arXiv subjects

Gokce Guven

Publications and source records attributed to Gokce Guven.

3 recordsLinked to original sources

arXiv2026

Statistical Hand Shape Modeling from Clinical CT Scans Using Deep Learning and Implicit Skinning

Accurate segmentation and statistical shape modeling of hand anatomy have significant implications for medical diagnostics, ergonomics, and biomechanics. This study proposes an AI-assisted reconstruction pipeline for segmenting and analyzing hand anatomy from 1,271 elbow-to-hand (e2h-CT) computed tomography scans. A Pix2Pix-based conditional generative adversarial network is first employed to remove plaster cast and background artifacts from CT volumes. The cleaned scans are then processed in 3D Slicer to extract skin and bone masks, which are converted into closed-surface mesh models. Segmented bone meshes are used to construct skeletal representations, enabling implicit skinning to align all hand models into a standardized anatomical configuration. Subsequently, non-rigid registration is performed on the hand skin surfaces using the Geodesic Based Coherent Point Drift++ (GBCPD++) algorithm to establish point-wise correspondence across subjects. Principal Component Analysis (PCA) is then applied to the registered models to quantify anatomical shape variability. The Pix2Pix preprocessing stage achieved a Dice coefficient of 0.9856 and an IoU of 0.9720 on the held-out test set. Statistical modeling was performed on a subset of 90 scans in which the fingers were fully visible and anatomically separated. The resulting statistical shape distributions demonstrate strong agreement with the U.S. Army Anthropometric Survey (ANSUR II), supporting the anatomical validity of the reconstructed models. The proposed methodology demonstrates significant potential for advancing biomechanical modeling, ergonomic optimization, prosthetic design, and precision medical diagnostics.

cs.CV↗

arXiv2025

X2BR: High-Fidelity 3D Bone Reconstruction from a Planar X-Ray Image with Hybrid Neural Implicit Methods

Accurate 3D bone reconstruction from a single planar X-ray remains a challenge due to anatomical complexity and limited input data. We propose X2BR, a hybrid neural implicit framework that combines continuous volumetric reconstruction with template-guided non-rigid registration. The core network, X2B, employs a ConvNeXt-based encoder to extract spatial features from X-rays and predict high-fidelity 3D bone occupancy fields without relying on statistical shape models. To further refine anatomical accuracy, X2BR integrates a patient-specific template mesh, constructed using YOLOv9-based detection and the SKEL biomechanical skeleton model. The coarse reconstruction is aligned to the template using geodesic-based coherent point drift, enabling anatomically consistent 3D bone volumes. Experimental results on a clinical dataset show that X2B achieves the highest numerical accuracy, with an IoU of 0.952 and Chamfer-L1 distance of 0.005, outperforming recent baselines including X2V and D2IM-Net. Building on this, X2BR incorporates anatomical priors via YOLOv9-based bone detection and biomechanical template alignment, leading to reconstructions that, while slightly lower in IoU (0.875), offer superior anatomical realism, especially in rib curvature and vertebral alignment. This numerical accuracy vs. visual consistency trade-off between X2B and X2BR highlights the value of hybrid frameworks for clinically relevant 3D reconstructions.

cs.CV↗

arXiv2014

Protonation States of Remote Residues Affect Binding-Release Dynamics of the Ligand but not the Conformation of apo Ferric Binding Protein

We have studied the apo (Fe3+ free) form of periplasmic ferric binding protein (FbpA) under different conditions and we have monitored the changes in the binding and release dynamics of H2PO4- that acts as a synergistic anion in the presence of Fe3+. Our simulations predict a dissociation constant of 2.2$\pm$0.2 mM which is in remarkable agreement with the experimentally measured value of 2.3$\pm$0.3 mM under the same ionization strength and pH conditions. We apply perturbations relevant for changes in environmental conditions as (i) different values of ionic strength (IS), and (ii) protonation of a group of residues to mimic a different pH environment. Local perturbations are also studied by protonation or mutation of a site distal to the binding region that is known to mechanically manipulate the hinge-like motions of FbpA. We find that while the average conformation of the protein is intact in all simulations, the H2PO4- dynamics may be substantially altered by the changing conditions. In particular, the bound fraction which is 20$\%$ for the wild type system is increased to 50$\%$ with a D52A mutation/protonation and further to over 90$\%$ at the protonation conditions mimicking those at pH 5.5. The change in the dynamics is traced to the altered electrostatic distribution on the surface of the protein which in turn affects hydrogen bonding patterns at the active site. The observations are quantified by rigorous free energy calculations. Our results lend clues as to how the environment versus single residue perturbations may be utilized for regulation of binding modes in hFbpA systems in the absence of conformational changes.

q-bio.BM↗