IHImran Hossain

Imran Hossain

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Imran Hossain has a total of 37 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    SE–UNet: Spatial and Channel Refinement for Robust Multi–Anatomical Structure Segmentation in Abdominal Ultrasound

    by Md. Rashed, Muhammad Jamil, Md. Imran Hossain, Adnan Kavak, Md. Riad Hassan, Özgür Çakır, Hossein Fotouhi

    Automated segmentation of multiple anatomical structures in abdominal ultrasound could enhance clinical decision–making by providing consistent quantitative measurements for disease monitoring. However, ultrasound–specific artifacts including speckle noise, low contrast and weak boundaries challenge existing deep learning models. We propose SE–UNet, a novel architecture that integrates spatial attention gates and squeeze–and–excitation blocks to improve boundary localization and tissue discrimination. On a multi–anatomical dataset combining kidney and spleen ultrasound images (six tissue classes), SE–UNet achieved a mean Dice score of 0.7165, outperforming nine baseline methods. Clinical validation with an expert radiologist on 50 cases demonstrated 0.78 Dice agreement with manual annotations, with 88% of segmentations deemed clinically acceptable for measurement purposes. The model processes images in 0.028 seconds, supporting real–time clinical workflow integration. Our approach demonstrates how combined spatial and channel refinement addresses key ultrasound segmentation challenges with direct clinical applicability.

  • Vision transformer based damage assessment from post-disaster satellite imagery: an applied study on hurricane harvey

    by Jakaria Habib, Imran Hossain, Basim Al Zabir Shammo, Taz Warul Mulk, Estiyak Rubaiat

  • Machine Learning-Based Hepatitis C Detection Using Ensemble Boosting

    by Basim Al Zabir Shammo, Lamia Islam, Mehedi Hasan, Anwar Hossain, Imran Hossain, Akash Mia, Moumita Haque Mohona

  • Open Access

    Vision Transformer based Damage Assessment from Post-Disaster Satellite Imagery: An Applied Study on Hurricane Harvey

    by Jakaria Habib, Imran Hossain, Basim Al Zabir Shammo, Taz Warul Mulk, Estiyak Rubaiat

    The timely and accurate evaluation of building damage is vital for post-disaster response. This study evaluates the efficacy of Vision Transformers (ViT-B32) compared to a baseline Convolutional Neural Network (EfficientNet-B0) for the binary classification of damaged versus undamaged buildings, utilizing 128×128 RGB satellite imagery from Hurricane Harvey. As opposed to CNNs that encode local features, ViT structures rely on self-attention mechanisms for capturing global spatial relationships, a property that is crucial in spotting intricate damage patterns in visually noisy disaster sites. Experiments show that the ViT model outperforms the CNN baseline in terms of classification accuracy (97.85% vs 96.90%) and prove to be more robust on unbalanced data as shown with F1-score and AUC than respective values of close competitors. Furthermore, the study highlights the model's interpretability: we generate attention heatmaps that visualize the specific image regions driving the classification decisions. These visualizations provide actionable insights by precisely localizing structural damage, thereby offering a valuable tool for prioritizing recovery efforts in disaster management workflows.

  • Open Access

    Trustworthy Deep Feature Extraction and Ensemble‐Based Machine Learning Approach for Breast Cancer Detections

    by Md. Rashed, Mohammad Kamrul Hasan, Md. Imran Hossain, Md. Sarwar Hosain, Abdul Hadi Abd Rahman, Shayla Islam

    Breast cancer (BC) has become a major public health concern and is critically associated with the highest global death rate for cancer detection. The diagnosis process and the techniques remain complex and often influenced by the diagnostician's background, which makes it challenging. Despite advancements in BC detection, existing methods cannot often effectively combine interpretability and high accuracy in complex imaging data, limiting their clinical applicability. This study proposes a reliable strategy for detecting BC by combining deep learning (DL) strengths with ensemble‐based machine learning (ML) techniques. ML models offer interpretability and generalisation, while DL enhances the ability to learn and uncover hidden patterns in complicated BC images. The pre‐trained model is used in the proposed technique for effective feature extraction, followed by applying eight different ML models to identify BC. The performance of the study is evaluated in terms of precision, recall, F1‐score, and confusion matrices for all classifiers. In addition, ROC curves are drawn for each classifier. Our rigorous experimentation yields compelling results that demonstrate exceptional performance compared with those of existing state‐of‐the‐art models. We achieve a higher accuracy rate of 97.50%, a precision of 97.15%, a recall of 97.00%, and an F1‐score of 96.98%. Furthermore, we determine that the support vector classifier is the most effective ML model when integrated with the pre‐trained VGG‐16 architecture. The strategies, exhaustive performance analysis, and reliable assessment presented in this research provide valuable advances in BC detection, helping doctors make better decisions, offering better patient care, and improving BC outcomes.

