-
Multi-OCT-SelfNet: Integrating Self-Supervised Learning with Multi-Source Data Fusion for Enhanced Multi-Class Retinal Disease Classification
Authors:
Fatema-E- Jannat,
Sina Gholami,
Jennifer I. Lim,
Theodore Leng,
Minhaj Nur Alam,
Hamed Tabkhi
Abstract:
In the medical domain, acquiring large datasets poses significant challenges due to privacy concerns. Nonetheless, the development of a robust deep-learning model for retinal disease diagnosis necessitates a substantial dataset for training. The capacity to generalize effectively on smaller datasets remains a persistent challenge. The scarcity of data presents a significant barrier to the practica…
▽ More
In the medical domain, acquiring large datasets poses significant challenges due to privacy concerns. Nonetheless, the development of a robust deep-learning model for retinal disease diagnosis necessitates a substantial dataset for training. The capacity to generalize effectively on smaller datasets remains a persistent challenge. The scarcity of data presents a significant barrier to the practical implementation of scalable medical AI solutions. To address this issue, we've combined a wide range of data sources to improve performance and generalization to new data by giving it a deeper understanding of the data representation from multi-modal datasets and developed a self-supervised framework based on large language models (LLMs), SwinV2 to gain a deeper understanding of multi-modal dataset representations, enhancing the model's ability to extrapolate to new data for the detection of eye diseases using optical coherence tomography (OCT) images. We adopt a two-phase training methodology, self-supervised pre-training, and fine-tuning on a downstream supervised classifier. An ablation study conducted across three datasets employing various encoder backbones, without data fusion, with low data availability setting, and without self-supervised pre-training scenarios, highlights the robustness of our method. Our findings demonstrate consistent performance across these diverse conditions, showcasing superior generalization capabilities compared to the baseline model, ResNet-50.
△ Less
Submitted 17 September, 2024;
originally announced September 2024.
-
A survey on conjugacy class graphs of groups
Authors:
P. J. Cameron,
F. E. Jannat,
R. K. Nath,
R. Sharafdini
Abstract:
There are several graphs defined on groups. Among them we consider graphs whose vertex set consists conjugacy classes of a group $G$ and adjacency is defined by properties of the elements of conjugacy classes. In particular, we consider commuting/nilpotent/solvable conjugacy class graph of $G$ where two distinct conjugacy classes $a^G$ and $b^G$ are adjacent if there exist some elements…
▽ More
There are several graphs defined on groups. Among them we consider graphs whose vertex set consists conjugacy classes of a group $G$ and adjacency is defined by properties of the elements of conjugacy classes. In particular, we consider commuting/nilpotent/solvable conjugacy class graph of $G$ where two distinct conjugacy classes $a^G$ and $b^G$ are adjacent if there exist some elements $x\in a^G$ and $y\in b^G$ such that $\langle x, y \rangle$ is abelian/nilpotent/solvable. After a section of introductory results and examples, we discuss all the available results on connectedness, graph realization, genus, various spectra and energies of certain induced subgraphs of these graphs. Proofs of the results are not included. However, many open problems for further investigation are stated.
△ Less
Submitted 19 March, 2024; v1 submitted 14 March, 2024;
originally announced March 2024.
-
Common neighborhood (signless) Laplacian spectrum and energy of CCC-graph
Authors:
Firdous Ee Jannat,
Rajat Kanti Nath
Abstract:
In this paper, we consider commuting conjugacy class graph (abbreviated as CCC-graph) of a finite group $G$ which is a graph with vertex set $\{x^G : x \in G \setminus Z(G)\}$ (where $x^G$ denotes the conjugacy class containing $x$) and two distinct vertices $x^G$ and $y^G$ are joined by an edge if there exist some elements $x'\in x^G$ and $y'\in y^G$ such that they commute. We compute common neig…
▽ More
In this paper, we consider commuting conjugacy class graph (abbreviated as CCC-graph) of a finite group $G$ which is a graph with vertex set $\{x^G : x \in G \setminus Z(G)\}$ (where $x^G$ denotes the conjugacy class containing $x$) and two distinct vertices $x^G$ and $y^G$ are joined by an edge if there exist some elements $x'\in x^G$ and $y'\in y^G$ such that they commute. We compute common neighborhood (signless) Laplacian spectrum and energy of CCC-graph of finite non-abelian groups whose central quotient is isomorphic to either $\mathbb{Z}_p \times \mathbb{Z}_p$ (where $p$ is any prime) or the dihedral group $D_{2n}$ ($n \geq 3$); and determine whether CCC-graphs of these groups are common neighborhood (signless) Laplacian hyperenergetic/borderenergetic. As a consequence, we characterize certain finite non-abelian groups viz. $D_{2n}$, $T_{4n}$, $U_{6n}$, $U_{(n, m)}$, $SD_{8n}$ and $V_{8n}$ such that their CCC-graphs are common neighborhood (signless) Laplacian hyperenergetic/borderenergetic.
