Mazhar Hussain Successfully Defended his Doctoral Thesis on Multi‑Sensor Data Fusion

Mazhar’s research addresses one of today’s most pressing challenges in data science—how to effectively combine information from diverse sensors to enhance estimation and prediction in real-world applications. His work delves into advanced techniques for analyzing and integrating multimodal data, with a particular focus on Artificial intelligence based data fusion methods using machine learning and deep learning-based data fusion methods.

As part of his thesis, Mazhar explored three distinct application areas:

• Predicting the viscosity of molten glass through analysis of geometric changes using Analytics based data fusion,,
• Estimating fuel consumption in city buses via machine learning techniques,
• Detecting and quantifying hazardous gases such as hydrogen sulfide (H₂S) and methyl mercaptan (CH₃SH) using deep learning models.

The results demonstrate that the choice of data fusion strategy, whether traditional, machine learning-based, or deep learning-based, should be guided by the nature and quality of the data and the specific application context. His work also includes a comprehensive review of state-of-the-art deep learning fusion strategies, emphasizing their growing role in domains like environmental monitoring, smart cities, robotics, and healthcare analytics.

Prof. Mattias O'Nils, Prof. Jan Lundgren, Prof. Niclas Björsell, Prof. Tomas Nordström, Associate Prof. Göran Thungström, Dr. Mazhar Hussain and Prof. Annalisa Liccardo on the screen. 

Professor Jan Lundgren served as the main supervisor and chair, with Professor Mattias O'Nils as co-supervisor. The opponent was Professor Niclas Björsell from the University of Gävle. The examination committee included Professor Annalisa Liccardo, University of Naples Federico II, Italy, Professor Tomas Nordström, Umeå University and Associate Professor Göran Thungström from Mid Sweden University.

Mazhar’s research contributes valuable insights into the potential of multi-sensor data fusion to solve complex real-world problems and supports continued innovation in AI and sensor-driven systems.

Read the thesis in Diva