Search

Uncertainty in medical image segmentation tasks, especially inter-rater variability, arising from differences in interpretations and annotations by various experts, presents a significant challenge in achieving consistent and reliable image segmentation. This variability not only reflects the inherent complexity and subjective nature of medical image interpretation but also directly impacts the development and evaluation of automated segmentation algorithms. Accurately modeling and quantifying this variability is essential for enhancing the robustness and clinical applicability of these algorithms. We report the set-up and summarize the benchmark results of the Quantification of Uncertainties in Biomedical Image Quantification Challenge (QUBIQ), which was organized in conjunction with International Conferences on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2020 and 2021. The challenge focuses on the uncertainty quantification of medical image segmentation which considers the omnipresence of inter-rater variability in imaging datasets. The large collection of images with multi-rater annotations features various modalities such as MRI and CT; various organs such as the brain, prostate, kidney, and pancreas; and different image dimensions 2D-vs-3D. A total of 24 teams submitted different solutions to the problem, combining various baseline models, Bayesian neural networks, and ensemble model techniques. The obtained results indicate the importance of the ensemble models, as well as the need for further research to develop efficient 3D methods for uncertainty quantification methods in 3D segmentation tasks., Comment: initial technical report

MRSI is an advanced imaging tool used for early detection of tumors. Analysis of MR spectra is constrained by high noise and long fitting times using model-based tools. We propose data-driven methods, such as denoising, for improving the data quality by reducing noise, and machine learning to accelerate the metabolite quantification in the presence of noise and artifacts., MRSI ist ein fortschrittliches Bildgebungsinstrument zur Früherkennung von Tumoren. Die Analyse von MR-Spektren wird durch hohes Rauschen und lange Anpassungszeiten mit modellbasierten Werkzeugen eingeschränkt. Wir schlagen datengesteuerte Methoden wie Entrauschen vor, um die Datenqualität durch Reduzierung des Rauschens zu verbessern, und maschinelles Lernen, um die Quantifizierung der Metaboliten bei Vorhandensein von Rauschen und Artefakten zu beschleunigen.

Pressure inference from a series of velocity fields is a common problem arising in medical imaging when analyzing 4D data. Traditional approaches primarily rely on a numerical scheme to solve the pressure-Poisson equation to obtain a dense pressure inference. This involves heavy expert intervention at each stage and requires significant computational resources. Concurrently, the application of current machine learning algorithms for solving partial differential equations is limited to domains with simple boundary conditions. We address these challenges in this paper and present V2P-Net: a novel, neural-network-based approach as an alternative method for inferring pressure from the observed velocity fields. We design an end-to-end hybrid-network architecture motivated by the conventional Navier-Stokes solver, which encapsulates the complex boundary conditions. It achieves accurate pressure estimation compared to the reference numerical solver for simulated flow data in multiple complex geometries of human in-vivo vessels.

MRSI is an advanced imaging tool used for early detection of tumors. Analysis of MR spectra is constrained by high noise and long fitting times using model-based tools. We propose data-driven methods, such as denoising, for improving the data quality by reducing noise, and machine learning to accelerate the metabolite quantification in the presence of noise and artifacts., MRSI ist ein fortschrittliches Bildgebungsinstrument zur Früherkennung von Tumoren. Die Analyse von MR-Spektren wird durch hohes Rauschen und lange Anpassungszeiten mit modellbasierten Werkzeugen eingeschränkt. Wir schlagen datengesteuerte Methoden wie Entrauschen vor, um die Datenqualität durch Reduzierung des Rauschens zu verbessern, und maschinelles Lernen, um die Quantifizierung der Metaboliten bei Vorhandensein von Rauschen und Artefakten zu beschleunigen.