△ Less
Submitted 5 March, 2024;
originally announced March 2024.
-
Common neighborhood Laplacian and signless Laplacian spectra and energies of commuting graphs
Authors:
Firdous Ee Jannat,
Rajat Kanti Nath
Abstract:
In this paper, we compute common neighbourhood Laplacian spectrum, common neighbourhood signless Laplacian spectrum and their respective energies of commuting graph of some finite non-abelian groups including some AC-groups, groups whose central quotients are isomorphic to $Sz(2)$, $\mathbb{Z}_p\times \mathbb{Z}_p$ or $D_{2m}$. Our findings lead us to conclude that these graphs are CNL (CNSL)-inte…
▽ More
In this paper, we compute common neighbourhood Laplacian spectrum, common neighbourhood signless Laplacian spectrum and their respective energies of commuting graph of some finite non-abelian groups including some AC-groups, groups whose central quotients are isomorphic to $Sz(2)$, $\mathbb{Z}_p\times \mathbb{Z}_p$ or $D_{2m}$. Our findings lead us to conclude that these graphs are CNL (CNSL)-integral. Additionally, we characterize the aforementioned groups such that their commuting graphs are CNL (CNSL)-hyperenergetic.
△ Less
Submitted 1 March, 2024;
originally announced March 2024.
-
Common neighborhood energies and their relations with Zagreb index
Authors:
Firdous Ee Jannat,
Rajat Kanti Nath,
Kinkar Chandra Das
Abstract:
In this paper we establish connections between common neighborhood Laplacian and common neighborhood signless Laplacian energies and the first Zagreb index of a graph $\mathcal{G}$. We introduce the concepts of CNL-hyperenergetic and CNSL-hyperenergetic graphs and showed that $\mathcal{G}$ is neither CNL-hyperenergetic nor CNSL-hyperenergetic if $\mathcal{G}$ is a complete bipartite graph. We obta…
▽ More
In this paper we establish connections between common neighborhood Laplacian and common neighborhood signless Laplacian energies and the first Zagreb index of a graph $\mathcal{G}$. We introduce the concepts of CNL-hyperenergetic and CNSL-hyperenergetic graphs and showed that $\mathcal{G}$ is neither CNL-hyperenergetic nor CNSL-hyperenergetic if $\mathcal{G}$ is a complete bipartite graph. We obtain certain relations between various energies of a graph. Finally, we conclude the paper with several bounds for common neighborhood Laplacian and signless Laplacian energies of a graph.
△ Less
Submitted 23 February, 2024;
originally announced February 2024.
-
OCT-SelfNet: A Self-Supervised Framework with Multi-Modal Datasets for Generalized and Robust Retinal Disease Detection
Authors:
Fatema-E Jannat,
Sina Gholami,
Minhaj Nur Alam,
Hamed Tabkhi
Abstract:
Despite the revolutionary impact of AI and the development of locally trained algorithms, achieving widespread generalized learning from multi-modal data in medical AI remains a significant challenge. This gap hinders the practical deployment of scalable medical AI solutions. Addressing this challenge, our research contributes a self-supervised robust machine learning framework, OCT-SelfNet, for d…
▽ More
Despite the revolutionary impact of AI and the development of locally trained algorithms, achieving widespread generalized learning from multi-modal data in medical AI remains a significant challenge. This gap hinders the practical deployment of scalable medical AI solutions. Addressing this challenge, our research contributes a self-supervised robust machine learning framework, OCT-SelfNet, for detecting eye diseases using optical coherence tomography (OCT) images. In this work, various data sets from various institutions are combined enabling a more comprehensive range of representation. Our method addresses the issue using a two-phase training approach that combines self-supervised pretraining and supervised fine-tuning with a mask autoencoder based on the SwinV2 backbone by providing a solution for real-world clinical deployment. Extensive experiments on three datasets with different encoder backbones, low data settings, unseen data settings, and the effect of augmentation show that our method outperforms the baseline model, Resnet-50 by consistently attaining AUC-ROC performance surpassing 77% across all tests, whereas the baseline model exceeds 54%. Moreover, in terms of the AUC-PR metric, our proposed method exceeded 42%, showcasing a substantial increase of at least 10% in performance compared to the baseline, which exceeded only 33%. This contributes to our understanding of our approach's potential and emphasizes its usefulness in clinical settings.
