Mahmuda Sultana Mimi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Mahmuda Sultana Mimi has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Subasish Das
Ingram School of Engineering, Texas State University, San Marcos, USA
12 shared publications - Monzurul Islam
Texas State University
6 shared publications - Rohit Chakraborty
Ingram School of Engineering, Texas State University
5 shared publications - Anannya Ghosh Tusti
Texas State University
4 shared publications - Gaurab Chhetri
Texas State University, San Marcos, USA
3 shared publications - Shriyank Somvanshi
Ingram School of Engineering, Texas State University, San Marcos, TX
3 shared publications
Latest Publications
- 16 August 2026
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
Crash Risk Analysis of Non-Motorists using Interpretable Tabular Deep Learning
The study investigates injury severity patterns among non-motorists (e.g., pedestrians and bicyclists) in Illinois using interpretable machine learning models applied to crash data from 2021 to 2023. The dataset included variables such as roadway design, environmental context, crash characteristics, and driver behavior. To address class imbalance in injury severity labels, synthetic oversampling was employed. Feature selection was performed using Random Forest and XGBoost, followed by training TabNet and FT-Transformer models. TabNet achieved the highest accuracy under the balanced training configuration and showed the strongest performance on the original imbalanced dataset across precision, recall, and F1 score, while also providing intrinsic interpretability through sequential attention-based feature selection. SHAP analysis revealed that temporal and environmental factors significantly influenced early model decisions, with behavioral and infrastructure-related variables impacting subsequent stages. The findings emphasize optimizing lighting, enhancing traffic control, and addressing time-specific crash risks. The study underscores the potential of interpretable tabular deep learning in traffic safety analytics, advocating for integration with contextual, geospatial, and behavioral data for targeted interventions.
- 16 August 2026
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
Crash Risk Analysis of Non-Motorists using Interpretable Tabular Deep Learning
The study investigates injury severity patterns among non-motorists (e.g., pedestrians and bicyclists) in Illinois using interpretable machine learning models applied to crash data from 2021 to 2023. The dataset included variables such as roadway design, environmental context, crash characteristics, and driver behavior. To address class imbalance in injury severity labels, synthetic oversampling was employed. Feature selection was performed using Random Forest and XGBoost, followed by training TabNet and FT-Transformer models. TabNet achieved the highest accuracy under the balanced training configuration and showed the strongest performance on the original imbalanced dataset across precision, recall, and F1 score, while also providing intrinsic interpretability through sequential attention-based feature selection. SHAP analysis revealed that temporal and environmental factors significantly influenced early model decisions, with behavioral and infrastructure-related variables impacting subsequent stages. The findings emphasize optimizing lighting, enhancing traffic control, and addressing time-specific crash risks. The study underscores the potential of interpretable tabular deep learning in traffic safety analytics, advocating for integration with contextual, geospatial, and behavioral data for targeted interventions.
- Open Access
- 25 June 2026
- Published by Elsevier in SSRN Electronic Journal
Interaction Breakdowns Between Older Drivers and Automated-Driving Features: A Mixed-Methods HCI Analysis
Older drivers face significant challenges in traffic environments due to age-related cognitive,sensory, and motor impairments. These difficulties are amplified by interactions with vehicleautomation systems, where reduced awareness, inattention, and overreliance on automationintroduce cooperative breakdowns. This study analyzes 451 crashes involving older adults using 2024 Texas CRIS data, categorized by driver age, crash severity, and SAE automation level. A mixed-methods approach combining Cluster Correspondence Analysis (CCA) with narrative review reveals patterns of human-automation interaction failures across SAE Levels 1–3. From an HCI perspective, these failures reflect mismatches in shared control expectations and limited support for dynamic role negotiation, especially in time-critical contexts such as intersections or lane changes. Findings show older drivers struggle not only with driving tasks but also with cooperative alignment during critical transitions. The study frames automation as a socio-technical system, calling for human-centered strategies that address age-specific vulnerabilities in safety-critical contexts.
- 25 June 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessInteraction Breakdowns Between Older Drivers and Automated-Driving Features: A Mixed-Methods HCI Analysis
Older drivers face significant challenges in traffic environments due to age-related cognitive,sensory, and motor impairments. These difficulties are amplified by interactions with vehicleautomation systems, where reduced awareness, inattention, and overreliance on automationintroduce cooperative breakdowns. This study analyzes 451 crashes involving older adults using 2024 Texas CRIS data, categorized by driver age, crash severity, and SAE automation level. A mixed-methods approach combining Cluster Correspondence Analysis (CCA) with narrative review reveals patterns of human-automation interaction failures across SAE Levels 1–3. From an HCI perspective, these failures reflect mismatches in shared control expectations and limited support for dynamic role negotiation, especially in time-critical contexts such as intersections or lane changes. Findings show older drivers struggle not only with driving tasks but also with cooperative alignment during critical transitions. The study frames automation as a socio-technical system, calling for human-centered strategies that address age-specific vulnerabilities in safety-critical contexts.
