摘要 :
Is it possible to find deterministic relationships between optical measurements and pathophysiology in an unsupervised manner and based on data alone? Optical property quantification is a rapidly growing biomedical imaging techniq...
展开
Is it possible to find deterministic relationships between optical measurements and pathophysiology in an unsupervised manner and based on data alone? Optical property quantification is a rapidly growing biomedical imaging technique for characterizing biological tissues that shows promise in a range of clinical applications, such as intraoperative breast-conserving surgery margin assessment. However, translating tissue optical properties to clinical pathology information is still a cumbersome problem due to, amongst other things, inter- and intrapatient variability, calibration, and ultimately the nonlinear behavior of light in turbid media. These challenges limit the ability of standard statistical methods to generate a simple model of pathology, requiring more advanced algorithms. We present a data-driven, nonlinear model of breast cancer pathology for real-time margin assessment of resected samples using optical properties derived from spatial frequency domain imaging data. A series of deep neural network models are employed to obtain sets of latent embeddings that relate optical data signatures to the underlying tissue pathology in a tractable manner. These self-explanatory models can translate absorption and scattering properties measured from pathology, while also being able to synthesize new data. The method was tested on a total of 70 resected breast tissue samples containing 137 regions of interest, achieving rapid optical property modeling with errors only limited by current semi-empirical models, allowing for mass sample synthesis and providing a systematic understanding of dataset properties, paving the way for deep automated margin assessment algorithms using structured light imaging or, in principle, any other optical imaging technique seeking modeling. Code is available.
收起
摘要 :
Breast cancer often necessitates surgical interventions such as breast-conserving surgery or mastectomy. In these surgeries, sentinel lymph node (SLN) samples are often excised for histopathological examination to ascertain the pr...
展开
Breast cancer often necessitates surgical interventions such as breast-conserving surgery or mastectomy. In these surgeries, sentinel lymph node (SLN) samples are often excised for histopathological examination to ascertain the presence of cancer metastasis. Despite its importance, traditional hematoxylin and eosin (H&E) staining, considerably prolongs the operation because of its complex processing requirements. Ultraviolet photoacoustic microscopy (UV-PAM) has emerged as a solution for bypassing the necessity of tissue staining or sectioning. However, its clinical application has been hindered by imaging speed. To overcome this challenge, we have developed a fast-scanning, reflection-mode UV-PAM designed for histopathology without staining based on high-sensitivity, wide-vision optical ultrasound detection. A specimen area of around 12 mm
2
can be scanned in 8 min with a lateral resolution of 1.5 μm. To enhance imaging speed, multi-focal PAM was implemented, resulting in a fourfold acceleration. This PAM technique has been utilized in SLN biopsy to differentiate cancerous tissue in breast cancer patients.
收起
摘要 :
This paper describes a fully spike-based neural network for optical flow estimation from dynamic vision sensor data. A low power embedded implementation of the method, which combines the asynchronous time-based image sensor with I...
展开
This paper describes a fully spike-based neural network for optical flow estimation from dynamic vision sensor data. A low power embedded implementation of the method, which combines the asynchronous time-based image sensor with IBM's TrueNorth Neurosynaptic System, is presented. The sensor generates spikes with submillisecond resolution in response to scene illumination changes. These spike are processed by a spiking neural network running on TrueNorth with a 1-ms resolution to accurately determine the order and time difference of spikes from neighbouring pixels, and therefore infer the velocity. The spiking neural network is a variant of the Barlow Levick method for optical flow estimation. The system is evaluated on two recordings for which ground truth motion is available, and achieves an average endpoint error of 11\% at an estimated power budget of under 80 mW for the sensor and computation.
收起
摘要 :
Glaucoma is a progressive eye condition that causes irreversible vision loss due to damage to the optic nerve. Recent developments in deep learning and the accessibility of computing resources have provided tool support for automa...
