Advanced detectors for PET/MRI
We have developed a novel MRI compatible positron emission tomography (PET) detector, based on detection of prompt Cherenkov light, that can achieve a twofold improvement in time-of-flight (TOF) measurement or significant reduction in price, compared to current state-of-the-art commercial scanners [COBISS.SI-ID: 28934695, 33032743]. Our activities dedicated to advancement of PET imaging also included collaboration within the EU joint programme for neurodegenerative disease research [JPND – PETMETPAT], where we have implemented harmonization of FDG PET brain imaging in neurology, specifically for Parkinson’s disease research. To enable and support our research we have established a image database for storing and distributing the PET images collected at University Medical Center Ljubljana. This work is included in an international initiative NIX (http://nix-alliance.org) connecting imaging centers from USA, Australia and Slovenia.
We continuously adapt latest techniques from the field of biomedical optics into experimental measurement systems and apply the in (pre-)clinical research. Using computational modeling and precise monitoring systems, we developed improved procedures for diagnostic and therapeutic use of lasers, e.g. hyperthermic laser lipolysis [COBISS.SI-ID: 3128420]. We have developed a non-invasive, cost effective multi-spectral imaging technique for early rheumatoid arthritis detection [COBISS.SI-ID: 28864551, 3386980]. The diagnostic power of the imaging systems is also increased by our in-hose developed profilometry, measuring precise shape of objects imaged [COBISS.SI-ID: 3297124, 16367899]. With the developed imaging systems and our expertise we collaborate in international projects dedicated to development of imaging technologies for the study of works of art, development real-time fluorescence lifetime acquisition system [H2020 777222] and EU COST actions for Skin Cancer Detection using Laser Imaging [BM 1205] and advancement of electromagnetic hyperthermic medical technologies [CA17115]. In this field we developed a fruitful collaboration with industrial partner Fotona d.o.o., developing medical laser sources and procedures [COBISS.SI-ID: 32005927, 14476059, 14723867].
Personalized radiation therapy
We have advanced personalization of radiation therapy by studying novel treatment plans, e.g. dynamic conformal arc therapy for shorter treatment time of lung cancer patients [COBISS.SI-ID: 2709883], and volumetric modulated arc therapy for reducing the dosimetric impact of positional errors in craniospinal irradiation [COBISS.SI-ID: 2365051]. More precise dose delivery is also possible using a novel method for determination of correction factors for photon beams in radiotherapy [COBISS.SI-ID: 3097211]. We are also leading or contributing to three IAEA projects dedicated to improvements in radiotherapy (»Improving Radiotherapy Practices for Advanced Radiotherapy Technologies Including Quality Assurance and Quality Control«, »Strengthening Knowledge of Radiation Therapy Professionals (Radiation Oncologist, Medical Physicists and Radiation Therapy Technologists)«, and »Improving Safety and Quality of Radiology Services through the Development of Medical Physics Departments and Enhancing the Theranostic Nuclear Medicine Approach«).
Biomarkers of neurodegenerative brain disorders
We studied the metabolic brain patterns characteristic for neurodegnerative brain disorders using FDG-PET imaging. Parkinson’s disease related pattern (PDRP) was found to differentiate well between patients with Parkinson’s disease, healthy volunteers and patients with atypical parkinsonian syndromes, and proved PDRP to be highly reproducible across different image reconstruction algorithms [COBISS.SI-ID: 5051052], contributing to wider use of such biomarker. In dementia with lewy bodies, we identified brain regions whose impairment contributes to clinical features [COBISS.SI-ID: 5881260]. We explored the role of genetics in dopaminergic treatment of Parkinson’s disease, finding that inflammation and oxidative stress gene variabilities do not play a major role in the occurrence of PD and the adverse events to such treatment [COBISS.SI-ID: 34211545]. In a study of the quality of life outcomes in Parkinson’s disease patients undergoing different therapies we identified distinct effect profiles of bilateral subthalamic stimulation, apomorphine, and intrajejunal levodopa infusion and highlighted the importance of holistic symptoms assessments to personalize the treatment choices [COBISS.SI-ID: 5670060]. With our research on biomarkers of neurodegenerative brain disorders we participated in an IAEA project and a Duoglobe international research.
In a clinical trial we assesed the efficacy and safety of ultrasonographically (US)-guided high-intensity focused ultrasound (HIFU) ablation for treatment of benign solid thyroid nodules, and fount it to be an efective and safe [COBISS.SI-ID: 2244524]. As part of an European project EUthyroid, we studied the methods and shown the importance of monitoring the iodine supply for prevention of iodine deficiency disorders [COBISS.SI-ID: 2887340, 3919020]. We also contributed to the regulatory approval of novel diagnostic radiopharmaceutial for personalized diagnosis and therapy of patients of medullary thyroid carcinoma [COBISS.SI-ID: 3004844], working within the EU project GRAN-T-MTC. In a clinical study we found 18F-fluorocholine PET/CT imaging to be effective for preoperative localization of hyperfunctioning parathyroid tissue [COBISS.SI-ID: 1796524, 3354028].
Advanced tissue modeling in biomedical optics
We have mastered the modeling of optical transport trough tissues and applied it to development of novel optical techniques and therapies. A most precise model of human finger to date was developed from high resolution in-vivo MRI scan and used to guide the design of a multi-sprectral imaging system for early detection of rheumatoid arthritis [COBISS.SI-ID: 3180900]. To develop an optical technique for non-invasive analysis of human skin structure and composition, a diffuse approximation model was compared to a golden standard Monte Carlo method, and found to be suitable for monitoring physiological changes in the skin [COBISS.SI-ID: 32196391].
Advanced modeling of tumor growth and response to therapies
Computational modeling of specific tumor types and individual patient data enabled us to advance personalization of cancer treatment, by predicting the response of the disease to a certain treatment. We developed and benchmarked a model simulating the dynamics of metastatic cancer cells that are drug-sensitive and drug-resistant with imaging metrics extracted from prostate cancer patients, demonstrating the need to model specific lesions besides the specificities relating to individual patients [COBISS.SI-ID: 3294564]. Another model studied was developed to predict tumor response to anti-PD-1 immunotherapy. Good agreement with experimental data was demonstrated, and major histocompatibility complex class I (MHC class I) was identified as a biomarker of response to immunotherapy. [COBISS.SI-ID: 3287652].