Roadmap on computational methods in optical imaging and holography [invited]

Applied Physics B, Год журнала: 2024, Номер 130(9)

Опубликована: Авг. 29, 2024

Computational methods have been established as cornerstones in optical imaging and holography recent years. Every year, the dependence of on computational is increasing significantly to extent that components are being completely efficiently replaced with at low cost. This roadmap reviews current scenario four major areas namely incoherent digital holography, quantitative phase imaging, through scattering layers, super-resolution imaging. In addition registering perspectives modern-day architects above research areas, also reports some latest studies topic. codes pseudocodes presented for a plug-and-play fashion readers not only read understand but practice algorithms their data. We believe this will be valuable tool analyzing trends predict prepare future holography.

Язык: Английский

---

Artificial Intelligence (AI) in Breast Imaging: A Scientometric Umbrella Review

Diagnostics, Год журнала: 2022, Номер 12(12), С. 3111 - 3111

Опубликована: Дек. 9, 2022

Artificial intelligence (AI), a rousing advancement disrupting wide spectrum of applications with remarkable betterment, has continued to gain momentum over the past decades. Within breast imaging, AI, especially machine learning and deep learning, honed unlimited cross-data/case referencing, found great utility encompassing four facets: screening detection, diagnosis, disease monitoring, data management as whole. Over years, cancer been apex cumulative risk ranking for women across six continents, existing in variegated forms offering complicated context medical decisions. Realizing ever-increasing demand quality healthcare, contemporary AI envisioned make strides clinical perception, capability detect indeterminate significance, predict prognostication, correlate available into meaningful endpoint. Here, authors captured review works decades, focusing on systematized included one usable document, which is termed an umbrella review. The present study aims provide panoramic view how poised enhance imaging procedures. Evidence-based scientometric analysis was performed accordance Preferred Reporting Items Systematic reviews Meta-Analyses (PRISMA) guideline, resulting 71 works. This synthesize, collate, works, thereby identifying patterns, trends, quality, types by structured search strategy. intended serve “one-stop center” synthesis holistic bird’s eye readers, ranging from newcomers researchers relevant stakeholders, topic interest.

Язык: Английский

---

A Review of Image Reconstruction Algorithms for Diffuse Optical Tomography

Applied Sciences, Год журнала: 2023, Номер 13(8), С. 5016 - 5016

Опубликована: Апрель 17, 2023

Diffuse optical tomography (DOT) is a biomedical imaging modality that can reconstruct hemoglobin concentration and associated oxygen saturation by using detected light passing through biological medium. Various clinical applications of DOT such as the diagnosis breast cancer functional brain are expected. However, it has been difficult to obtain high spatial resolution quantification accuracy with because diffusive propagation in tissues strong scattering absorption. In recent years, various image reconstruction algorithms have proposed overcome these technical problems. Moreover, progress related technologies, artificial intelligence supercomputers, circumstances surrounding changed. To support clinics new entries technologies DOT, we review efforts from viewpoint (i) forward calculation process, including radiative transfer equation its approximations simulate precision, (ii) optimization use sparsity regularization prior information improve quantification.

Язык: Английский

---

Vortex Beams and Deep Learning for Optical Wireless Communication Through Turbulent and Diffuse Media

Journal of Lightwave Technology, Год журнала: 2024, Номер 42(10), С. 3631 - 3641

Опубликована: Фев. 5, 2024

Interest in optical wireless communications (OWC) as a possible complement to RF technology has increased recently. However, the propagation of beams through atmosphere distorts beam's amplitude and phase, resulting information loss significant noise. Arising due combination multiple absorption scattering events, distortion beam makes communication difficult. In some cases, orbital angular momentum light (OAM), along with various deep learning algorithms (DL), could be helpful mitigate problem provide high-capacity links. this work, we propagate Laguerre-Gaussian (LG) different topological charges (l) under diffuse turbulent conditions develop deep-learning classification network characterize 'l' LG beam. The proposed method is later implemented using laboratory setup demonstrating on table. results show that algorithm can identify modes high accuracy even when propagates highly media. To demonstrate robustness OWC system, small grayscale images are transmitted over channel. A bit error rate (BER) only 2.3 × 10 ^-4 9.7 for tabletop experiment simulations, respectively. demonstrated low BER system suggests promising applications secure reliable data transmission adverse atmospheric conditions, highlighting potential advancing technologies.