展开
Glaucoma is a progressive eye condition that causes irreversible vision loss due to damage to the optic nerve. Recent developments in deep learning and the accessibility of computing resources have provided tool support for automated glaucoma diagnosis. Despite deep learning’s advances in disease diagnosis using medical images, generic convolutional neural networks are still not widely used in medical practices due to the limited trustworthiness of these models. Although deep learning-based glaucoma classification has gained popularity in recent years, only a few of them have addressed the explainability and interpretability of the models, which increases confidence in using such applications. This study presents state-of-the-art deep learning techniques to segment and classify fundus images to predict glaucoma conditions and applies visualization techniques to explain the results to ease understandability. Our predictions are based on U-Net with attention mechanisms with ResNet50 for the segmentation process and a modified Inception V3 architecture for the classification. Attention U-Net with modified ResNet50 backbone obtained 99.58% and 98.05% accuracies for optic disc segmentation and optic cup segmentation, respectively for the RIM-ONE dataset. Additionally, we generate heatmaps that highlight the regions that impacted the glaucoma diagnosis using both Gradient-weighted Class Activation Mapping (Grad-CAM) and Grad-CAM++. Our model that classifies the segmented images achieves accuracy, sensitivity, and specificity values of 98.97%, 99.42%, and 95.59%, respectively, with the RIM-ONE dataset. This model can be used as a support tool for automated glaucoma identification using fundus images.
收起
摘要 :
Micro-hyperspectral imaging (MHSI) is an advanced sensor technology that exploits the distinct spectral signatures of samples for characterization, analysis, and identification. However, applying MHSI to the studies of materials a...
展开
Micro-hyperspectral imaging (MHSI) is an advanced sensor technology that exploits the distinct spectral signatures of samples for characterization, analysis, and identification. However, applying MHSI to the studies of materials at microscale and nanoscale remains constrained. Perovskite, an innovative material, has attracted widespread attention due to its exceptional optical properties, and the characterizing utilization of MHSI on microscale and nanoscale perovskite crystals represents a novel perspective for further comprehensive analysis of the optical characteristics of perovskite. In this study, we developed a home-built MHSI sensor system through the integration of a hyperspectral imaging (HSI) sensor and a metallurgical microscope, enabling the acquisition of 3-D data of the microscale perovskite crystals (
$\text {MA}\text {Pb}\text {Br}_{{3}}$
) samples, which including the characterization imaging of the morphological structure of the samples and its high-resolution spectral response values for hundreds of spectral bands. Auxiliary verification methodologies, including white light interference (WLI), energy-dispersive spectroscopy (EDS), and COMSOL simulation, were employed for the cross-analysis of complex interrelationships among compositional ratios, thickness, morphology, and optical properties in perovskite crystals (
$\text {MA}\text {Pb}\text {Br}_{{3}}$
) and to substantiate the data acquired by MHSI. Specifically, the relationship between thickness and elemental ratios of the crystal has been investigated to explain the optical properties of the “fishbone” sample, and morphology features have been studied to verify the optical properties of the “pyramid” sample. The strong correlation between the experimental results and auxiliary verification highlights the immense potential of MHSI sensor technology for the characterization and extensive in-depth optical analysis of micro–nano materials.
收起
摘要 :
Objective:
The key characteristics of light propagation are the average penetration depth, average maximum penetration depth, average maximum lateral spread, and average path length of photons. These parameters depend on tissue o...
展开
Objective:
The key characteristics of light propagation are the average penetration depth, average maximum penetration depth, average maximum lateral spread, and average path length of photons. These parameters depend on tissue optical properties and, thus, on the pathological state of the tissue. Hence, they could provide diagnostic information on tissue integrity. This study investigates these parameters for articular cartilage which has a complex structure.
Methods:
We utilize Monte Carlo simulation to simulate photon trajectories in articular cartilage and estimate the average values of the light propagation parameters (penetration depth, maximum penetration depth, maximum lateral spread, and path length) in the spectral band of 400–1400 nm based on the optical properties of articular cartilage zonal layers and bulk tissue.
Results:
Our findings suggest that photons in the visible band probe a localized small volume of articular cartilage superficial and middle zones, while those in the NIR band penetrate deeper into the tissue and have larger lateral spread. In addition, we demonstrate that a simple model of articular cartilage tissue, based on the optical properties of the bulk tissue, is capable to provide an accurate description of the light-tissue interaction in articular cartilage.
Conclusion:
The results indicate that as the photons in the spectral band of 400–1400 nm can reach the full depth of articular cartilage matrix, they can provide viable information on its pathological state. Therefore, diffuse optical spectroscopy holds significant importance for objectively assessing articular cartilage health.
Significance:
In this study, for the first time, we estimate the light propagation parameters in articular cartilage.
收起
摘要 :
Photoacoustic imaging (PAI) has become a powerful biomedical imaging technique in the last decades, combining optical and ultrasound imaging (USI) principles. Among the various configurations, optical-resolution photoacoustic micr...