Nearest neighbor search (NNS) has a wide range of applications in information retrieval, computer vision, machine learning, databases, and other areas. Existing state-of-the-art algorithm for nearest neighbor search, Hierarchical Navigable Small World Networks(HNSW), is unable to scale to large datasets of 100M records in high dimensions. In this paper, we propose LANNS, an end-to-end platform for Approximate Nearest Neighbor Search, which scales for web-scale datasets. Library for Large Scale Approximate Nearest Neighbor Search (LANNS) is deployed in multiple production systems for identifying topK ($100 \leq topK \leq 200$) approximate nearest neighbors with a latency of a few milliseconds per query, high throughput of 2.5k Queries Per Second (QPS) on a single node, on large ($\sim$180M data points) high dimensional (50-2048 dimensional) datasets., Comment: 10 pages, 9 figures, 9 tables

Three-dimensional (3D) biomedical image sets are often acquired with in-plane pixel spacings that are far less than the out-of-plane spacings between images. The resultant anisotropy, which can be detrimental in many applications, can be decreased using image interpolation. Optical flow and/or other registration-based interpolators have proven useful in such interpolation roles in the past. When acquired images are comprised of signals that describe the flow velocity of fluids, additional information is available to guide the interpolation process. In this paper, we present an optical-flow based framework for image interpolation that also minimizes resultant divergence in the interpolated data., Comment: 24 pages, 10 figures, International Journal of Biomedical Imaging, accepted manuscript

Magnetic Resonance Spectroscopic Imaging (MRSI) is a clinical imaging modality for measuring tissue metabolite levels in-vivo. An accurate estimation of spectral parameters allows for better assessment of spectral quality and metabolite concentration levels. The current gold standard quantification method is the LCModel - a commercial fitting tool. However, this fails for spectra having poor signal-to-noise ratio (SNR) or a large number of artifacts. This paper introduces a framework based on random forest regression for accurate estimation of the output parameters of a model based analysis of MR spectroscopy data. The goal of our proposed framework is to learn the spectral features from a training set comprising of different variations of both simulated and in-vivo brain spectra and then use this learning for the subsequent metabolite quantification. Experiments involve training and testing on simulated and in-vivo human brain spectra. We estimate parameters such as concentration of metabolites and compare our results with that from the LCModel.

The advancement in wireless communication technologies and rapid growth in bandwidth hungry application has aggravated the need for efficient utilization of the spectrum resource. Cognitive radio (CR) has emerged as a promising solution to the present problem of spectrum crisis. It increased the efficiency of spectrum utilization by allowing secondary/unlicensed user to coexist with the primary/licensed user in a particular spectrum band. The stringent requirement is a reliable and efficient spectrum sensing technique. In this thesis both hard and soft combination based cooperative spectrum sensing techniques has been investigated. The local CR sensors sense the energy of a specified narrow band and report the data or the preliminary decision to the fusion center. Double threshold based hard decision combining is reinvestigated in this thesis where the local CR sensors use two thresholds instead of one. When the sensed energy falls within the two thresholds no decision is made. Majority based decision combination is used at the fusion center to combine the local reliable decisions from the CR sensors. Analytical expressions of probability of sensing error and probability of false alarm are derived as a function of the width of the ambiguity region (region between the two thresholds). Using the analytical and simulation results it is shown that there exists an optimal value of the width of ambiguity region for which the performance is optimum. Finally an optimization problem is formulated solving which the optimum width of ambiguity region can be found. Next, a soft decision based cooperative cognitive radio is investigated, where a non-uniform quantization technique is proposed prior to reporting. The quantization of the locally sensed energies controls the backhaul bandwidth required for reporting the data. The non-uniform quantization is done based on the likelihood function, which is the probability that a particular decision is correct given a value of sensed en

