arXiv subjects

Mayank Goswami

Publications and source records attributed to Mayank Goswami.

At least 37 records · Page 2Linked to original sources

arXiv2023

Characterization of Electrical Impedance Tomography System

Electrical Impedance Tomography can be cost-effective, portable, non-invasive imaging technique. It has preclinical and a few of them already proven industrial applications. This technique can only recover images of low spatial and contrast resolution, partially due to existing physical models. The capability of discriminating between Impedance profiles in recovery is around 73 percentage. However, similar to other modalities, EITs performance depends on the hardware and recovery algorithm design and operating parameters. This work presents an empirically obtained mutual relation between the hardware design related six independent variables, namely (a) molarity of the coupling media, (b) scanning duration, (c) Size of the object, and (d) parameters defining the size of the scanning assembly (No of electrodes, area of the vessel, and percentage periphery covered by the electrodes), affecting its performance. The expression predicts that the error can be kept under a 10 percentage value in a worst-case scenario if these six parameters are kept under a given range. The root mean square error between the experiment values and predicted values from the presented equation is 1.4053. It is shown that time significantly affects the recovery process compared to other optimizing parameters. It is also shown that the accepted molarity value for the presented system is 2M.

physics.ins-det↗

arXiv2023

Relative estimation of scattering noise and its utility to select radiation detector for Gamma CT scanner

This study investigates two unavoidable noise factors: electronic noise and radiation scattering, associated with detectors and their electronics. This study proposes a novel methodology to estimate electronic and scattering noise separately. It utilizes mathematical tools, namely, Kanpur theorem-1, standard deviation, similarity dice coefficient parameters, and experimental Computerized Tomography technique. Four types of gamma detectors: CsI (Tl), LaBr_3 (Ce), NaI (Tl) and HPGe are used with their respective electronics. A detector having integrated circuit electronics is shown to impart significantly less (~33% less) electronic noise in data as compared to detectors with distributed electronics. Kanpur Theorem-1 signature is proposed as a scattering error estimate. An empirical expression is developed showing that scattering noise depends strongly on mass attenuation coefficients of detector crystal material and weakly on their active area. The difference between predicted and estimated relative scattering is 14.6%. The methodology presented in this study will assist the related industry in selecting the appropriate detector of optimal diameter, thickness, material composition, and hardware as per requirement.

physics.ins-det↗

arXiv2023

AI pipeline for accurate retinal layer segmentation using OCT 3D images

Image data set from a multi-spectral animal imaging system is used to address two issues: (a) registering the oscillation in optical coherence tomography (OCT) images due to mouse eye movement and (b) suppressing the shadow region under the thick vessels/structures. Several classical and AI-based algorithms in combination are tested for each task to see their compatibility with data from the combined animal imaging system. Hybridization of AI with optical flow followed by Homography transformation is shown to be working (correlation value>0.7) for registration. Resnet50 backbone is shown to be working better than the famous U-net model for shadow region detection with a loss value of 0.9. A simple-to-implement analytical equation is shown to be working for brightness manipulation with a 1% increment in mean pixel values and a 77% decrease in the number of zeros. The proposed equation allows formulating a constraint optimization problem using a controlling factor α for minimization of number of zeros, standard deviation of pixel value and maximizing the mean pixel value. For Layer segmentation, the standard U-net model is used. The AI-Pipeline consists of CNN, Optical flow, RCNN, pixel manipulation model, and U-net models in sequence. The thickness estimation process has a 6% error as compared to manual annotated standard data.

