Artificial Intelligence and Translational Theranostics

The AITT lab, in the Department of Nuclear Medicine at the University of Bern, is also affiliated with the ARTORG Center and the Center for Artificial Intelligence in Medicine at the same university. The lab consists of an interdisciplinary team comprising informaticians, physicians, physicists, and biologists. As shown in the above sketch, the AITT lab's research focuses on four main methodologies: biological modeling, quantification, imaging physics, and experimental platforms, all of which are supported by artificial intelligence (AI). The goal of these methodological developments is to optimize clinical practice in medical imaging, particularly in nuclear medicine, including early diagnosis of cancer and neurodegenerative disorders, treatment planning for radiopharmaceutical therapy, survival prediction, and personalized treatment. The ongoing projects are listed below.

Imaging physics:
•    Reconstruction for total-body PET
•    Ultra-low dose PET imaging including denoising and CT-free attenuation correction and scatter correction
•    Enhancement of dosimetry SPECT imaging
•    Imaging systems for on-chip and intravital imaging

Computational biology:
•    Digital twin by combining physiologically based pharmacokinetic modelling (PBPK) and histology-driven reaction-diffusion simulation
•    Computational modelling for the optimization of tandem radiopharmaceutical therapy with different radioisotopes
•    Computational modelling for the optimization of combination of radiopharmaceutical therapy and PARP inhibitor therapy

Quantification:
•    Parametric imaging & parametric imaging reconstruction for total-body PET
•    Pretherapy dosimetry prediction for radiopharmaceutical therapy
•    Radiobiological modelling for radiopharmaceutical therapy
•    Metabolic connectome
•    Lesion detection, segmentation and characterization

Experimental platforms:
•    On-chip theranostics
•    Intravital multimodal theranostic imaging

For clinical translation, it specifically focuses on the following topics:
•    Early diagnosis of pancreatic cancer
•    Early diagnosis and differential diagnosis of neurodegenerative diseases
•    Early diagnosis of coronary artery disease
•    Treatment planning and personalized therapy for radiopharmaceutical therapy
•    Personalized treatment of lymphoma
 

Researcher Portrait (Center for Artificial Intelligence in Medicine)

Acknowledgement of funding agencies