arXiv subjects

M. Imran Cheema

Publications and source records attributed to M. Imran Cheema.

12 recordsLinked to original sources

arXiv2025

Respiration Rate Sensor Based on Fiber Cavity Attenuated Phase Shift Spectroscopy

Respiratory rate (RR) is a vital sign with significant diagnostic value. Existing RR monitors often suffer from baseline drift over time, breaths can be occluded by limb or body movements, and many systems struggle to resolve shallow or extreme thoracic motion. Here, we propose a novel RR-monitoring sensor based on fiber-cavity attenuated phase-shift spectroscopy~(CAPS). The sensor comprises a fiber cavity whose portion is embedded into a flexible chest binder in a sinusoidal-like pattern, which the patient wears. Thoracic expansion and contraction during breathing modulate the cavity, and RR is extracted through CAPS measurements. Our sensor exhibits high reproducibility, strong sensitivity to strain and pressure induced by chest movements, inherent resistance to baseline drift, and the ability to detect body movements. The system achieves a root-mean-square error of 0.91 breaths per minute relative to the ground truth, across RR values ranging from 8 to 44 breaths per minute, when tested on multiple subjects in various postures, including sitting, supine, prone, and lateral positions. We anticipate that this work will contribute to the development of comprehensive optical fiber-based sleep monitoring systems.

physics.optics↗

arXiv2025

HistoSpeckle-Net: Mutual Information-Guided Deep Learning for high-fidelity reconstruction of complex OrganAMNIST images via perturbed Multimode Fibers

Existing deep learning methods in multimode fiber (MMF) imaging often focus on simpler datasets, limiting their applicability to complex, real-world imaging tasks. These models are typically data-intensive, a challenge that becomes more pronounced when dealing with diverse and complex images. In this work, we propose HistoSpeckle-Net, a deep learning architecture designed to reconstruct structurally rich medical images from MMF speckles. To build a clinically relevant dataset, we develop an optical setup that couples laser light through a spatial light modulator (SLM) into an MMF, capturing output speckle patterns corresponding to input OrganAMNIST images. Unlike previous MMF imaging approaches, which have not considered the underlying statistics of speckles and reconstructed images, we introduce a distribution-aware learning strategy. We employ a histogram-based mutual information loss to enhance model robustness and reduce reliance on large datasets. Our model includes a histogram computation unit that estimates smooth marginal and joint histograms for calculating mutual information loss. It also incorporates a unique Three-Scale Feature Refinement Module, which leads to multiscale Structural Similarity Index Measure (SSIM) loss computation. Together, these two loss functions enhance both the structural fidelity and statistical alignment of the reconstructed images. Our experiments on the complex OrganAMNIST dataset demonstrate that HistoSpeckle-Net achieves higher fidelity than baseline models such as U-Net and Pix2Pix. It gives superior performance even with limited training samples and across varying fiber bending conditions. By effectively reconstructing complex anatomical features with reduced data and under fiber perturbations, HistoSpeckle-Net brings MMF imaging closer to practical deployment in real-world clinical environments.

cs.CV↗

arXiv2025

Experimental insights into data augmentation techniques for deep learning-based multimode fiber imaging: limitations and success

Multimode fiber~(MMF) imaging using deep learning has high potential to produce compact, minimally invasive endoscopic systems. Nevertheless, it relies on large, diverse real-world medical data, whose availability is limited by privacy concerns and practical challenges. Although data augmentation has been extensively studied in various other deep learning tasks, it has not been systematically explored for MMF imaging. This work provides the first in-depth experimental and computational study on the efficacy and limitations of augmentation techniques in this field. We demonstrate that standard image transformations and conditional generative adversarial-based synthetic speckle generation fail to improve, or even deteriorate, reconstruction quality, as they neglect the complex modal interference and dispersion that results in speckle formation. To address this, we introduce a physical data augmentation method in which only organ images are digitally transformed, while their corresponding speckles are experimentally acquired via fiber. This approach preserves the physics of light-fiber interaction and enhances the reconstruction structural similarity index measure~(SSIM) by up to 17\%, forming a viable system for reliable MMF imaging under limited data conditions.

physics.optics↗

arXiv2024

Towards optimal multimode fiber imaging by leveraging input polarization and deep learning

Deep learning techniques provide a plausible route towards achieving practical imaging through multimode fibers. The results produced by these methods are often influenced by physical factors like temperature, fiber length, external perturbations, and polarization state of the input light. Literature focuses on these different elements impacting deep-learning-enabled multimode imaging, yet the effects of input polarization remain under-explored. Here, we show experimentally that the state of polarization of light, being injected at multimode fiber input, affects the fidelity of reconstructed images from speckle patterns. Certain polarization states produce high-quality images at fiber output, while some yield degraded results. We have designed a conditional generative adversarial network~(CGAN) for image regeneration at various degrees of input light polarization. At a particular polarization state and with a thinner core multimode fiber, our network can reconstruct images with an average structural similarity index(SSIM) exceeding 0.9. Hence, in the case of multimode fibers that are held fixed, optimal imaging can be achieved by leveraging deep learning models with the input light polarization state, where the fidelity of images is maximum. We also show that the model can be trained to image adequately for all input light polarization states when the fiber has bends or twists. We anticipate that our work will be a stepping stone toward developing high-resolution and less invasive multimode fiber endoscopes.