eess.IV↗

arXiv2022

On Flipping the Fréchet distance

The classical and extensively-studied Fréchet distance between two curves is defined as an inf max, where the infimum is over all traversals of the curves, and the maximum is over all concurrent positions of the two agents. In this article we investigate a "flipped" Fréchet measure defined by a sup min -- the supremum is over all traversals of the curves, and the minimum is over all concurrent positions of the two agents. This measure produces a notion of "social distance" between two curves (or general domains), where agents traverse curves while trying to stay as far apart as possible. We first study the flipped Fréchet measure between two polygonal curves in one and two dimensions, providing conditional lower bounds and matching algorithms. We then consider this measure on polygons, where it denotes the minimum distance that two agents can maintain while restricted to travel in or on the boundary of the same polygon. We investigate several variants of the problem in this setting, for some of which we provide linear time algorithms. Finally, we consider this measure on graphs. We draw connections between our proposed flipped Fréchet measure and existing related work in computational geometry, hoping that our new measure may spawn investigations akin to those performed for the Fréchet distance, and into further interesting problems that arise.

cs.CG↗

arXiv2022

Do we really need to recalibrate the CT system Configuration for every experiment?

Different technical and physical factors may affect the image quality reconstructed using a Computed Tomography system. We have developed and designed a 2D Gamma Computed Tomography set up to study the effect of some physical parameters. One must decide the number of detectors and set CT geometry parameters or configuration like the fan-beam angle and number of rotations. Usually, geometry parameters are determined based on the objects size. This study shows the influence of the density distribution of same-sized phantom or objects on CT geometry parameters. Due to limited space, industrial applications may not allow a CT system to move around the object (under investigation). On-spot customization of CT system configuration may be required according to similar situations. The same problem is experienced in medical science, material science, and many other fields in which the CT system is widely used for non-destructive imaging. The number of detectors in the scanning array is one major factor that optimizes a CT system. Of course, more detectors are desired for better resolution. Changing the number of detectors requires recalibration of CT geometry. A simulated work is presented to study the influence of the density distribution of same-sized objects and the number of detectors on CT system configuration. The same is verified experimentally also. A comparison between simulated and experimental results shows a good agreement.

physics.ins-det↗

arXiv2022

Characterization of 3D Printers and X-Ray Computerized Tomography

The 3D printing process flow requires several inputs for the best printing quality. These settings may vary from sample to sample, printer to printer, and depend upon users' previous experience. The involved operational parameters for 3D Printing are varied to test the optimality. Thirty-eight samples are printed using four commercially available 3D printers, namely: (a) Ultimaker 2 Extended+, (b) Delta Wasp, (c) Raise E2, and (d) ProJet MJP. The sample profiles contain uniform and non-uniform distribution of the assorted size of cubes and spheres with a known amount of porosity. These samples are scanned using X-Ray Computed Tomography system. Functional Imaging analysis is performed using AI-based segmentation codes to (a) characterize these 3D printers and (b) find Three-dimensional surface roughness of three teeth and one sandstone pebble (from riverbed) with naturally deposited layers is also compared with printed sample values. Teeth has best quality. It is found that ProJet MJP gives the best quality of printed samples with the least amount of surface roughness and almost near to the actual porosity value. As expected, 100% infill density value, best spatial resolution for printing or Layer height, and minimum nozzle speed give the best quality of 3D printing.

eess.IV↗

arXiv2022

Effect of scattering and electronic noise upon selection of detectors for Gamma Computerized Tomography

Computed tomography (CT) has become a vital tool in a variety of fields as a result of technological developments and continual improvement. High-quality CT images are desirable for image interpretation and obtaining information from CT images. A variety of things influence the CT image quality. Various research groups have investigated and attempted to improve image quality by examining noise/error associated with CT geometry. This study aims to select detectors for CT, which yield the least amount of noise in projection data. Three distinct gamma-ray detectors that are routinely used in CT have been compared in terms of scattering and electrical noise. The sensitivity of Kanpur Theorem-1 to scattering noise is demonstrated in this work and used to quantify the relative level of scattering noise. The detector measures the signal multiple times, and the standard deviation of the signal is used to calculate the electronic noise. It is observed that IC CsI(Tl) scintillation detector produces low electronic noise and relative scattering noise as compared to conventional electronic detectors; NaI(Tl) and HPGe.

