PA system.
  • Presentation
    PACT image reconstruction: from sinograms to images using neural networks
    (International society for optics and photonics (SPIE), 2025-03-20) Hui, Xie ; Rajendran, Praveenbalaji ; Ling, Tong ; Pramanik, Manojit ; Department of Electrical and Computer Engineering
    Photoacoustic computed tomography (PACT) is an advanced biomedical imaging technology that combines the strengths of optical and acoustic imaging. In PACT, the image reconstruction process involves transforming raw sensor data (sinograms) into clear, informative images, a task made challenging by the requirement for precise inverse transforms and the influence of sensor imperfections and noise. Traditional reconstruction methods rely on complex, stage-wise signal processing chains tailored to specific acquisition strategies and often require expert parameter tuning for optimal results. While deep learning has proven highly effective in addressing imaging challenges, most applications have been limited to fields like computed tomography (CT) and magnetic resonance imaging (MRI), with relatively few adaptations for PACT. In this study, we introduce a novel approach to PAT image reconstruction, reframing it as a data-driven supervised learning task. By leveraging a deep neural network, we map raw sinograms directly to reconstructed images, bypassing the need for traditional handcrafted reconstruction pipelines. Our network, trained on simulated datasets, achieves image reconstruction directly from sinogram data, offering a streamlined, efficient solution. This research represents a notable advancement in applying deep learning to PACT, with the potential to enhance imaging speed and improve reconstructed image quality.
  • Presentation
    Pre-Service Teachers' Perceptions of Adopting Generative AI Tools in Teaching Mathematics: Insights from a TPACK-based Workshop.
    (Association for the Advancement of Computing in Education (AACE), 2025-03-17) Shrestha, Shristi ; Yi, Jiyeong ; School of Education
    This study explores perceptions of pre-service teachers (PSTs) toward integrating Generative AI (GenAI) in mathematics instruction through a TPACK-based professional development (PD) workshop. Findings reveal that while PSTs recognize GenAI’s potential for lesson planning, differentiated instruction, and improving efficiency, concerns persist regarding accuracy, student overreliance, and ethical risks. The workshop fostered critical reflections on GenAI’s pedagogical applications, but PSTs highlighted the need for structured training and institutional guidance. These insights underscore the importance of comprehensive AI literacy and TPACK-driven PD programs to equip future educators with the skills to critically and effectively integrate GenAI in mathematics classrooms.