- 25 May 2026
- Published by Elsevier in Transport Policy
- Vol. 185
Classifying crash severity by SAE automation level using transformer models: Insights for automated vehicle policy
- 25 May 2026
- Published by Elsevier in Transport Policy
- Vol. 185
Classifying crash severity by SAE automation level using transformer models: Insights for automated vehicle policy
- 7 January 2026
- Published by Taylor & Francis in International Journal of Urban Sciences
- p. 1-30
Non-spatial AI modeling to estimate traffic volume measures on local roadways
- 7 January 2026
- Published by Taylor & Francis in International Journal of Urban Sciences
- p. 1-30
Non-spatial AI modeling to estimate traffic volume measures on local roadways
- 2 January 2026
- Published by Elsevier in Accident Analysis & Prevention
- Vol. 227, 108382
Uncovering latent structures of crash typology in narcotic-involved fatal crashes for safe system interventions
- 2 January 2026
- Published by Elsevier in Accident Analysis & Prevention
- Vol. 227, 108382
Uncovering latent structures of crash typology in narcotic-involved fatal crashes for safe system interventions
- Top Cited PapersOpen Access
- 24 December 2025
- Published by Association for Computing Machinery (ACM) in ACM Computing Surveys
- Vol. 58 (7) , 1-33
From Tiny Machine Learning to Tiny Deep Learning: A Survey
The rapid growth of edge devices has driven the demand for deploying artificial intelligence (AI) at the edge, giving rise to Tiny Machine Learning (TinyML) and its evolving counterpart, Tiny Deep Learning (TinyDL). While TinyML initially focused on enabling simple inference tasks on microcontrollers, the emergence of TinyDL marks a paradigm shift toward deploying deep learning models on severely resource-constrained hardware. This survey presents a comprehensive overview of the transition from TinyML to TinyDL, encompassing architectural innovations, hardware platforms, model optimization techniques, and software toolchains. We analyze state-of-the-art methods in quantization, pruning, and neural architecture search (NAS), and examine hardware trends from MCUs to dedicated neural accelerators. Furthermore, we categorize software deployment frameworks, compilers, and AutoML tools enabling practical on-device learning. Applications across domains such as computer vision, audio recognition, healthcare, and industrial monitoring are reviewed to illustrate the real-world impact of TinyDL. Finally, we identify emerging directions including neuromorphic computing, federated TinyDL, edge-native foundation models, and domain-specific co-design approaches. This survey aims to serve as a foundational resource for researchers and practitioners, offering a holistic view of the ecosystem and laying the groundwork for future advancements in edge AI.
- 24 December 2025
- Published by Association for Computing Machinery (ACM) in ACM Computing Surveys
- Vol. 58 (7) , 1-33
Top Cited PapersOpen AccessFrom Tiny Machine Learning to Tiny Deep Learning: A Survey
The rapid growth of edge devices has driven the demand for deploying artificial intelligence (AI) at the edge, giving rise to Tiny Machine Learning (TinyML) and its evolving counterpart, Tiny Deep Learning (TinyDL). While TinyML initially focused on enabling simple inference tasks on microcontrollers, the emergence of TinyDL marks a paradigm shift toward deploying deep learning models on severely resource-constrained hardware. This survey presents a comprehensive overview of the transition from TinyML to TinyDL, encompassing architectural innovations, hardware platforms, model optimization techniques, and software toolchains. We analyze state-of-the-art methods in quantization, pruning, and neural architecture search (NAS), and examine hardware trends from MCUs to dedicated neural accelerators. Furthermore, we categorize software deployment frameworks, compilers, and AutoML tools enabling practical on-device learning. Applications across domains such as computer vision, audio recognition, healthcare, and industrial monitoring are reviewed to illustrate the real-world impact of TinyDL. Finally, we identify emerging directions including neuromorphic computing, federated TinyDL, edge-native foundation models, and domain-specific co-design approaches. This survey aims to serve as a foundational resource for researchers and practitioners, offering a holistic view of the ecosystem and laying the groundwork for future advancements in edge AI.
- Open Access
- 2 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5-14
ST-GraphNet: A Spatio-Temporal Graph Neural Network for Understanding and Predicting Automated Vehicle Crash Severity
- 2 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5-14
Open AccessST-GraphNet: A Spatio-Temporal Graph Neural Network for Understanding and Predicting Automated Vehicle Crash Severity
- 18 October 2025
- Published by Elsevier in Journal of Safety Research
- Vol. 95, 301-317
Crash risk patterns among older bicyclists: Insights from hybrid XGBoost-Cluster Correspondence Analysis
- 18 October 2025
- Published by Elsevier in Journal of Safety Research
- Vol. 95, 301-317
Crash risk patterns among older bicyclists: Insights from hybrid XGBoost-Cluster Correspondence Analysis