Salman Ibne Eunus

Mr. Salman Ibne Eunus

Computer Science and Engineering, BRAC University, Dhaka, Bangladesh
orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 April 2024 in AI
Verified
Journal Article
Open access
PDF file
Salman Ibne EunusSalman Ibne Eunus
Salman Ibne Eunus
SHShahriar Hossain
Shahriar Hossain
ARA. E. M. Ridwan
A. E. M. Ridwan
Ashik AdnanAshik Adnan
Ashik Adnan
Md. Saiful IslamMd. Saiful Islam
Md. Saiful Islam
Jia UddinJia Uddin
Jia Uddin
Accidents due to defective railway lines and derailments are common disasters that are observed frequently in Southeast Asian countries. It is imperative to run proper diagnosis over the detection of such faults to prevent such accidents. However, manual detection of such faults periodically can be both time-consuming and costly. In this paper, we have proposed a Deep Learning (DL)-based algorithm for automatic fault detection in railway tracks, which we termed an Ensembled Convolutional Autoencoder ResNet-based Recurrent Neural Network (ECARRNet). We compared its output with existing DL techniques in the form of several pre-trained DL models to investigate railway tracks and determine whether they are defective or not while considering commonly prevalent faults such as—defects in rails and fasteners. Moreover, we manually collected the images from different railway tracks situated in Bangladesh and made our dataset. After comparing our proposed model with the existing models, we found that our proposed architecture has produced the highest accuracy among all the previously existing state-of-the-art (SOTA) architecture, with an accuracy of 93.28% on the full dataset. Additionally, we split our dataset into two parts having two different types of faults, which are fasteners and rails. We ran the models on those two separate datasets, obtaining accuracies of 98.59% and 92.06% on rail and fastener, respectively. Model explainability techniques like Grad-CAM and LIME were used to validate the result of the models, where our proposed model ECARRNet was seen to correctly classify and detect the regions of faulty railways effectively compared to the previously existing transfer learning models.

19 July 2023 in OSF Preprints
Verified
Preprint
Open access
Salman Ibne EunusSalman Ibne Eunus
Salman Ibne Eunus
Accidents due to defective railway lines and derailments are common disasters that are observed frequently in south- east Asian countries. It is imperative to run proper diagnosis over the detection of such faults to prevent such accidents. However, manual detection of such faults on a periodic basis can be both time-consuming and costly. In this paper, we have proposed a Deep Learning (DL)-based algorithm for automatic fault detection in railway tracks, which we termed an Ensembled Convolutional Autoencoder ResNet-based Recurrent Neural Network (ECARRNet).

24 July 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
Salman Ibne EunusSalman Ibne Eunus
Salman Ibne Eunus
AIAzhar Mohd Ibrahim
Azhar Mohd Ibrahim
This review synthesizes recent advances in hybrid control architectures for autonomous multi-drone swarms, with particular emphasis on approaches that integrate deep reinforcement learning (DRL) with model-based control and safety-aware mechanisms. The review examines swarm coordination as a constrained decision-making problem under nonlinear quadrotor dynamics and graph-based interaction topologies. It surveys major hybridization strategies, including DRL combined with classical controllers, planning-learning frameworks, swarm-intelligence-assisted optimization, and safety-enhancing techniques such as control barrier functions and runtime shielding. Drawing on literature published between 2018 and 2026, the paper evaluates the current state of the field and highlights key trends, strengths, and limitations of existing approaches. Particular attention is given to the challenges of achieving safe decision-making, scalable multi-agent coordination, and reliable deployment in real-world environments. The review identifies three major research gaps: the lack of scalable frameworks that combine learning with formal safety guarantees, limited solutions for robust simulation-to-reality transfer of collective behaviors, and insufficient consideration of hardware constraints during training and evaluation. To support more consistent assessment of hybrid swarm architectures, the review discusses evaluation perspectives centered on safety, reliability, efficiency, scalability, and deployment feasibility. Finally, it consolidates emerging best practices from the literature and outlines future directions, including verifiable hybrid control synthesis, scalable multi-agent reinforcement learning, hardware-aware policy design, human-swarm interaction, and standardized benchmarking. The review provides researchers and practitioners with a structured roadmap for developing safe, scalable, and certifiable autonomous drone swarms.

1 August 2023 in OSF Preprints
Verified
Preprint
Open access
Salman Ibne EunusSalman Ibne Eunus
Salman Ibne Eunus
Bone metastasis is a frequently occurring disease and can bea consequence of a number of different cancers such as - prostrate, lungand breast cancers, and predicting them can be really useful for the diag-nosis of patients with such diseases. Classifying images of bone scan forbone metastasis prediction requires a huge amount of data to producea prediction output which is reliable and accurate, but a single medicalorganization usually do not have access to such amounts of data fromother organizations and those organizations are not also ready to sharetheir patients’ private data as well due to data security issues. For suchscenarios, it is not often possible to train a model with enough data, thusleading to an inaccurate prediction model for bone metastasis. This canbe devastating at times due to the occurrence of many false positives orfalse negatives, if bone metastasis is wrongly classified. In order to finda better solution, so that there is less data protection and privacy issuesand therefore more availability of data, we are proposing to use a Feder-ated Learning (FL) based approach for bone metastasis prediction usingconvolutional neural network. As per our knowledge and backgroundstudy, we are the first to use federated learning for bone metastasis pre-diction on the BS-80K dataset. Federated Averaging (FedAvg) strategywas used for implementing the federated learning methodology wheredifferent client models were built along with a Global Model (server)

15 August 2023 in OSF Preprints
Verified
Preprint
Open access
Salman Ibne EunusSalman Ibne Eunus
Salman Ibne Eunus
The gut microbiome is the aggregation of all the microorganisms, bacteria, viruses, protozoa and fungi, and their collective genetic material present in the Gastrointestinal Tract (GIT) of human body. Human gut mainly consists of bacteria, both commensal (good) and pathogenic (bad) bacteria in the GIT. It is colonized by 10^14 microbes, which is 10 times more than total human cells. Gut Microbiota has been a topic of research to discover the potential gut microbe-host interactions which includes effects on metabolism, immune system and neuroendocrine responses, etc