physics.optics↗

arXiv2023

Lipoarabinomannan-based Tuberculosis Diagnosis using a Fiber Cavity Ring Down Biosensor

Despite existing for millennia, tuberculosis (TB) remains a persistent global health challenge. A significant obstacle in controlling TB spread is the need for a rapid, portable, sensitive, and accurate diagnostic test. Currently, sputum culture stands as a benchmark test for TB diagnosis. Although highly reliable, it necessitates advanced laboratory facilities and involves considerable testing time. In this context, we present a rapid, portable, and cost-effective optical fiber sensor designed to measure lipoarabinomannan (LAM), a TB biomarker found in patients' urine samples. Our sensing approach is based on the applications of phase shift-cavity ringdown spectroscopy (PS-CRDS) to an optical fiber cavity created by two fiber Bragg gratings. A tapered fiber is spliced inside the optical cavity to serve as the sensing head. We functionalize the tapered fiber surface with anti-LAM antigen CS-35 through a unique chemistry, creating a strong affinity for LAM molecules. We measure the phase difference between the cavity transmission and the reference modulating signal at the cavity output. The measured phase is directly proportional to the injected LAM concentrations in aqueous solutions over the sensing head. Our demonstrated sensor provides a detection limit of 10 pg/mL and a sensitivity of 2.6$^\circ$/ng/mL. This sensor holds promise for numerous applications in the healthcare sector, particularly in low-resource settings.

physics.med-ph↗

arXiv2022

Raman Spectroscopy and Machine Learning-based Optical Sensor for Rapid Tuberculosis Diagnosis via Sputum

Tuberculosis (TB) is a contagious disease that causes 1.5 million deaths per year globally. Early diagnosis of TB patients is critical to control its spread. However, standard TB diagnostic tests such as sputum culture take days to weeks to produce results. Here, we demonstrate a quick, portable, easy-to-use, and non-invasive optical sensor based on sputum samples for TB detection. The probe uses Raman spectroscopy to detect TB in a patient's sputum supernatant. We deploy a machine-learning algorithm, principal component analysis (PCA), on the acquired Raman data to enhance the detection sensitivity and specificity. On testing 112 potential TB patients, our results show that the developed probe's accuracy is 100% for true-positive and 93.4% for true-negative. Moreover, the probe correctly identifies patients on TB medication. We anticipate that our work will lead to a viable and rapid TB diagnostic platform.

physics.med-ph↗

arXiv2022

Fabry-Pérot fiber cavity refractive index sensing via linewidth tracking in the broken PT-symmetric region

Parity-time (PT) symmetric optical sensors operating around exceptional points have recently gained much attraction, offering an unparalleled high sensitivity in measuring small perturbations. In the past, most of the PT-symmetric sensors have been based on tracking the mode-splitting that arises due to a perturbation-induced change in coupling strength between two subcavities of the PT-symmetric system. We design a linear fiber Fabry-Pérot and coupled cavities sensor, tailored to operate in the broken PT-symmetric region. We explore a new sensing metric-that is, the mode's linewidth change as a function of perturbation-induced changes in the loss within one of the subcavities of the PT-symmetric system. The coupling strength between the two subcavities remains unchanged in our proposed sensor. Supported by a mathematical formulation, we find that the full-width-half-maximum (FWHM) of the cavity resonances exhibits a square root dependence on the refractive index (RI) change in one of the subcavities. The proposed fiber cavity refractive index sensor has a maximum sensitivity of $2.26\times10^7$ GHz/RIU and a lowest detection limit of $10^{-9}$ RIU, widely outperforming the comparable cavity sensors subject to the same refractive index change, gain, and loss settings.

physics.optics↗

arXiv2021

Phase shift-cavity ring down spectroscopy in linear and active fiber cavities for sensing applications at 1550 nm

Liquid phase sensing applications at 1550~nm are highly desirable due to widely available off-the-shelf components. Generally, liquids at 1550~nm induce a high absorption loss that limits the overall sensor's sensitivity and detection limit. One solution is to use an active fiber loop in conjunction with cavity ring down spectroscopy to overcome these absorption losses. However, the amplifier inside the fiber loop suffers from inherent gain fluctuations that limit the sensing system's overall performance. Here, we provide a novel sensor using the wavelength-scanned phase shift-cavity ring down spectroscopy (PS-CRDS) in conjunction with a linear active fiber cavity that potentially offers a more sensitive solution than traditional fiber loop sensors. We use a tapered fiber as a sensing head inside the active cavity built from fiber Bragg gratings. We derive a theoretical phase shift expression for our system and simulate it using the finite element method to determine optimum tapered fiber diameter for glucose sensing in DI water. Compared to a non-amplified system, we find that our amplified system can increase the sensitivity by fourteen times via the amplifier gain tuning. We also conduct experimental measurements using 0-15.5~mM glucose solutions and find them in excellent agreement with our theoretical predictions. Experimentally we obtain the sensor's sensitivity of 0.768~$^o$/mM (1164~$^o$/RIU) and detection limit of 0.75~mM ( 4.5~$\times$~10$^{-4}$~RIU) without any temperature stabilization in the system. We anticipate that the present work will find a wide range of sensing applications in fiber cavities, ring resonators, and other microcavity structures.

