Automated analysis of small intestinal lamina propria to distinguish normal, celiac disease, and non-celiac biopsy images

Join ARU's Medical Technology Research Centre (MTRC) online for a seminar from Oliver Faust about recent advancements in the automated diagnosis of celiac disease.

Celiac Disease (CD) poses a significant public health challenge due to its high prevalence and the absence of a cure.

Diagnosis relies on identifying characteristic mucosal alterations in the small intestine, but distinguishing CD from other conditions can be challenging. Recent advancements in automated diagnostic support systems offer promise for early detection and improved patient outcomes.

In this study, Oliver and his team explored information extraction methods tailored for endoscopy, video capsule endoscopy (VCE), and biopsy image analyses to differentiate CD from non-CD conditions.

Statistical and non-linear methods emerged as crucial tools for enhancing clinical workflows by reducing variability among observers. However, bias introduced during system design may limit the generalizability of performance results, underscoring the need for comprehensive understanding and mitigating design biases.

The researchers compared and reviewed articles and discussed recommendations to advance CD diagnosis, highlighting the transition towards advanced deep learning techniques. Their proposed Automated Feature Vector Selection-based Ensemble ResNet (AFVSER) leverages transfer learning to automate CD detection with high accuracy, surpassing other pretrained models.

AFVSER demonstrates potential to reduce diagnostic costs and improve diagnosis quality, paving the way for next-generation CD detection systems. This comprehensive approach aims to enhance the accuracy, efficiency, and accessibility of CD diagnosis, addressing a critical need in healthcare.

Oliver has worked at the forefront of artificial intelligence in biomedical engineering. He is an international expert for medical decision support systems, artificial intelligence, the Internet of Medical Things, and signal processing, substantiated by his considerable publication record which includes 100 impact factor journal papers. He is a founding member of the Transformative Artificial Intelligence Applications research cluster at ARU.

Oliver holds two doctoral degrees (Doctor of Philosophy in electronic engineering from the University of Aberdeen; Doctor of Engineering in biomedical science from Chiba University, Japan). He's a founding faculty member of Habib University in Karachi, Pakistan, and assisted the start-up of Altreonic Inc, a software company in Belgium.

He collaborates with more than 20 distinguished biomedical scientists worldwide. Previously, he was a visiting scholar at Tianjin University, Zhejiang University, and University of Electronic Science and Technology of China.

For more information, email [email protected]