physics.ins-det↗

arXiv2022

LabVIEW is faster and C is economical interfacing tool for UCT automation

An in-house developed 2D ultrasound computerized Tomography system is fully automated. Performance analysis of instrument and software interfacing soft tools, namely the LabVIEW, MATLAB, C, and Python, is presented. The instrument interfacing algorithms, hardware control algorithms, signal processing, and analysis codes are written using above mentioned soft tool platforms. Total of eight performance indices are used to compare the ease of (a) realtime control of electromechanical assembly, (b) sensors, instruments integration, (c) synchronized data acquisition, and (d) simultaneous raw data processing. It is found that C utilizes the least processing power and performs a lower number of processes to perform the same task. In runtime analysis (data acquisition and realtime control), LabVIEW performs best, taking 365.69s in comparison to MATLAB (623.83s), Python ( 1505.54s), and C (1252.03s) to complete the experiment. Python performs better in establishing faster interfacing and minimum RAM usage. LabVIEW is recommended for its fast process execution. C is recommended for the most economical implementation. Python is recommended for complex system automation having a very large number of components involved. This article provides a methodology to select optimal soft tools for instrument automation-related aspects.

cs.PL↗

arXiv2022

Noise analysis, error estimates, and Gamma Radiation Measurement for limited detector computerized tomography application

Computed Tomography is one of the efficient and vital modalities of non-destructive techniques (NDT). Various factors influence the CT reconstruction result, including limited projection data, detector electronics optimization, background noise, detection noise, discretized nature of projection data, and many more. Radiation hardening and other aging factors that affect the operational settings may require recalibration of electronics parameters. Two well-known exercises are utilized with the motivation to improve reliability and accuracy in inverse recovery. The first exercise brute-forces an optimal candidate from the set of calibration methods for minimum error in inverse recovery. The second exercise, Kanpur Theorem-1 (KT-1) examines if optimal calibration sets electronics to impart minimum noise. The mutual conformity between statistics-derived CLT and Riemann integral transform-based KT-1 is shown first time using gamma radiation measurement. The analysis shows that measurement data with normal distribution inflicts the least noise in inverse recovery.

eess.IV↗

arXiv2022

A Manifold View of Adversarial Risk

The adversarial risk of a machine learning model has been widely studied. Most previous works assume that the data lies in the whole ambient space. We propose to take a new angle and take the manifold assumption into consideration. Assuming data lies in a manifold, we investigate two new types of adversarial risk, the normal adversarial risk due to perturbation along normal direction, and the in-manifold adversarial risk due to perturbation within the manifold. We prove that the classic adversarial risk can be bounded from both sides using the normal and in-manifold adversarial risks. We also show with a surprisingly pessimistic case that the standard adversarial risk can be nonzero even when both normal and in-manifold risks are zero. We finalize the paper with empirical studies supporting our theoretical results. Our results suggest the possibility of improving the robustness of a classifier by only focusing on the normal adversarial risk.

cs.LG↗

arXiv2022

Obtaining Approximately Optimal and Diverse Solutions via Dispersion

There has been a long-standing interest in computing diverse solutions to optimization problems. Motivated by reallocation of governmental institutions in Sweden, in 1995 J. Krarup posed the problem of finding $k$ edge-disjoint Hamiltonian Circuits of minimum total weight, called the peripatetic salesman problem (PSP). Since then researchers have investigated the complexity of finding diverse solutions to spanning trees, paths, vertex covers, matchings, and more. Unlike the PSP that has a constraint on the total weight of the solutions, recent work has involved finding diverse solutions that are all optimal. However, sometimes the space of exact solutions may be too small to achieve sufficient diversity. Motivated by this, we initiate the study of obtaining sufficiently-diverse, yet approximately-optimal solutions to optimization problems. Formally, given an integer $k$, an approximation factor $c$, and an instance $I$ of an optimization problem, we aim to obtain a set of $k$ solutions to $I$ that a) are all $c$ approximately-optimal for $I$ and b) maximize the diversity of the $k$ solutions. Finding such solutions, therefore, requires a better understanding of the global landscape of the optimization function. We show that, given any metric on the space of solutions, and the diversity measure as the sum of pairwise distances between solutions, this problem can be solved by combining ideas from dispersion and multicriteria optimization. We first provide a general reduction to an associated budget-constrained optimization (BCO) problem, where one objective function is to be maximized (minimized) subject to a bound on the second objective function. We then prove that bi-approximations to the BCO can be used to give bi-approximations to the diverse approximately optimal solutions problem with a little overhead.