  • Open Access

    Hybrid Machine Learning Models for Accurate Type 2 Diabetes Mellitus Prediction Using a Stacking Classifier and a Meta-Model Approach

    by Md. Rashed, Imran Hossain, Akif Mahdi, Gulam Mustofa

    Type 2 diabetes mellitus (T2DM) is a major global health issue that significantly reduces life expectancy and impairs overall quality of life. Early diagnosis of T2DM is critical, as it can help prevent or delay the development of associated complications. This study aims to assess the effectiveness of a stacked ensemble machine learning model, incorporating a meta-model approach, for the early detection of T2DM. In the first stage of modeling (Level 1), two base models - random forest (RF) and support vector machines (SVM) - are trained using distinct feature selection, tuning, and optimization strategies. In the second stage, ensemble methods, such as voting and stacking classifiers, are employed to combine the predictions from these base models. The final prediction is made by a meta-model, specifically a gradient boosting classifier (GBC), which classifies individuals as either positive or negative for T2DM. Further classification of positive instances is based on low-, moderate-, or high-risk levels. The system is designed for deployment to enable real-time assessment of diabetes risk. The Meta-Model (GBC), with an accuracy of 99.13%, a precision of 100%, specificity of 100%, and an F1 score of 99.25%, consistently outperforms the base models (SVM and RF) and ensemble models (voting and stacking classifiers), proving highly effective in the early detection of T2DM. This significant accuracy highlights the enormous promise of machine learning methods for assisting in the early identification of T2DM. The findings of this study will significantly enhance precision medicine screening for T2DM, empowering healthcare professionals to diagnose the condition early and ultimately improve patient outcomes.


  • Open Access

    Internet of Things-based Home Automation with Network Mapper and MQTT Protocol

    by Tahsin Alam, Rokonuzzaman, Sohag Sarker, A F M Zainul Abadin, Tarun Debnath, Imran Hossain


  • Open Access

    An Advanced Deep Learning Based Three-Stream Hybrid Model for Dynamic Hand Gesture Recognition

    by Abdur Rahim, Abu Saleh Musa Miah, Hemel Sharker Akash, Jungpil Shin, Imran Hossain, Najmul Hossain

    In the modern context, hand gesture recognition has emerged as a focal point. This is due to its wide range of applications, which include comprehending sign language, factories, hands-free devices, and guiding robots. Many researchers have attempted to develop more effective techniques for recognizing these hand gestures. However, there are challenges like dataset limitations, variations in hand forms, external environments, and inconsistent lighting conditions. To address these challenges, we proposed a novel three-stream hybrid model that combines RGB pixel and skeleton-based features to recognize hand gestures. In the procedure, we preprocessed the dataset, including augmentation, to make rotation, translation, and scaling independent systems. We employed a three-stream hybrid model to extract the multi-feature fusion using the power of the deep learning module. In the first stream, we extracted the initial feature using the pre-trained Imagenet module and then enhanced this feature by using a multi-layer of the GRU and LSTM modules. In the second stream, we extracted the initial feature with the pre-trained ReseNet module and enhanced it with the various combinations of the GRU and LSTM modules. In the third stream, we extracted the hand pose key points using the media pipe and then enhanced them using the stacked LSTM to produce the hierarchical feature. After that, we concatenated the three features to produce the final. Finally, we employed a classification module to produce the probabilistic map to generate predicted output. We mainly produced a powerful feature vector by taking advantage of the pixel-based deep learning feature and pos-estimation-based stacked deep learning feature, including a pre-trained model with a scratched deep learning model for unequalled gesture detection capabilities.


  • GVA: guided visual attention approach for automatic image caption generation

    by Bipul Hossen, Zhongfu Ye, Amr Abdussalam, Imran Hossain


  • Open Access

    Single-channel Speech Separation Based on Double-density Dual-tree CWT and SNMF

    by Imran Hossain, Abdur Rahim, Najmul Hossain

    Speech is essential to human communication; therefore, distinguishing it from noise is crucial. Speech separation becomes challenging in real-world circumstances with background noise and overlapping speech. Moreover, the speech separation using short-term Fourier transform (STFT) and discrete wavelet transform (DWT) addresses time and frequency resolution and time-variation issues, respectively. To solve the above issues, a new speech separation technique is presented based on the double-density dual-tree complex wavelet transform (DDDTCWT) and sparse non-negative matrix factorization (SNMF). The signal is separated into high-pass and low-pass frequency components using DDDTCWT wavelet decomposition. For this analysis, we only considered the low-pass frequency components and zeroed out the high-pass ones. Subsequently, the STFT is then applied to each sub-band signal to generate a complex spectrogram. Therefore, we have used SNMF to factorize the joint form of magnitude and the absolute value of real and imaginary (RI) components that decompose the basis and weight matrices. Most researchers enhance the magnitude spectra only, ignore the phase spectra, and estimate the separated speech using noisy phase. As a result, some noise components are present in the estimated speech results. We are dealing with the signal's magnitude as well as the RI components and estimating the phase of the RI parts. Finally, separated speech signals can be achieved using the inverse STFT (ISTFT) and the inverse DDDTCWT (IDDDTCWT). Separation performance is improved for estimating the phase component and the shift-invariant, better direction selectivity, and scheme freedom properties of DDDTCWT. The speech separation efficiency of the proposed algorithm outperforms performance by 6.53