The advancement in wireless communication technologies and rapid growth in bandwidth hungry application has aggravated the need for efficient utilization of the spectrum resource. Cognitive radio (CR) has emerged as a promising solution to the present problem of spectrum crisis. It increased the efficiency of spectrum utilization by allowing secondary/unlicensed user to coexist with the primary/licensed user in a particular spectrum band. The stringent requirement is a reliable and efficient spectrum sensing technique. In this thesis both hard and soft combination based cooperative spectrum sensing techniques has been investigated. The local CR sensors sense the energy of a specified narrow band and report the data or the preliminary decision to the fusion center. Double threshold based hard decision combining is reinvestigated in this thesis where the local CR sensors use two thresholds instead of one. When the sensed energy falls within the two thresholds no decision is made. Majority based decision combination is used at the fusion center to combine the local reliable decisions from the CR sensors. Analytical expressions of probability of sensing error and probability of false alarm are derived as a function of the width of the ambiguity region (region between the two thresholds). Using the analytical and simulation results it is shown that there exists an optimal value of the width of ambiguity region for which the performance is optimum. Finally an optimization problem is formulated solving which the optimum width of ambiguity region can be found. Next, a soft decision based cooperative cognitive radio is investigated, where a non-uniform quantization technique is proposed prior to reporting. The quantization of the locally sensed energies controls the backhaul bandwidth required for reporting the data. The non-uniform quantization is done based on the likelihood function, which is the probability that a particular decision is correct given a value of sensed en

The advancement in wireless communication technologies and rapid growth in bandwidth hungry application has aggravated the need for efficient utilization of the spectrum resource. Cognitive radio (CR) has emerged as a promising solution to the present problem of spectrum crisis. It increased the efficiency of spectrum utilization by allowing secondary/unlicensed user to coexist with the primary/licensed user in a particular spectrum band. The stringent requirement is a reliable and efficient spectrum sensing technique. In this thesis both hard and soft combination based cooperative spectrum sensing techniques has been investigated. The local CR sensors sense the energy of a specified narrow band and report the data or the preliminary decision to the fusion center. Double threshold based hard decision combining is reinvestigated in this thesis where the local CR sensors use two thresholds instead of one. When the sensed energy falls within the two thresholds no decision is made. Majority based decision combination is used at the fusion center to combine the local reliable decisions from the CR sensors. Analytical expressions of probability of sensing error and probability of false alarm are derived as a function of the width of the ambiguity region (region between the two thresholds). Using the analytical and simulation results it is shown that there exists an optimal value of the width of ambiguity region for which the performance is optimum. Finally an optimization problem is formulated solving which the optimum width of ambiguity region can be found. Next, a soft decision based cooperative cognitive radio is investigated, where a non-uniform quantization technique is proposed prior to reporting. The quantization of the locally sensed energies controls the backhaul bandwidth required for reporting the data. The non-uniform quantization is done based on the likelihood function, which is the probability that a particular decision is correct given a value of sensed en

The advancement in wireless communication technologies and rapid growth in bandwidth hungry application has aggravated the need for efficient utilization of the spectrum resource. Cognitive radio (CR) has emerged as a promising solution to the present problem of spectrum crisis. It increased the efficiency of spectrum utilization by allowing secondary/unlicensed user to coexist with the primary/licensed user in a particular spectrum band. The stringent requirement is a reliable and efficient spectrum sensing technique. In this thesis both hard and soft combination based cooperative spectrum sensing techniques has been investigated. The local CR sensors sense the energy of a specified narrow band and report the data or the preliminary decision to the fusion center. Double threshold based hard decision combining is reinvestigated in this thesis where the local CR sensors use two thresholds instead of one. When the sensed energy falls within the two thresholds no decision is made. Majority based decision combination is used at the fusion center to combine the local reliable decisions from the CR sensors. Analytical expressions of probability of sensing error and probability of false alarm are derived as a function of the width of the ambiguity region (region between the two thresholds). Using the analytical and simulation results it is shown that there exists an optimal value of the width of ambiguity region for which the performance is optimum. Finally an optimization problem is formulated solving which the optimum width of ambiguity region can be found. Next, a soft decision based cooperative cognitive radio is investigated, where a non-uniform quantization technique is proposed prior to reporting. The quantization of the locally sensed energies controls the backhaul bandwidth required for reporting the data. The non-uniform quantization is done based on the likelihood function, which is the probability that a particular decision is correct given a value of sensed en