cs.DS↗

arXiv2022

A Topological Filter for Learning with Label Noise

Noisy labels can impair the performance of deep neural networks. To tackle this problem, in this paper, we propose a new method for filtering label noise. Unlike most existing methods relying on the posterior probability of a noisy classifier, we focus on the much richer spatial behavior of data in the latent representational space. By leveraging the high-order topological information of data, we are able to collect most of the clean data and train a high-quality model. Theoretically we prove that this topological approach is guaranteed to collect the clean data with high probability. Empirical results show that our method outperforms the state-of-the-arts and is robust to a broad spectrum of noise types and levels.

cs.CV↗

arXiv2021

AI based Scintillation Detector Calibration

Data set generated from the scintillation detector is used to build a mathematical model based on three different algorithms: (a) Multiple Polynomial Regression (b) Support Vector Regression (c) Neural Network algorithm. Using visualizations and correlations, it is found that the Median of the data will give accurate results and average time has a major contribution in radiation measurement.

physics.ins-det↗

arXiv2021

Sensitivity Analysis of calibration methods and factors effecting the statistical nature of radiation measurement

Scintillator detectors electronics is recalibrated against the datasheet given by the manufacturer. Optimal and mutual dependent values of (a) high voltage at PMT (Photomultiplier Tube), (b) amplifier gain, (c) average time to count the radiation particles (set by operator), and (d) number of instances/sample number are estimated. Total 5: two versions of Central Limit Theorem (CLT), (3) industry preferred Pulse Width Saturation, (4) calibration based on MPPC coupled Gamma-ray detector and (5) gross method are used. It is shown that CLT method is the most optimal method to calibrate detector and its respective electronics couple. An inverse modeling-based Computerized Tomography method is used for verification. It is shown that statistically averaging results are more accurate and precise data than mode and median, if the data is not skewed and random number of samples are used during the calibration process. It is also shown that the average time to count the radiation particle is the most important parameter affecting the optimal calibration setting for precision and accurate measurements of gamma radiation.

physics.ins-det↗

arXiv2021

AI and conventional methods for UCT projection data estimation

A 2D Compact ultrasound computerized tomography (UCT) system is developed. Fully automatic post processing tools involving signal and image processing are developed as well. Square of the amplitude values are used in transmission mode with natural 1.5 MHz frequency and rise time 10.4 ns and fall time 8.4 ns and duty cycle of 4.32%. Highest peak to corresponding trough values are considered as transmitting wave between transducers in direct line talk. Sensitivity analysis of methods to extract peak to corresponding trough per transducer are discussed in this paper. Total five methods are tested. These methods are taken from broad categories: (a) Conventional and (b) Artificial Intelligence (AI) based methods. Conventional methods, namely: (a) simple gradient based peak detection, (b) Fourier based, (c) wavelet transform are compared with AI based methods: (a) support vector machine (SVM), (b) artificial neural network (ANN). Classification step was performed as well to discard the signal which does not has contribution of transmission wave. It is found that conventional methods have better performance. Reconstruction error, accuracy, F-Score, recall, precision, specificity and MCC for 40 x 40 data 1600 data files are measured. Each data file contains 50,002 data point. Ten such data files are used for training the Neural Network. Each data file has 7/8 wave packets and each packet corresponds to one transmission amplitude data. Reconstruction error is found to be minimum for ANN method. Other performance indices show that FFT method is processing the UCT signal with best recovery.

