Searcharxiv⌕ Search

arXiv subjects

Masoud Ganji

Publications and source records attributed to Masoud Ganji.

11 recordsLinked to original sources
arXiv2025

Null fluid gravitational fields on Kerr manifolds and optical lifts of Sasaki structures

Building on the characterisation in [C. D. Hill, J. Lewandowski and P. Nurowski, Indiana Univ. Math. J. 57 (2008), 3131--3176] of 4-dimensional Lorentzian metrics adapted to an optical structure and satisfying the null fluid Einstein equations, we give an explicit parameterisation of this class under the assumption that the optical structure is of Kerr type. As immediate consequences, we obtain: (1) a new method for constructing solutions to the Einstein equations with a null fluid energy momentum tensor, yielding a large family of explicit metrics that naturally includes the classical Kerr black hole metrics and all Ricci flat examples described in [M. Ganji, C. Giannotti, G. Schmalz and A. Spiro, Ann. Physics 75 (2025), Paper No. 169908, 28]; (2) a solution to a conjecture in Hill, Lewandowski and Nurowski's paper on the local existence of smooth optical lifts in the case of Sasaki CR structures.

gr-qc↗
arXiv2024

A New Era in Computational Pathology: A Survey on Foundation and Vision-Language Models

Recent advances in deep learning have completely transformed the domain of computational pathology (CPath). More specifically, it has altered the diagnostic workflow of pathologists by integrating foundation models (FMs) and vision-language models (VLMs) in their assessment and decision-making process. The limitations of existing deep learning approaches in CPath can be overcome by FMs through learning a representation space that can be adapted to a wide variety of downstream tasks without explicit supervision. Deploying VLMs allow pathology reports written in natural language be used as rich semantic information sources to improve existing models as well as generate predictions in natural language form. In this survey, a holistic and systematic overview of recent innovations in FMs and VLMs in CPath is presented. Furthermore, the tools, datasets and training schemes for these models are summarized in addition to categorizing them into distinct groups. This extensive survey highlights the current trends in CPath and its possible revolution through the use of FMs and VLMs in the future.

cs.LG↗
arXiv2024

Einstein manifolds with optical geometries of Kerr type

We classify the Ricci flat Lorentzian $n$-manifolds satisfying three particular conditions, encoding and combining some crucial features of the Kerr metrics and the Robinson-Trautman optical structures. We prove that: (a) If $n>4$, there is no Lorentzian manifold satisfying the considered Kerr type conditions, in unexpected contrast with what occurs for the metrics satisfying (very similar) Taub-NUT type conditions; (b) If $n=4$ there are two large classes of such Kerr type manifolds. Each class consists of manifolds fibering over open Riemann surfaces, equipped with a metric of constant Gaussian curvature $κ= 1$ or $κ= -1$. The first class includes a three parameter family of metrics admitting real analytic extensions to $(\mathbb R^3 \setminus\{0\}) \times \mathbb R = (S^2 \times \mathbb R_+) \times \mathbb R$ and a large class of other metrics not admitting this kind of extensions. The metrics of this first class admitting such extensions are all isometric to the well known Kerr metrics, with the three parameters corresponding to the three space-like components of the angular momentum of the gravitational field. The second class contains a subclass of metrics defined on $\big(\mathbb D\times \mathbb R_+\big)\times \mathbb R$, where $\mathbb D$ is the Lobachevsky Poincaré disc. This subclass is in bijection with the holomorphic functions on $\mathbb D$ satisfying an appropriate open condition. These and other results are consequences of a very simple way to construct totally explicit examples of Ricci flat Lorentzian manifolds.

math.DG↗
arXiv2024

Quasi-Einstein shearfree spacetimes lifted from Sasakian manifolds

In this article we prove that a certain class of {\it smooth} Sasakian manifolds admits lifts to 4-dimensional quasi-Einstein shearfree spacetimes of Petrov type II or D. This is related to an analogous result by Hill, Lewandowski and Nurowski \cite{HLN} for general {\it real-analytic} CR manifolds. In particular, this holds for all tubular CR manifolds. Furthermore, we show that any Sasakian manifold with underlying Kähler-Einstein manifold with non-zero Einstein constant has a lift to a shearfree Einstein metric of Petrov type II or D.

