Inshad Rahman Noman

Mr. Inshad Rahman Noman

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

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

Representative Publications

View All Publications
8 November 2024 in IEEE Access
Verified
Journal Article
Open access
ALAhmed Ali Linkon
Ahmed Ali Linkon
INInshad Rahman Noman
Inshad Rahman Noman
RIRashedul Islam
Rashedul Islam
JBJoy Chakra Bortty
Joy Chakra Bortty
Kanchon Kumar BishnuKanchon Kumar Bishnu
Kanchon Kumar Bishnu
Masuk AbdullahMasuk Abdullah
Masuk Abdullah
The increasing prevalence of diabetes necessitates the development of effective early detection methods to mitigate its health impacts. This paper investigates the impact of feature transformation and machine learning (ML) models on the early detection of diabetes using a binary tabular classification dataset. We explore three feature transformation techniques, no transformation, normalization, and min-max scaling, to assess their influence on the performance of various ML models. To comprehensively evaluate the effectiveness of these preprocessing techniques, we experimented with twelve different ML models, including both traditional algorithms and ensemble methods. A publicly available dataset has been used for this research, containing 768 samples and 8 features. To ensure their effectiveness, the models are assessed using several evaluation metrics, including accuracy, precision, recall, F1-score, and Area Under the Receiver Operating Characteristic Curve (AUC-ROC). Among the ML models, Light Gradient Boosting Machine (LGBM) achieved the highest accuracy of 82.91% when min-max scaling was applied to the data. Our results demonstrate the varying effectiveness of different combinations of feature transformation techniques and ML models in enhancing diabetes detection performance. Furthermore, it has been observed that the ensemble models generally achieved better performance than traditional ML models. These findings provide valuable insights for optimizing preprocessing and model selection strategies in the development of robust early diabetes detection systems.

23 January 2025 in Frontiers in Bioscience-Elite
Verified
Journal Article
Open access
PDF file
MBMohammad Muzahidur Rahman Bhuiyan
Mohammad Muzahidur Rahman Bhuiyan
INInshad Rahman Noman
Inshad Rahman Noman
Munna AzizMunna Aziz
Munna Aziz
MRMd Mizanur Rahaman
Md Mizanur Rahaman
MIMD Rashedul Islam
MD Rashedul Islam
Kallol DasKallol Das
Kallol Das
Our study focused on plant breeding, from traditional methods to the present most advanced genetic and data-driven concepts. Conventional breeding techniques, such as mass selection and cross-breeding, have been instrumental in crop improvement, although they possess inherent limitations in precision and efficiency. Advanced molecular methods allow breeders to improve crops quicker by more accurately targeting specific traits. Data analytics and information technology (IT) are crucial in modern plant breeding, providing tools for data management, analysis, and interpretation of large volumes of data from genomic, phenotypic, and environmental sources. Meanwhile, emerging technologies in machine learning, high-throughput phenotyping, and the Internet of Things (IoT) provide real-time insights into the performance and responses of plants to environmental variables, enabling precision breeding. These tools will allow breeders to select complex traits related to yield, disease resistance, and abiotic stress tolerance more precisely and effectively. Moreover, this data-driven approach will enable breeders to use resources judiciously and make crops resilient, thus contributing to sustainable agriculture. Data analytics integrated into IT will enhance traditional breeding and other key applications in sustainable agriculture, such as crop yield improvement, biofortification, and climate change adaptation. This review aims to highlight the role of interdisciplinary collaboration among breeders, data scientists, and agronomists in absorbing these technologies. Further, this review discusses the future trends that will make plant breeding even more effective with this new wave of artificial intelligence (AI), blockchain, and collaborative platforms, bringing new data transparency, collaboration, and predictability levels. Data and IT-based breeding will greatly contribute to future global food security and sustainable food production. Thus, creating high-performing, resource-efficient crops will be the foundation of a future agricultural vision that balances environmental care. More technological integration in plant breeding is needed for resilient and sustainable food systems to handle the growing population and changing climate challenges.

