Nowadays, healthcare systems collect and provide most medical data in digital form. The availability of medical data enables a large number of artificial intelligence applications, and there is a growing interest in the quantitative analysis of clinical images using techniques such as Positron Emission Tomography, Computerized Tomography, and Magnetic Resonance Imaging, mainly applied to texture analysis and radiomics. In particular, thanks to machine and deep learning, researchers can generate insights to improve the discovery of new therapeutic tools, support diagnostic decisions, aid in the rehabilitation process, etc. However, the increasing amount of available data may lead to a more significant effort to make a diagnosis. Moreover, this task is even more challenging due to the high inter/intra patient variability, the availability of various imaging techniques, and the need to consider data from multiple sensors and sources, which brought to the well-known domain shift issue. 

To address the problems described, radiologists and pathologists today use tools to assist them in analyzing biomedical images. They are known as Computer-Aided Diagnosis (CAD) systems, and they allow to mitigate or eliminate the difficulties due to inter- and intra-observer variability, represented by various assessments of the same region, under the same assumptions, by the same physician at different times, and various assessments of the same region by several physicians, thanks to appropriate algorithms. 

Further relevant issues are data access, which may be delayed or even prevented for various reasons, such as privacy, security and intellectual property, and representativeness of the captured sample compared to the real population. For these reasons, researchers have recently explored the use of synthetic data, both for training the models and to estimate and teach systems in situations that have not been observed in actual reality. 

This workshop aims to provide an overview of recent advances in the field of biomedical image processing using machine learning, deep learning, artificial intelligence, and radiomics features, placing particular attention on contributions dealing with practical applications, for example potential alternative solution against domain shift, or exploiting synthetic images to teach actual CAD systems. In particular, the ultimate goal is to analyse how these techniques can be employed in the typical medical image processing workflow, from image acquisition to classification, including retrieval, disease detection, prediction, and classification. 

Topics

The workshop calls for submissions addressing, but not limited to, the following topics: 

• Biomedical Image Processing

  • Machine and Deep Learning techniques for image analysis (i.e., segmentation of cells, tissues, organs, lesions; classification of cells, diseases, tumours, etc.)

  • Image Registration Techniques

  • Image Preprocessing Techniques

  • Image-based 3D reconstruction

  • Computer-Aided Detection and Diagnosis Systems (CADs)

• Biomedical Image Analysis

  • Radiomics and Artificial intelligence for personalised medicine

  • Machine and Deep Learning as tools to support medical diagnoses and decisions

  • Image retrieval (e.g., context-based retrieval, lesion similarity)

• CAD architectures

  • Advanced architecture for biomedical image remote processing, elaboration and transmission

  • 3D Vision, Virtual, Augmented and Mixed Reality application for remote surgery

  • Image processing techniques for privacy-preserving AI in medicine.

Albert Comelli <acomelli@fondazionerimed.com>

Cecilia Di Ruberto <dirubert@unica.it>

Andrea Loddo <andrea.loddo@unica.it> 

Lorenzo Putzu <lorenzo.putzu@unica.it> 

Alessandro Stefano <alessandro.stefano@ibfm.cnr.it>