Publication - Portfolio

Real-Time Fire Detection: Integrating Lightweight Deep Learning Models on Drones with Edge Computing

Author(s): Md Fahim Shahoriar Titu, Mahir Afser Pavel, Goh Kah Ong Michael, and Riasat Khan

Abstract: Fire accidents are life-threatening catastrophes leading to losses of life, financial damage, climate change, and ecological destruction. Promptly and efficiently detecting and extinguishing fires is essential to reduce the loss of lives and damage. This study uses...

Non-small cell lung cancer detection through knowledge distillation approach with teaching assistant

Author(s): Mahir Afser Pavel, Rafiul Islam, Shoyeb Bin Babor, Riaz Mehadi, and Riasat Khan

Abstract: Non-small cell lung cancer (NSCLC) exhibits a comparatively slower rate of metastasis in contrast to small cell lung cancer, contributing to approximately 85% of the global patient population. In this work ...

Fusion of Image Filtering and Knowledge-Distilled YOLO Models for Root Canal Failure Diagnosis

Author(s): Afifa Zain Apurba, Md Fahim Shahoriar Titu, Mahir Afser Pavel, Intisar Tahmid Naheen, and Riasat Khan

Abstract: Root canal treatment involves the removal of inflamed pulp from infected teeth and sealing them to prevent re-entry of bacteria and infection. Early and accurate identification of root canal issues is crucial for improving treatment outcomes and minimizing complications. This study proposes ...

Multi-Stage Knowledge Distillation with Layer Fusion-Based Deep Learning Approach for Skin Cancer Classification

Author(s): Mahir Afser Pavel, Ramisa Asad, Goh Kah Ong Michael, Md Ikramuzzaman, Murad Mustakim, and Riasat Khan

Abstract: Skin cancer is one of the most common types of cancer globally, caused by prolonged exposure to the sun's UV rays. Despite recent developments in research, early diagnosis, prevention, and treatment, skin cancer remains a significant health concern. This study proposes ...

Hybrid ViT-RetinaNet with Explainable Ensemble Learning for Fine-Grained Vehicle Damage Classification

Author(s): Ananya Saha, Mahir Afser Pavel, Afifa Zain Apurba, Md Fahim Shahoriar Titu, and Riasat Khan

Abstract: Efficient and explainable vehicle damage inspection is essential due to the increasing complexity and volume of vehicular incidents. Traditional manual inspection approaches are not time-effective, prone to human error, and lead to inefficiencies in insurance claims and repair workflows. Existing deep learning methods, such as CNNs, often struggle with generalization, require large annotated datasets, and lack interpretability. This study presents ...

Error-Tolerant Multimodal Vision-Language Models for Endodontic Triaging: A Cross-Sectional Study

Author(s): Md Fahim Shahoriar Titu, Mahir Afser Pavel, Afifa Zain Apurba, Saif Ahmed, Shafin Rahman, James Dudley, and Taseef Hasan Farook

Abstract: To assess the performance of multimodal vision-language artificial intelligence models, optimised using quantisation-aware training, in triaging endodontic treatment needs. The focus is ...

Bridging High-Resource to Low-Resource Language Gaps: Refining Clinical Outcome Prediction via Cross-Lingual Methods and Adaptive Translation Strategies with SHAP

Author(s): Mahir Afser Pavel, Rafiul Islam, Mohammad Junayed Hasan, and M. R. C. Mahdy

Abstract: English has more dominant and rich cultures in every NLP research than other low-resource languages. So, it lacks research on other low-resource languages like Bengali and remains underexplored in the NLP field, especially in deploying models, creating datasets, and other availabilities. The study introduces ...

Reseach papers and publications

Real-Time Fire Detection: Integrating Lightweight Deep Learning Models on Drones with Edge Computing

Non-small cell lung cancer detection through knowledge distillation approach with teaching assistant

Fusion of Image Filtering and Knowledge-Distilled YOLO Models for Root Canal Failure Diagnosis

Multi-Stage Knowledge Distillation with Layer Fusion-Based Deep Learning Approach for Skin Cancer Classification

Hybrid ViT-RetinaNet with Explainable Ensemble Learning for Fine-Grained Vehicle Damage Classification

Error-Tolerant Multimodal Vision-Language Models for Endodontic Triaging: A Cross-Sectional Study

Under Review

Bridging High-Resource to Low-Resource Language Gaps: Refining Clinical Outcome Prediction via Cross-Lingual Methods and Adaptive Translation Strategies with SHAP