△ Less
Submitted 22 January, 2024;
originally announced January 2024.
-
VegaEdge: Edge AI Confluence Anomaly Detection for Real-Time Highway IoT-Applications
Authors:
Vinit Katariya,
Fatema-E- Jannat,
Armin Danesh Pazho,
Ghazal Alinezhad Noghre,
Hamed Tabkhi
Abstract:
Vehicle anomaly detection plays a vital role in highway safety applications such as accident prevention, rapid response, traffic flow optimization, and work zone safety. With the surge of the Internet of Things (IoT) in recent years, there has arisen a pressing demand for Artificial Intelligence (AI) based anomaly detection methods designed to meet the requirements of IoT devices. Catering to this…
▽ More
Vehicle anomaly detection plays a vital role in highway safety applications such as accident prevention, rapid response, traffic flow optimization, and work zone safety. With the surge of the Internet of Things (IoT) in recent years, there has arisen a pressing demand for Artificial Intelligence (AI) based anomaly detection methods designed to meet the requirements of IoT devices. Catering to this futuristic vision, we introduce a lightweight approach to vehicle anomaly detection by utilizing the power of trajectory prediction. Our proposed design identifies vehicles deviating from expected paths, indicating highway risks from different camera-viewing angles from real-world highway datasets. On top of that, we present VegaEdge - a sophisticated AI confluence designed for real-time security and surveillance applications in modern highway settings through edge-centric IoT-embedded platforms equipped with our anomaly detection approach. Extensive testing across multiple platforms and traffic scenarios showcases the versatility and effectiveness of VegaEdge. This work also presents the Carolinas Anomaly Dataset (CAD), to bridge the existing gap in datasets tailored for highway anomalies. In real-world scenarios, our anomaly detection approach achieves an AUC-ROC of 0.94, and our proposed VegaEdge design, on an embedded IoT platform, processes 738 trajectories per second in a typical highway setting. The dataset is available at https://github.com/TeCSAR-UNCC/Carolinas_Dataset#chd-anomaly-test-set .
△ Less
Submitted 13 November, 2023;
originally announced November 2023.
-
Common neighbourhood spectrum and energy of commuting conjugacy class graph
Authors:
Firdous Ee Jannat,
Rajat Kanti Nath
Abstract:
In this paper we compute common neighbourhood (abbreviated as CN) spectrum and energy of commuting conjugacy class graph of several families of finite non-abelian groups. As a consequence of our results we show that the commuting conjugacy class graphs of the groups $D_{2n}$, $T_{4n}$, $SD_{8n}$, $U_{(n,m)}$, $U_{6n}$, $V_{8n}$, $G(p, m, n)$ and some families of groups whose central quotient is is…
▽ More
In this paper we compute common neighbourhood (abbreviated as CN) spectrum and energy of commuting conjugacy class graph of several families of finite non-abelian groups. As a consequence of our results we show that the commuting conjugacy class graphs of the groups $D_{2n}$, $T_{4n}$, $SD_{8n}$, $U_{(n,m)}$, $U_{6n}$, $V_{8n}$, $G(p, m, n)$ and some families of groups whose central quotient is isomorphic to $D_{2n}$ or $Z_p \times Z_p$, for some prime $p$, are CN-integral but not CN-hyperenergetic.
△ Less
Submitted 22 August, 2022;
originally announced August 2022.