math.DG↗
arXiv2024

Adversary-Robust Graph-Based Learning of WSIs

Enhancing the robustness of deep learning models against adversarial attacks is crucial, especially in critical domains like healthcare where significant financial interests heighten the risk of such attacks. Whole slide images (WSIs) are high-resolution, digitized versions of tissue samples mounted on glass slides, scanned using sophisticated imaging equipment. The digital analysis of WSIs presents unique challenges due to their gigapixel size and multi-resolution storage format. In this work, we aim at improving the robustness of cancer Gleason grading classification systems against adversarial attacks, addressing challenges at both the image and graph levels. As regards the proposed algorithm, we develop a novel and innovative graph-based model which utilizes GNN to extract features from the graph representation of WSIs. A denoising module, along with a pooling layer is incorporated to manage the impact of adversarial attacks on the WSIs. The process concludes with a transformer module that classifies various grades of prostate cancer based on the processed data. To assess the effectiveness of the proposed method, we conducted a comparative analysis using two scenarios. Initially, we trained and tested the model without the denoiser using WSIs that had not been exposed to any attack. We then introduced a range of attacks at either the image or graph level and processed them through the proposed network. The performance of the model was evaluated in terms of accuracy and kappa scores. The results from this comparison showed a significant improvement in cancer diagnosis accuracy, highlighting the robustness and efficiency of the proposed method in handling adversarial challenges in the context of medical imaging.

cs.CV↗
arXiv2023

Artifact-Robust Graph-Based Learning in Digital Pathology

Whole slide images~(WSIs) are digitized images of tissues placed in glass slides using advanced scanners. The digital processing of WSIs is challenging as they are gigapixel images and stored in multi-resolution format. A common challenge with WSIs is that perturbations/artifacts are inevitable during storing the glass slides and digitizing them. These perturbations include motion, which often arises from slide movement during placement, and changes in hue and brightness due to variations in staining chemicals and the quality of digitizing scanners. In this work, a novel robust learning approach to account for these artifacts is presented. Due to the size and resolution of WSIs and to account for neighborhood information, graph-based methods are called for. We use graph convolutional network~(GCN) to extract features from the graph representing WSI. Through a denoiser {and pooling layer}, the effects of perturbations in WSIs are controlled and the output is followed by a transformer for the classification of different grades of prostate cancer. To compare the efficacy of the proposed approach, the model without denoiser is trained and tested with WSIs without any perturbation and then different perturbations are introduced in WSIs and passed through the network with the denoiser. The accuracy and kappa scores of the proposed model with prostate cancer dataset compared with non-robust algorithms show significant improvement in cancer diagnosis.

eess.IV↗
arXiv2023

The Levi-Civita connections of manifolds with prescribed optical geometries

We explicitly derive the Christoffel symbols in terms of adapted frame fields for the Levi-Civita connection of a Lorentzian $n$-manifold $(M, g)$, equipped with a prescribed optical geometry of Kähler-Sasaki type. The formulas found in this paper have several important applications, such as determining the geometric invariants of Lorentzian manifolds with prescribed optical geometries or solving curvature constraints.

math.DG↗
arXiv2023

Spatio-temporal determinantal point processes

Determinantal point processes are models for regular spatial point patterns, with appealing probabilistic properties. We present their spatio-temporal counterparts and give examples of these models, based on spatio-temporal covariance functions which are separable and non-separable in space and time.

math.ST↗
arXiv2021

Lorentzian manifolds with shearfree congruences and Kähler-Sasaki geometry

We study Lorentzian manifolds $(M, g)$ of dimension $n\geq 4$, equipped with a maximally twisting shearfree null vector field $p_o$, for which the leaf space $S = M/\{\exp t p_o\}$ is a smooth manifold. If $n = 2k$, the quotient $S = M/\{\exp t p_o\}$ is naturally equipped with a subconformal structure of contact type and, in the most interesting cases, it is a regular Sasaki manifold projecting onto a quantisable Kähler manifold of real dimension $2k -2$. Going backwards through this line of ideas, for any quantisable Kähler manifold with associated Sasaki manifold $S$, we give the local description of all Lorentzian metrics $g$ on the total spaces $M$ of $A$-bundles $π: M \to S$, $A = S^1, \mathbb R$, such that the generator of the group action is a maximally twisting shearfree $g$-null vector field $p_o$. We also prove that on any such Lorentzian manifold $(M, g)$ there exists a non-trivial generalized electromagnetic plane wave having $p_o$ as propagating direction field, a result that can be considered as a generalization of the classical $4$-dimensional Robinson Theorem. We finally construct a 2-parametric family of Einstein metrics on a trivial bundle $M = \mathbb R \times S$ for any prescribed value of the Einstein constant. If $\dim M = 4$, the Ricci flat metrics obtained in this way are the well-known Taub-NUT metrics.

math.DG↗
arXiv2018

On a criterion for local embeddability of 3-dimensional CR-structures

We introduce a CR-invariant class of Lorentzian metrics on a circle bundle over a 3-dimensional CR-structure, which we call quasi-Fefferman metrics. These metrics generalise the Fefferman metric but allow for more control of the Ricci curvature. Our main result is a criterion for embaddability of 3-dimensional CR-structures in terms of the Ricci curvature of the quasi-Fefferman metrics in the spirit of the results by Hill et al.

math.CV↗