Язык: Английский

---

Challenges and advances in optical 3D mesoscale imaging

Journal of Microscopy, Год журнала: 2022, Номер 286(3), С. 201 - 219

Опубликована: Май 5, 2022

Abstract Optical mesoscale imaging is a rapidly developing field that allows the visualisation of larger samples than possible with standard light microscopy, and fills gap between cell organism resolution. It spans from advanced fluorescence micrometric clusters to centimetre‐size complete organisms. However, volume specimens, new problems arise. Imaging deeper into tissues at high resolution poses challenges ranging optical distortions shadowing opaque structures. This manuscript discusses latest developments in highlights limitations, namely labelling, clearing, absorption, scattering, also sample handling. We then focus on approaches seek turn more quantitative technique, analogous tomography medical imaging, highlighting future role for digital physical phantoms as well artificial intelligence.

Язык: Английский

---

High-fidelity diffuse optical tomography imaging based on MRI physics information-constrained stacked autoencoder neural network

Optics Communications, Год журнала: 2025, Номер unknown, С. 131753 - 131753

Опубликована: Март 1, 2025

Язык: Английский

---

Propagation of Laguerre-Gaussian beam intensities through optically thick turbid media

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Июнь 5, 2025

Язык: Английский

---

Regression-based neural network for improving image reconstruction in diffuse optical tomography

Biomedical Optics Express, Год журнала: 2022, Номер 13(4), С. 2006 - 2006

Опубликована: Фев. 10, 2022

Diffuse optical tomography (DOT) is a non-invasive imaging technique utilizing multi-scattered light at visible and infrared wavelengths to detect anomalies in tissues. However, the DOT image reconstruction based on solving inverse problem, which requires massive calculations time. In this article, for first time, best of our knowledge, simple, regression-based cascaded feed-forward deep learning neural network derived solve problem compressed breast geometry. The predicted data subsequently utilized visualize tissues their anomalies. dataset study created using Monte-Carlo algorithm, simulates propagation placed inside parallel plate source-detector geometry (forward process). simulated DL-DOT system's performance evaluated Pearson correlation coefficient (R) Mean squared error (MSE) metrics. Although comparatively smaller (50 nos.) used, simulation results show that developed algorithm delivers an increment ∼30% over analytical solution approach, terms R. Furthermore, proposed network's MSE outperforms solution's by large margin revealing robustness adaptability system potential applications medical settings.

Язык: Английский

---

Unrolled-DOT: an interpretable deep network for diffuse optical tomography

Journal of Biomedical Optics, Год журнала: 2023, Номер 28(03)

Опубликована: Март 8, 2023

SignificanceImaging through scattering media is critical in many biomedical imaging applications, such as breast tumor detection and functional neuroimaging. Time-of-flight diffuse optical tomography (ToF-DOT) one of the most promising methods for high-resolution media. ToF-DOT traditional DOT require an image reconstruction algorithm. Unfortunately, this algorithm often requires long computational runtimes may produce lower quality reconstructions presence model mismatch or improper hyperparameter tuning.AimWe used a data-driven unrolled network our inverse solver. The faster than solvers achieves higher by accounting mismatch.ApproachOur "Unrolled-DOT" uses learned iterative shrinkage thresholding In addition, we incorporate refinement U-Net Visual Geometry Group (VGG) perceptual loss to further increase quality. We trained tested on simulated real-world data benchmarked against physics-based learning-based solvers.ResultsIn experiments data, Unrolled-DOT outperformed algorithms achieved over 10× reduction runtime mean-squared error, compared solvers.ConclusionWe demonstrated solver that state-of-the-art performance speed quality, which can aid future applications noninvasive imaging.

Язык: Английский

---

Early Detection of Breast Cancer using Diffuse Optical Probe and Ensemble Learning Method

Опубликована: Июнь 28, 2023

In this paper, we propose using the diffuse optical breast scanning (DOB-Scan) probe, which employs an ensemble learning method to enable earlier detection of cancer. For this, utilized nine models with various regression algorithms as base estimators predict properties for liquid breast-mimicking phantoms. These included Polynomial Regression, Support Vector, Random Forest, K-Nearest Neighbors, Decision Tree, Multi-layer Perceptron, XGBoost, CatBoost, and Extra Trees Regressors. We evaluated performance our based on accuracy, precision, recall, F1-score, Matthews Correlation Coefficient (MCC). Our analysis revealed that model had highest accuracy 93%, making it best model. Additionally, Bagging KNN achieved 100% in classifying into healthy unhealthy categories. results suggest DOB-Scan utilizing approach, has potential detect cancer at stage.

Язык: Английский

---