Deep learning has revolutionized medical image analysis, promising to significantly improve the precision of diagnoses and therapies through advanced segmentation methods. However, the efficacy of deep neural networks is often compromised by the prevalence of imperfect medical labels, while acquiring large-scale, accurately labeled data remains a prohibitive challenge. To address the imperfect label issue, we introduce a novel learning framework that iteratively optimizes both a neural network and its label set to enhance segmentation accuracy. This framework operates through two steps: initially, it robustly trains on a dataset with label noise, distinguishing between clean and noisy labels, and subsequently, it refines noisy labels based on high-confidence predictions from the robust network. By applying this method, not only is the network trained more effectively on imperfect data, but the dataset is progressively cleaned and expanded. Our evaluations are conducted on retina Optical Coherence Tomography datasets using U-Net and SegNet architectures, and demonstrate substantial improvements in segmentation accuracy and data quality, advancing the capabilities of weakly supervised segmentation in medical imaging.
Near-infrared reflectance spectroscopy can serve as a complementary imaging tool that accurately provides endocardial substrates through optical mapping and improves the quality of ablation therapy. Optical indices were extracted from spectrum response, visualized similarity of blood and PBS maps interpolated from those indices were evaluated. Statistical analysis between blood and PBS optical indices were performed for each substrate type, and classification algorithms were developed using key features to classify pulmonary vein, lesion, and fibrosis with high accuracy. The results indicate NIRS mapping catheters can serve as a complementary imaging tool to the current EAM systems to improve treatment efficacy.
Atrial fibrillation is a common and potentially lethal arrhythmia, yet catheter radiofrequency ablation (RFA), a mainstay of treatment, frequently fails to provide long-term remission. We present a catheter capable of near-infrared diffuse reflectance spectroscopy, with a source fiber delivering broadband light and a detection fiber whose light is sent to a spectrometer. Separate catheters have been fabricated with different source-detection separations, yielding spectra sensitive to different optical properties of the underlying tissue. Optical indices have been developed from benchtop measurements to distinguish the spectral signatures of different cardiac substrates. These measurements will equip clinicians with intraprocedural feedback to improve RFA effectiveness.
SignificanceRadiofrequency ablation (RFA) procedures for atrial fibrillation frequently fail to prevent recurrence, partially due to limitations in assessing extent of ablation. Optical spectroscopy shows promise in assessing RFA lesion formation but has not been validated in conditions resembling those in vivo.AimCatheter-based near-infrared spectroscopy (NIRS) was applied to porcine hearts to demonstrate that spectrally derived optical indices remain accurate in blood and at oblique incidence angles.ApproachPorcine left atria were ablated and mapped using a custom-fabricated NIRS catheter. Each atrium was mapped first in phosphate-buffered saline (PBS) then in porcine blood.ResultsNIRS measurements showed little angle dependence up to 60 deg. A trained random forest model predicted lesions with a sensitivity of 81.7%, a specificity of 86.1%, and a receiver operating characteristic curve area of 0.921. Predicted lesion maps achieved a mean structural similarity index of 0.749 and a mean normalized inner product of 0.867 when comparing maps obtained in PBS and blood.ConclusionsCatheter-based NIRS can precisely detect RFA lesions on left atria submerged in blood. Optical parameters are reliable in blood and without perpendicular contact, confirming their ability to provide useful feedback during in vivo RFA procedures.
In this project, we propose a deep learning based weakly supervised learning algorithm for cardiac adipose tissue segmentation using image-level labels. Based on ReLayNet, our proposed method can automatically segment the adipose tissue from normal myocardium tissue in pixel level. Compared with fully supervised learning methods, our model achieves competitive segmentation results on both accuracy and Dice coefficient within a database of OCT images of human cardiac tissue. Combined with the OCT image, the predicted adipose map could provide additional information for the guidance of cardiac radio frequency ablation.
The current study investigates the beneficial combination of optical coherence tomography (OCT) and photoacoustic microscopy (PAM) as a safe method for observing retinal and choroidal vasculature. A recent addition to the field has been the integration of gold nanoparticles (AuNPs) to provide enhanced contrast in OCT and PAM images. The improved analysis of capillaries is the result of the strong optical scattering and optical absorption of gold nanoparticles due to surface plasmon resonance. Femtosecond laser ablation created the ultra-pure colloidal gold nanoparticles, which were then capped with polyethylene glycol (PEG). The AuNPs were administered to thirteen New Zealand rabbits to determine the advantages of this technology, while also investigating the safety and biocompatibility. The study determines that the synthesized PEG-AuNPs (20.0 ± 1.5 nm) were beneficial in enhancing contrast in PAM and OCT images without demonstrating cytotoxic effects to bovine retinal endothelial cells. In living rabbits, the administered PEG-AuNPs resulted in an 82% increased signal for PAM and a 45% increased signal for OCT in the retinal and choroidal vessels. A histology and biodistribution report determined that the AuNPs had mostly accumulated in the liver and spleen. TUNEL staining and histology established that no cell injury or death in the lung, liver, kidney, spleen, heart, or eyes had occurred up to 1 week after receiving a dose of AuNP. The nanoparticle technology, therefore, provides an effective and safe method to enhance contrast in ocular imaging, resulting in improved visualization of retinal microvasculature.
Current clinical available retinal imaging techniques have limitations, including limited depth of penetration or requirement for the invasive injection of exogenous contrast agents. Here, we developed a novel multimodal imaging system for high-speed, high-resolution retinal imaging of larger animals, such as rabbits. The system integrates three state-of-the-art imaging modalities, including photoacoustic microscopy (PAM), optical coherence tomography (OCT), and fluorescence microscopy (FM). In vivo experimental results of rabbit eyes show that the PAM is able to visualize laser-induced retinal burns and distinguish individual eye blood vessels using a laser exposure dose of ~80 nJ, which is well below the American National Standards Institute (ANSI) safety limit 160 nJ. The OCT can discern different retinal layers and visualize laser burns and choroidal detachments. The novel multi-modal imaging platform holds great promise in ophthalmic imaging.
Most reported photoacoustic ocular imaging work to date uses small animals, such as mice and rats, the eyes of which are small and less than one-third the size of a human eye, which poses a challenge for clinical translation. Here we achieved chorioretinal imaging of larger animals, i.e. rabbits, using a dual-modality photoacoustic microscopy (PAM) and optical coherence tomography (OCT) system. Preliminary experimental results in living rabbits demonstrate that the PAM can noninvasively visualize depth-resolved retinal and choroidal vessels using a safe laser exposure dose; and the OCT can finely distinguish different retinal layers, the choroid, and the sclera. This reported work might be a major step forward in clinical translation of photoacoustic microscopy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.