Searcharxiv⌕ Search

arXiv subjects

Isa Yildirim

Publications and source records attributed to Isa Yildirim.

15 recordsLinked to original sources
arXiv2025

Deep Unfolded BM3D: Unrolling Non-local Collaborative Filtering into a Trainable Neural Network

Block-Matching and 3D Filtering (BM3D) exploits non-local self-similarity priors for denoising but relies on fixed parameters. Deep models such as U-Net are more flexible but often lack interpretability and fail to generalize across noise regimes. In this study, we propose Deep Unfolded BM3D (DU-BM3D), a hybrid framework that unrolls BM3D into a trainable architecture by replacing its fixed collaborative filtering with a learnable U-Net denoiser. This preserves BM3D's non-local structural prior while enabling end-to-end optimization. We evaluate DU-BM3D on low-dose CT (LDCT) denoising and show that it outperforms classic BM3D and standalone U-Net across simulated LDCT at different noise levels, yielding higher PSNR and SSIM, especially in high-noise conditions.

eess.IV↗
arXiv2025

Prompt-Conditioned FiLM and Multi-Scale Fusion on MedSigLIP for Low-Dose CT Quality Assessment

We propose a prompt-conditioned framework built on MedSigLIP that injects textual priors via Feature-wise Linear Modulation (FiLM) and multi-scale pooling. Text prompts condition patch-token features on clinical intent, enabling data-efficient learning and rapid adaptation. The architecture combines global, local, and texture-aware pooling through separate regression heads fused by a lightweight MLP, trained with pairwise ranking loss. Evaluated on the LDCTIQA2023 (a public LDCT quality assessment challenge) with 1,000 training images, we achieve PLCC = 0.9575, SROCC = 0.9561, and KROCC = 0.8301, surpassing the top-ranked published challenge submissions and demonstrating the effectiveness of our prompt-guided approach.

cs.CV↗
arXiv2025

Learnable Total Variation with Lambda Mapping for Low-Dose CT Denoising

While Total Variation (TV) excels in noise reduction and edge preservation, its reliance on a scalar regularization parameter limits adaptivity. In this study, we present a Learnable Total Variation (LTV) framework coupling an unrolled TV solver with a LambdaNet that predicts a per-pixel regularization map. The proposed framework is trained end-to-end to optimize reconstruction and regularization jointly, yielding spatially adaptive smoothing. Experiments on the DeepLesion dataset, using realistic LoDoPaB-CT simulation, show consistent gains over classical TV and FBP+U-Net, achieving up to +3.7 dB PSNR and 8% relative SSIM improvement. LTV provides an interpretable alternative to black-box CNNs for low-dose CT denoising.

cs.CV↗
arXiv2025

Task-Adaptive Low-Dose CT Reconstruction

Deep learning-based low-dose computed tomography reconstruction methods already achieve high performance on standard image quality metrics like peak signal-to-noise ratio and structural similarity index measure. Yet, they frequently fail to preserve the critical anatomical details needed for diagnostic tasks. This fundamental limitation hinders their clinical applicability despite their high metric scores. We propose a novel task-adaptive reconstruction framework that addresses this gap by incorporating a frozen pre-trained task network as a regularization term in the reconstruction loss function. Unlike existing joint-training approaches that simultaneously optimize both reconstruction and task networks, and risk diverging from satisfactory reconstructions, our method leverages a pre-trained task model to guide reconstruction training while still maintaining diagnostic quality. We validate our framework on a liver and liver tumor segmentation task. Our task-adaptive models achieve Dice scores up to 0.707, approaching the performance of full-dose scans (0.874), and substantially outperforming joint-training approaches (0.331) and traditional reconstruction methods (0.626). Critically, our framework can be integrated into any existing deep learning-based reconstruction model through simple loss function modification, enabling widespread adoption for task-adaptive optimization in clinical practice. Our codes are available at: https://github.com/itu-biai/task_adaptive_ct

eess.IV↗
arXiv2025

LMM-IQA: Image Quality Assessment for Low-Dose CT Imaging

Low-dose computed tomography (CT) represents a significant improvement in patient safety through lower radiation doses, but increased noise, blur, and contrast loss can diminish diagnostic quality. Therefore, consistency and robustness in image quality assessment become essential for clinical applications. In this study, we propose an LLM-based quality assessment system that generates both numerical scores and textual descriptions of degradations such as noise, blur, and contrast loss. Furthermore, various inference strategies - from the zero-shot approach to metadata integration and error feedback - are systematically examined, demonstrating the progressive contribution of each method to overall performance. The resultant assessments yield not only highly correlated scores but also interpretable output, thereby adding value to clinical workflows. The source codes of our study are available at https://github.com/itu-biai/lmms_ldct_iqa.

cs.CV↗
arXiv2022

3D U-NetR: Low Dose Computed Tomography Reconstruction via Deep Learning and 3 Dimensional Convolutions