20 March 2025 in Diagnostics
Verified
Journal Article
Open access
PDF file
JBJoy Chakra Bortty
Joy Chakra Bortty
GCGouri Shankar Chakraborty
Gouri Shankar Chakraborty
INInshad Rahman Noman
Inshad Rahman Noman
SBSalil Batra
Salil Batra
Joy DasJoy Das
Joy Das
AIAraf Islam
Araf Islam
Background/Objectives: Alzheimer’s disease (AD) is a progressive, neurodegenerative disorder, which causes memory loss and loss of cognitive functioning, along with behavioral changes. Early detection is important to delay disease progression, timely intervention and to increase patients’ and caregivers’ quality of life (QoL). One of the major and primary challenges for preventing any disease is to identify the disease at the initial stage through a quick and reliable detection process. Different researchers across the world are still working relentlessly, coming up with significant solutions. Artificial intelligence-based solutions are putting great importance on identifying the disease efficiently, where deep learning with medical imaging is highly being utilized to develop disease detection frameworks. In this work, a novel and optimized detection framework has been proposed that comes with remarkable performance that can classify the level of Alzheimer’s accurately and efficiently. Methods: A powerful vision transformer model (ViT-B16) with three efficient Convolutional Neural Network (CNN) models (VGG19, ResNet152V2, and EfficientNetV2B3) has been trained with a benchmark dataset, ‘OASIS’, that comes with a high volume of brain Magnetic Resonance Images (MRI). Results: A weighted average ensemble technique with a Grasshopper optimization algorithm has been designed and utilized to ensure maximum performance with high accuracy of 97.31%, precision of 97.32, recall of 97.35, and F1 score of 0.97. Conclusions: The work has been compared with other existing state-of-the-art techniques, where it comes with high efficiency, sensitivity, and reliability. The framework can be utilized in IoMT infrastructure where one can access smart and remote diagnosis services.

Verified
Journal Article
Open access
PDF file
MRMd Mizanur Rahaman
Md Mizanur Rahaman
MIMD Rashedul Islam
MD Rashedul Islam
MMMia Md Tofayel Gonee Manik
Mia Md Tofayel Gonee Manik
Munna AzizMunna Aziz
Munna Aziz
INInshad Rahman Noman
Inshad Rahman Noman
JBJoy Chakra Bortty
Joy Chakra Bortty
Accurately predicting how quickly people will adopt electric vehicles (EVs) is vital for planning charging stations, managing supply chains, and shaping climate policy. We present a forecasting model that uses three separate Long Short-Term Memory (LSTM) branches—one for past EV sales, one for infrastructure and policy signals, and one for economic trends. An attention mechanism first highlights the most important weeks in each branch, then decides which branch matters most at any point in time. Trained end-to-end on publicly available data, the model beats traditional statistical methods and newer deep learning baselines while remaining small enough to run efficiently. An ablation study shows that every branch and both attention steps improve accuracy, and that adding policy and economic data helps more than relying on EV history alone. Because the network is modular and its attention weights are easy to interpret, it can be extended to produce confidence intervals, include physical constraints, or forecast adoption of other clean-energy technologies.

10 March 2025
Verified
Conference Paper
Closed access
MRMd Mizanur Rahaman
Md Mizanur Rahaman
MIMD Rashedul Islam
MD Rashedul Islam
MBMohammad Muzahidur Rahman Bhuiyan
Mohammad Muzahidur Rahman Bhuiyan
INInshad Rahman Noman
Inshad Rahman Noman
Munna AzizMunna Aziz
Munna Aziz
Kallol DasKallol Das
Kallol Das
Multi-omics approaches have transformed the assessment of biological systems by offering a clear hallmark of intricate diseases. In this work, we introduce a novel ML+DL to improve biomarkers and disease prognosis methods using multi-omics data. Using Recursive Feature Elimination (RFE) for feature selection, we combined RF, SVM, and XGB with DNN and LSTM to analyze feature- specific information as well as temporal dependencies in the datasets. Experiments showed that proposed models of SVM+LSTM have the highest accuracy of 98.8%, with SVM+DNN having a slightly lower value of 98.7%. In contrast, RF+DNN and RF+LSTM had acc of 97.5% and 97.1%, respectively, and SVM-based models have an average accuracy of 97.0 to 97.1. Based on these results, the robustness of the hybrid models in capturing intricate relationships across multi- omics layers is demonstrated. Altogether, the outcomes evidentially demonstrate the applicability of novel ML/DL models specifically for individual diagnosis, furthering the therapeutic approach and improving biotechnological processes in both diagnostics and therapeutic applications. Although data integration remains an issue and there are scalability barriers to computation, this work presents a method by which precision medicine can be achieved by linking molecular information to clinics.

28 May 2026
Verified
Book Chapter
Closed access
PDF file
Kallol DasKallol Das
Kallol Das
MMMia Md Tofayel Gonee Manik
Mia Md Tofayel Gonee Manik
MIMD Rashedul Islam
MD Rashedul Islam
MRMd Mizanur Rahaman
Md Mizanur Rahaman
MBMohammad Muzahidur Rahman Bhuiyan
Mohammad Muzahidur Rahman Bhuiyan
INInshad Rahman Noman
Inshad Rahman Noman