eess.SP↗

arXiv2021

AI Algorithm for Mode Classification of PCF SPR Sensor Design

Photonic Crystal Fiber design based on surface plasmon resonance phenomenon (PCF SPR) is optimized before it is fabricated for a particular application. An artificial intelligence algorithm is evaluated here to increase the ease of the simulation process for common users. COMSOL MultiPhysics is used. The algorithm suggests best among eight standard machine learning and one deep learning model to automatically select the desired mode, chosen visually by the experts otherwise. Total seven performance indices: namely Precision, Recall, Accuracy, F1-Score, Specificity, Matthew correlation coefficient, are utilized to make the optimal decision. Robustness towards variations in sensor geometry design is also considered as an optimal parameter. Several PCF-SPR based Photonic sensor designs are tested, and a large range optimal (based on phase matching) design is proposed. For this design algorithm has selected Support Vector Machine (SVM) as the best option with an accuracy of 96%, F1-Score is 95.83%, and MCC of 92.30%. The average sensitivity of the proposed sensor design with respect to change in refractive index (1.37-1.41) is 5500 nm/RIU. Resolution is 2.0498x10^(-5) RIU^(-1). The algorithm can be integrated into commercial software as an add-on or as a module in academic codes. The proposed novel step has saved approximately 75 minutes in the overall design process. The present work is equally applicable for mode selection of sensor other than PCF-SPR sensing geometries.

eess.SP↗

arXiv2021

Deep Learning models for benign and malign Ocular Tumor Growth Estimation

Relatively abundant availability of medical imaging data has provided significant support in the development and testing of Neural Network based image processing methods. Clinicians often face issues in selecting suitable image processing algorithm for medical imaging data. A strategy for the selection of a proper model is presented here. The training data set comprises optical coherence tomography (OCT) and angiography (OCT-A) images of 50 mice eyes with more than 100 days follow-up. The data contains images from treated and untreated mouse eyes. Four deep learning variants are tested for automatic (a) differentiation of tumor region with healthy retinal layer and (b) segmentation of 3D ocular tumor volumes. Exhaustive sensitivity analysis of deep learning models is performed with respect to the number of training and testing images using 8 eight performance indices to study accuracy, reliability/reproducibility, and speed. U-net with UVgg16 is best for malign tumor data set with treatment (having considerable variation) and U-net with Inception backbone for benign tumor data (with minor variation). Loss value and root mean square error (R.M.S.E.) are found most and least sensitive performance indices, respectively. The performance (via indices) is found to be exponentially improving regarding a number of training images. The segmented OCT-Angiography data shows that neovascularization drives the tumor volume. Image analysis shows that photodynamic imaging-assisted tumor treatment protocol is transforming an aggressively growing tumor into a cyst. An empirical expression is obtained to help medical professionals to choose a particular model given the number of images and types of characteristics. We recommend that the presented exercise should be taken as standard practice before employing a particular deep learning model for biomedical image analysis.

eess.IV↗

arXiv2021

Topological Detection of Trojaned Neural Networks

Deep neural networks are known to have security issues. One particular threat is the Trojan attack. It occurs when the attackers stealthily manipulate the model's behavior through Trojaned training samples, which can later be exploited. Guided by basic neuroscientific principles we discover subtle -- yet critical -- structural deviation characterizing Trojaned models. In our analysis we use topological tools. They allow us to model high-order dependencies in the networks, robustly compare different networks, and localize structural abnormalities. One interesting observation is that Trojaned models develop short-cuts from input to output layers. Inspired by these observations, we devise a strategy for robust detection of Trojaned models. Compared to standard baselines it displays better performance on multiple benchmarks.

cs.LG↗