展开
Photoacoustic imaging (PAI) has become a powerful biomedical imaging technique in the last decades, combining optical and ultrasound imaging (USI) principles. Among the various configurations, optical-resolution photoacoustic microscopy (OR-PAM) served as a vital tool for high-resolution and high-sensitivity imaging of small animals in vivo. However, the current systems have limited mobility, which hinders preclinical and clinical research in cramped spaces. Here, we propose a compact (
$510\times 660\times1660$
mm3) transportable PAI system with stimulated Raman scattering (SRS) spectroscopy to address this challenge. The proposed system is designed for complete mobility, overcoming space restrictions. We demonstrate the feasibility of the system by imaging in vitro phantoms and in vivo mice. Additionally, we verify its functional imaging capabilities by visualizing oxygen saturation levels in vivo mice ears. Our results indicate that the proposed transportable PAI system maintains fast data acquisition time, superior image quality, and multispectral imaging capabilities. This shows great potential for expanding the applications of OR-PAM to a wider range of research and clinical scenarios.
收起
摘要 :
Photoacoustic (PA) imaging provides optical contrast at relatively large depths within the human body, compared to other optical methods, at ultrasound (US) spatial resolution. By integrating real-time PA and US (PAUS) modalities,...
展开
Photoacoustic (PA) imaging provides optical contrast at relatively large depths within the human body, compared to other optical methods, at ultrasound (US) spatial resolution. By integrating real-time PA and US (PAUS) modalities, PAUS imaging has the potential to become a routine clinical modality bringing the molecular sensitivity of optics to medical US imaging. For applications where the full capabilities of clinical US scanners must be maintained in PAUS, conventional limited view and bandwidth transducers must be used. This approach, however, cannot provide high-quality maps of PA sources, especially vascular structures. Deep learning (DL) using data-driven modeling with minimal human design has been very effective in medical imaging, medical data analysis, and disease diagnosis, and has the potential to overcome many of the technical limitations of current PAUS imaging systems. The primary purpose of this article is to summarize the background and current status of DL applications in PAUS imaging. It also looks beyond current approaches to identify remaining challenges and opportunities for robust translation of PAUS technologies to the clinic.
收起
摘要 :
This study presents a noninvasive device capable of subdiffraction-limited imaging. The research involves the construction of a two-fiber optical tweezer system to manipulate the stable stationary or ordered motion of a biological...
展开
This study presents a noninvasive device capable of subdiffraction-limited imaging. The research involves the construction of a two-fiber optical tweezer system to manipulate the stable stationary or ordered motion of a biological cell lens within an optical trap. By controlling the outgoing optical field at the fiber tip, the device enables large-range scanning imaging of the biological cell lens, expanding the field of view for ultramicroscopic imaging. The experimental results demonstrate that the biological cell lens can achieve magnification of nanostructures with a resolution as low as 100 nm under white light microscopy. Notably, when observing the surface of digital video disk (DVD) disks, the imaging effect of the biological cell lens surpasses direct observation, providing magnification factors of 1.5–2 times at the center of the lens and 1.3–1.5 times at the edges. These findings highlight the superior imaging capabilities of the bio-lens. The device presented in this study offers a new approach and technical means for leveraging biological cell lens imaging applications. It circumvents issues of sample damage and contamination during imaging, while its simple structure and ease of operation make it a promising tool in this field.
收起
摘要 :
An approach for edge-enhanced imaging based on the optical element that generates a curvilinear vortex beam with intensity and phase distributions along arbitrary curves was proposed, which is different from the classic edge-enhan...
展开
An approach for edge-enhanced imaging based on the optical element that generates a curvilinear vortex beam with intensity and phase distributions along arbitrary curves was proposed, which is different from the classic edge-enhanced imaging method with a 4f imaging system. It is demonstrated that the image edge can be enhanced by modulating the optical element that generates the curvilinear vortex beam with a ring or ellipse trajectory. By setting the factor that determines the ring or elliptical trajectory of the curvilinear vortex beam, the intensity distribution of the optical element can be controlled, resulting in directional enhancement of the image. Furthermore, by clipping the optical element, the image edge can be enhanced in the selective region. It is helpful for the further applications of the optical element that generates the curvilinear vortex beams in optical imaging, especially higher contrast and resolution images.
收起
