Md. Razu Ahmed

Mr. Md. Razu Ahmed

Deep Statistical Learning and Research Lab, Pabna University of Science and Technology, Pābna, Rājshāhi, Bangladesh
+1
orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
17 December 2025 in International Journal of Statistical Sciences
Verified
Journal Article
Open access
Md. Razu AhmedMd. Razu Ahmed
Md. Razu Ahmed
ARAbdur Rakib
Abdur Rakib
ASAbu Bakar Shiddik
Abu Bakar Shiddik
SRShamim Reza
Shamim Reza
Chronic kidney disease (CKD) remains a major global health concern, often progressing silently until advanced stages. Early detection is therefore critical to improving outcomes. This study focuses on identifying the most important predisposing risk factors for CKD and optimizing prediction performance using a stacking machine learning ensemble. Two datasets were analyzed: the UCI CKD dataset and a synthetically simulated CKD dataset designed to mirror real-world variability. A leakage-safe preprocessing pipeline was implemented, including median and mode imputation for missing values, Z-score capping for outliers, min–max normalization, and class balancing through the Synthetic Minority Over-sampling Technique (SMOTE). Feature selection was performed using six complementary approaches: Logistic Regression (LR), Recursive Feature Elimination (RFE), Random Forest (RF), Mutual Information (MI), Chi-Square (χ²), and Principal Component Analysis (PCA) with a majority-vote strategy used to identify features consistently recognized as predictive. For the UCI CKD dataset, the common risk factors were serum creatinine, hemoglobin, packed cell volume, red blood cell count, specific gravity, albumin, sugar, hypertension, diabetes mellitus, and appetite. For the simulated CKD dataset, key predictors included blood glucose random, blood urea, serum creatinine, potassium, hemoglobin, packed cell volume, specific gravity, albumin, red blood cells, pus cell clumps, appetite, pedal edema, and anemia. Using these selected features, the stacking ensemble achieved 100.0% accuracy on the UCI CKD dataset and 96.7% accuracy on the simulated dataset, both with negligible misclassification rates. Bootstrap confidence intervals confirmed the robustness of these results. The findings highlight that combining systematic feature selection with stacking significantly improves predictive accuracy while maintaining interpretability. This integrated framework offers a reliable tool for early CKD detection and can support clinical decision-making in real-world healthcare settings. Future work will focus on expanding validation across multi-site data and developing a clinical decision support interface. IJSS, Vol. 25(2), November, 2025, pp 1-32

13 September 2026 in Data in Brief
Verified
Journal Article
Open access
MYMarufa Yeasmin
Marufa Yeasmin
Md. Razu AhmedMd. Razu Ahmed
Md. Razu Ahmed
PKProtiva Rani Kundu
Protiva Rani Kundu
JMJannatul Mauya
Jannatul Mauya
MRMd. Shamim Reza
Md. Shamim Reza

9 November 2026 in BMC Public Health
Verified
Journal Article
Open access
PDF file
MHMuhammad Ziaul Haque
Muhammad Ziaul Haque
Md. Razu AhmedMd. Razu Ahmed
Md. Razu Ahmed
MSMimosa Saha
Mimosa Saha
Md. Feroz AliMd. Feroz Ali
Md. Feroz Ali
DBDiganto Biswas
Diganto Biswas
OOObaidullah Obaidi
Obaidullah Obaidi
Dengue places recurrent pressure on surveillance and hospital services in Bangladesh, yet most Bangladesh-focused forecasting studies have used national or temporally aggregated data. Whether retaining district-level information before national aggregation provides additional value for short-range forecasting remains uncertain. This study evaluated district-informed and direct national approaches for one-week-ahead forecasting of reported dengue hospital admissions across all 64 districts of Bangladesh. A nationwide surveillance–meteorological panel for 2022–2025 comprised 93,504 district-day observations aggregated into 13,312 complete district-week records. Admissions were organized by reporting district, not patient residence. A single pooled district-week Light Gradient Boosting Machine (LightGBM) generated district-specific forecasts that were summed to obtain national predictions. This bottom-up approach was compared with direct national LightGBM, a negative binomial generalized additive model, one-week persistence, and hierarchical forecasting benchmarks. Model development was temporally ordered: 2024 was used for validation and model selection, finalized specifications were refitted using all pre-2025 model-ready data, and models remained fixed throughout the 52-week 2025 evaluation. Because model specifications were revised during peer review after initial 2025 results were known, 2025 was held out from model fitting but is not treated as an entirely untouched test set. Bottom-up LightGBM achieved the lowest observed national point error in 2025 (MAE = 239.37; RMSE = 380.27; wMAPE = 12.10%; R² = 0.9625), compared with the direct national negative binomial generalized additive model (wMAPE = 14.85%), one-week persistence (15.65%), and direct national LightGBM (18.67%). At district-week level, improvement over persistence was modest (wMAPE = 28.34% vs. 29.16%). All three primary 95% circular block-bootstrap confidence intervals for the primary national wMAPE differences crossed zero, and adjusted predictive-accuracy tests did not establish statistical superiority. Hierarchical reconciliation did not reduce national point error below the unreconciled bottom-up forecast. SHapley Additive exPlanations (SHAP) identified recent admissions as the dominant predictive signal, while additional analyses showed geographic heterogeneity in forecast-error magnitude. The empirical 95% residual-based uncertainty interval covered 49 of 52 held-out weeks (94.23%). District-informed bottom-up forecasting achieved the lowest observed national point error while retaining subnational forecast information, but its advantage over strong comparators remained statistically uncertain. Prospective multi-season validation is required before routine early-warning use.

29 December 2025 in Engineering Reports
Verified
Journal Article
Open access
Md. Razu AhmedMd. Razu Ahmed
Md. Razu Ahmed
JMJannatul Mauya
Jannatul Mauya
SRShamim Reza
Shamim Reza
Ruhul AminRuhul Amin
Ruhul Amin
This study evaluates audio source separation with Fast Independent Component Analysis (FastICA) in a fully specified, reproducible pipeline and benchmarks it against Principal Component Analysis (PCA) and Nonnegative Matrix Factorization (NMF). Three conversational recordings collected at the Department of Statistics, Pabna University of Science and Technology were canonicalized to 48 kHz WAV 96.63 s each, converted to mono, trimmed at 25 dB, RMS‐normalized, and time‐aligned. Sources were mixed with a fixed 3 × 3 row‐normalized Gaussian matrix. FastICA used parallel updates with a logcosh nonlinearity and unit‐variance whitening; PCA served as a multichannel linear baseline; NMF operated on a mono short‐time Fourier transform with KL divergence and NNDSVDA initialization, followed by soft masking and inverse transform. Performance was computed with BSS Eval after best‐permutation and scale alignment and summarized over 10 runs as mean ± SD. FastICA achieved Signal‐to‐Distortion Ratio (SDR) 53.51 ± 0.07 dB, Signal‐to‐Interference Ratio (SIR) 53.52 ± 0.07 dB, and Signal‐to‐Artifact Ratio (SAR) 79.58 ± 0.00 dB, well above a 20 dB high‐quality SDR threshold. PCA yielded SDR 2.79 ± 0.00 dB, SIR 2.79 ± 0.00 dB, SAR 80.64 ± 0.00 dB; NMF produced SDR −2.26 ± 0.00 dB, SIR 0.41 ± 0.00 dB, SAR 4.80 ± 0.00 dB. Waveform and spectrogram visualizations, together with descriptive, high‐order, and entropy statistics, corroborate these outcomes. The results establish FastICA as an effective classical approach for audio source separation and provide a transparent reference pipeline for comparative studies using SDR, SIR, and SAR.