Intelligent imaging for precision medicine

Abstract

Ultrasound localization microscopy is a novel method to generate high-resolution images of the microvasculature. For this, ultrasound contrast agents consisting of gas-filled microbubbles are used. These microbubbles can be localized as single objects in ultrasound images with high precision. By tracking them over time, super-resolved images of the microvasculature can be constructed. The method is particularly interesting for the classification and therapy monitoring of breast cancer but can be applied to any clinical question in need of information on the microvascular structure. In this contribution the method will be described and problems that arise when translating it from small animal studies to clinical imaging will be discussed together with potential solutions. Particularly, , optimal spatial sampling for localization, patient motion compensation, and the extension to the third dimension with two-dimensional matrix transducers will be covered.

Biography

Georg Schmitz (Senior Member, IEEE) received the Dipl.-Ing. degree in 1990 and the Dr.-Ing. degree in 1995 in Electrical Engineering and Information Technology from Ruhr-Universität Bochum, Germany.

From 1995 to 2001 he was with Philips Research Laboratories in Hamburg and Aachen as a senior and Principal Scientist, working in medical image and signal processing. From 2001 to 2004 he was Professor for Medical Engineering at the University of Applied Science Koblenz. Since 2004 he has been Professor for Electrical Engineering and holds the chair for Medical Engineering at Ruhr-Universität Bochum, Germany.

His research is on ultrasound imaging with a special focus on image reconstruction and statistical signal processing. Current topics are ultrasound superresolution vascular imaging by localization microscopy using contrast agents aa well as signal processing and image reconstruction with artificial neural networks

Abstract

Breast cancer is a major threat to women’s health. In the past decade, quantitative contrast-enhanced ultrasound has shown promise for the diagnosis of prostate cancer by probing angiogenesis, a hallmark of most solid tumors. More recently, we have translated these concepts for the detection of breast cancer. Dedicated spatiotemporal analysis of ultrasound-contrast-agent kinetics will be presented that can identify the typical heterogeneity of breast cancer (micro)vasculature.

Biography

Ultrasound localization microscopy is a novel method to generate high-resolution images of the microvasculature. For this, ultrasound contrast agents consisting of gas-filled microbubbles are used. These microbubbles can be localized as single objects in ultrasound images with high precision. By tracking them over time, super-resolved images of the microvasculature can be constructed. The method is particularly interesting for the classification and therapy monitoring of breast cancer but can be applied to any clinical question in need of information on the microvascular structure.

Abstract

Artificial intelligence (AI) is gaining ground in breast cancer detection, a major challenge in healthcare. Early detection is crucial for successful treatment and survival. Although population based breast cancer screening programs have been installed in Europe, one in eight cancers is missed during screening. Moreover, a shortage of radiologists prevents the implementation of screening programs worldwide. Screenpoint Medical, located in Nijmegen, developed the CE and FDA approved deep-learning AI solution Transpara for breast cancer detection in 2D and 3D mammography. Screenpoint has a strong focus on clinical evidence. Clinical studies have shown that Transpara reaches the performance level of a radiologist, enabling higher accuracy of breast cancer screening and early detection. An overview of Transpara’s clinical evidence will be presented as well as an outlook towards the future of AI in order to face the challenges in breast cancer.

Biography

Anne-Kathrin studied psychology at Hamburg University in Germany followed by a master's programme in Neuroscience at Goldsmiths University of London, where she focused on methods and technics in neuroscience, specifically on medical imaging (MRI). She received a DAAD scholarship and came to the Netherlands in 2016 to pursuit a PhD at the Donders Institute at Radboud University. During her PhD she investigated how to stratify neural biotypes of anxiety based on MRI. In 2019, she worked as a visiting researcher at Western University in Canada. After her PhD, she joined Screenpoint Medical in the role of clinical scientist in 2021. In her daily work, she supports clinical studies on the implementation of AI in medical imaging for cancer detection in breast cancer screening and works on the validation of Screenpoint’s AI product Transpara in order to provide evidence-based AI to the clinical market.