physics.optics↗

arXiv2020

Detection of Aflatoxin M1 by Fiber Cavity Attenuated Phase Shift Spectroscopy

Aflatoxin M1 (AFM1) is a carcinogenic compound commonly found in milk in excess of the WHO permissible limit, especially in developing countries. Currently, state-of-the-art tests for detecting AFM1 in milk include chromatographic systems and enzyme-linked-immunosorbent assays. Although these tests provide fair accuracy and sensitivity however, they require trained laboratory personnel, expensive infrastructure, and many hours for producing final results. Optical sensors leveraging spectroscopy have a tremendous potential of providing an accurate, real time, and specialists-free AFM1 detector. Despite this, AFM1 sensing demonstrations using optical spectroscopy are still immature. Here, we demonstrate an optical sensor that employs the principle of cavity attenuated phase shift spectroscopy in optical fiber cavities for rapid AFM1 detection in aqueous solutions at 1550 nm. The sensor constitutes a cavity built by two fiber Bragg gratings. We splice a tapered fiber of $<$ 10 $μ$m waist inside the cavity as a sensing head. For ensuring specific binding of AFM1 in a solution, the tapered fiber is functionalized with DNA aptamers followed by validation of the conjugation via FTIR, TGA, and EDX analyses. We then detect AFM1 in a solution by measuring the phase shift between a sinusoidally modulated laser input and the sensor output at resonant frequencies of the cavity. Our results show that the sensor has the detection limit of 20 ng/L (20 ppt) which is well below both the US and the European safety regulations. We anticipate that the present work will lead towards a rapid and accurate AFM1 sensor, especially for low-resource settings.

physics.med-ph↗

arXiv2019

Digital synthesis of multistage etalons for enhancing the FSR

Fabry-Perot fiber etalons (FPE) built from three or more reflectors are attractive for a variety of applications including communications and sensing. For accelerating a research and development work, one often desires to use off-the-shelf components to build an FPE with a required transmission profile for a particular application. Usually, multistage FPEs are designed with equal lengths of cavities followed by determination of the required reflectivities for realizing a desired transmission profile. As seen in previous works, fabricated reflectors are usually slightly different from the designed ones leading to departure from the desired transmission profile of the FPE. Here, we show a novel digital synthesis of multistage etalons with off-the-shelf reflectors and unequal lengths of involved cavities. We find that, in contrast to equal cavity lengths, unequal lengths of cavities provide more number of poles in the $z$-domain to achieve a desired multicavity FPE transmission response. For given reflectivities and by determining correct unequal lengths of cavities with our synthesis technique, we demonstrate a design example of increasing the FSR followed by its experimental validation. This work is generalizable to ring resonators, mirrored, and fiber Bragg grating based cavities; enabling the design and optimization of cavity systems for a wide range of applications including lasers, sensors, and filters.

eess.SP↗

arXiv2013

Accurate determination of the quality factor and tunneling distance of axisymmetric resonators for biosensing applications

Due to ultra high quality factor ($10^6-10^9$), axisymmetric optical microcavities are popular platforms for biosensing applications. It has been recently demonstrated that a microcavity biosensor can track a biodetection event as a function of its quality factor by using phase shift cavity ring down spectroscopy (PS-CRDS). However, to achieve maximum sensitivity, it is necessary to optimize the microcavity parameters for a given sensing application. Here, we introduce an improved finite element model which allows us to determine the optimized geometry for the PS-CRDS sensor. The improved model not only provides fast and accurate determination of quality factors but also determines the tunneling distance of axisymmetric resonators. The improved model is validated numerically, analytically, and experimentally.

physics.optics↗

arXiv2012

Simultaneous measurement of quality factor and wavelength shift by phase shift microcavity ring down spectroscopy

Optical resonant microcavities with ultra high quality factors are widely used for biosensing. Until now, the primary method of detection has been based upon tracking the resonant wavelength shift as a function of biodetection events. One of the sources of noise in all resonant-wavelength shift measurements is the noise due to intensity fluctuations of the laser source. An alternative approach is to track the change in the quality factor of the optical cavity by using phase shift cavity ring down spectroscopy, a technique which is insensitive to the intensity fluctuations of the laser source. Here, using biotinylated microtoroid resonant cavities, we show simultaneous measurement of the quality factor and the wavelength shift by using phase shift cavity ring down spectroscopy. These measurements were performed for disassociation phase of biotin-streptavidin reaction. We found that the disassociation curves are in good agreement with the previously published results. Hence, we demonstrate not only the application of phase shift cavity ring down spectroscopy to microcavities in the liquid phase but also simultaneous measurement of the quality factor and the wavelength shift for the microcavity biosensors in the application of kinetics measurements.

physics.optics↗