In this paper, we introduced a novel deep learning-based reconstruction technique for low-dose CT imaging using 3 dimensional convolutions to include the sagittal information unlike the existing 2 dimensional networks which exploits correlation only in transverse plane. In the proposed reconstruction technique, sparse and noisy sinograms are back-projected to the image domain with FBP operation, then the denoising process is applied with a U-Net like 3-dimensional network called 3D U-NetR. The proposed network is trained with synthetic and real chest CT images, and 2D U-Net is also trained with the same dataset to show the importance of the third dimension in terms of recovering the fine details. The proposed network shows better quantitative performance on SSIM and PSNR, especially in the real chest CT data. More importantly, 3D U-NetR captures medically critical visual details that cannot be visualized by a 2D network on the reconstruction of real CT images with 1/10 of the normal dose.

cs.CV↗
arXiv2021

On the stability and convergence of Mann iteration process in convex A- metric spaces

In this paper, firstly, we introduce the concept of convexity in A-metric spaces and show that Mann iteration process converges to the unique fixed point of Zamfirescu type contractions in this newly defined convex A-metric space. Secondly, we define the concept of stability in convex A-metric spaces and establish stability result for the Mann iteration process considered in such spaces. Our results carry some well-known results from the literature to convex A-metric spaces.

math.GN↗
arXiv2021

Fixed point results for Zamfirescu mappings in A-metric spaces

In the present paper, we extend the Zamfirescu results ([9]) to A-metric spaces. Firstly, we define the notion of Zamfirescu mapping in A-metric spaces. After, we also obtain a fixed point theorem for such mappings. The established results carry some well-known results from the literature (see [2], [3], [4], [9]) to A-metric spaces. Appropriate example is also given.

math.FA↗
arXiv2020

An Unsupervised Reconstruction Method For Low-Dose CT Using Deep Generative Regularization Prior

Low-dose CT imaging requires reconstruction from noisy indirect measurements which can be defined as an ill-posed linear inverse problem. In addition to conventional FBP method in CT imaging, recent compressed sensing based methods exploit handcrafted priors which are mostly simplistic and hard to determine. More recently, deep learning (DL) based methods have become popular in medical imaging field. In CT imaging, DL based methods try to learn a function that maps low-dose images to normal-dose images. Although the results of these methods are promising, their success mostly depends on the availability of high-quality massive datasets. In this study, we proposed a method that does not require any training data or a learning process. Our method exploits such an approach that deep convolutional neural networks (CNNs) generate patterns easier than the noise, therefore randomly initialized generative neural networks can be suitable priors to be used in regularizing the reconstruction. In the experiments, the proposed method is implemented with different loss function variants. Both analytical CT phantoms and human CT images are used with different views. Conventional FBP method, a popular iterative method (SART), and TV regularized SART are used in the comparisons. We demonstrated that our method with different loss function variants outperforms the other methods both qualitatively and quantitatively.

eess.IV↗
arXiv2020

Self-Supervised Training For Low Dose CT Reconstruction

Ionizing radiation has been the biggest concern in CT imaging. To reduce the dose level without compromising the image quality, low-dose CT reconstruction has been offered with the availability of compressed sensing based reconstruction methods. Recently, data-driven methods got attention with the rise of deep learning, the availability of high computational power, and big datasets. Deep learning based methods have also been used in low-dose CT reconstruction problem in different manners. Usually, the success of these methods depends on labeled data. However, recent studies showed that training can be achieved successfully with noisy datasets. In this study, we defined a training scheme to use low-dose sinograms as their own training targets. We applied the self-supervision principle in the projection domain where the noise is element-wise independent which is a requirement for self-supervised training methods. Using the self-supervised training, the filtering part of the FBP method and the parameters of a denoiser neural network are optimized. We demonstrate that our method outperforms both conventional and compressed sensing based iterative reconstruction methods qualitatively and quantitatively in the reconstruction of analytic CT phantoms and real-world CT images in low-dose CT reconstruction task.

eess.IV↗
arXiv2019

Convexity in G-metric spaces and approximation of fixed points by Mann iterative proces

In this paper, we first define the concept of convexity in G-metric spaces. We then use Mann iterative process in this newly defined convex G-metric space to prove some convergence results for some classes of mappings. In this way, we can extend several existence results to those approximating fixed points. Our results are just new in the setting.

math.GN↗
arXiv2017

Convergence rate of implicit iteration process and a data dependence result

The aim of this paper is to introduce an implicit S-iteration process and study its convergence in the framework of W-hyperbolic spaces. We show that the implicit S-iteration process has higher rate of convergence than implicit Mann type iteration and implicit Ishikawa-type iteration processes. We present a numerical example to support the analytic result proved herein. Finally, we prove a data dependence result for a contractive type mapping using implicit S-iteration process.

math.FA↗