Our skin – a window on health, if we can just see through it!
Optoacoustic imaging provides a window into what is happening within tissues in a way a microscope can’t. It enables three-dimensional, high-resolution optical imaging up to 1 cm inside tissues. INNODERM, an earlier project, showed that raster scan optoacoustic mesoscopy(opens in new window) (RSOM) can image previously invisible pathophysiological, oxygenation and morphological features of the skin at depths of 1-5 mm. This could benefit the millions of sufferers of skin conditions, as currently, in dermatology, the gold standard for many diseases is the visual inspection of the skin. This is restricted to the external side of the skin and dependent on the clinician’s visual assessment. However, the severity of the disease, or the effects of the therapy, might only be seen in deeper skin layers, which can be visualised and analysed with RSOM. So, the WINTHER(opens in new window) project built on the achievements of INNODERM to make RSOM an even more powerful clinical tool. “Using the skin as a window for disease assessment, aided by modern computational algorithms and artificial intelligence (AI), WINTHER aims to improve the clinical accuracy of RSOM and broaden its clinical use to assess the progression and therapy in cardiovascular diseases and diabetes,” explains Vasilis Ntziachristos(opens in new window), project coordinator and chair of biological imaging at the Center for Translational Cancer Research(opens in new window) of the Technical University of Munich(opens in new window) (TUM). “While INNODERM focused on dermatological diseases, we expanded the applications of optoacoustic mesoscopy to cardiometabolic conditions in WINTHER, and, in parallel, we improved the methodology and hardware to make RSOM much faster than its earlier versions,” adds Ntziachristos.
Fast RSOM combined with AI for next-generation diagnostics
Fast RSOM (F-RSOM) enables the imaging of dynamic processes in the skin, such as blood flow and oxygenation changes. This is essential for assessing endothelial function, and monitoring therapy efficacy in cardiovascular diseases and diabetes. The ability to capture these changes in real time allows for more accurate and timely diagnosis and treatment decisions. Using skin as a window into the body, the images generated are then interpreted by a modern computation framework, based on deterministic and AI algorithms. These improvements were aimed at enhancing the speed, accuracy and clinical utility of RSOM. The hardware improvements in F-RSOM include the development of ultra-broadband ultrasound transducers for high-resolution imaging, the introduction of novel scanning protocols for dynamic control of the scanned area, and the development of a modern computation framework to enhance the processing and analysis of F-RSOM images. The project, supported also by iThera Medical(opens in new window), developed deep learning models trained to recognise and segment different features in the F-RSOM images.
New applications for F-RSOM
Ntziachristos is proud of the way WINTHER has shifted the paradigm in therapeutic monitoring of diabetes, cardio-metabolic diseases and inflammatory skin conditions. “Indeed, in our paper(opens in new window), published in ‘Nature Biomedical Engineering’, RSOM demonstrated the ability to quantify diabetes effects on skin features and then use them to detect diabetes progression, from a simple skin scan. “This new ability could significantly enhance diabetes monitoring, addressing a critical need for efficient management and prevention strategies in a global epidemic affecting over 500 million people.”
Keywords
WINTHER, RSOM, F-RSOM, endothelial function, AI, diagnostics, artificial intelligence, AI algorithms, ultra-broadband ultrasound transducers