The Quantitative and Computational Cancer Imaging (QACCI) team is dedicated to developing advanced multi-parametric MRI protocols integrated with machine learning and deep learning techniques. Their goal is to generate sensitive and specific imaging biomarkers for cancer detection and diagnosis. By combining imaging and computational approaches, QACCI aims to model tumor- and patient-specific pathological and genomic profiles. Recognizing the limitations of current computational methods, which often lack generalizability and clinical applicability, the team focuses on creating clinically viable, computer-aided tools to support personalized treatment and surgical planning. This work is conducted within a multidisciplinary framework involving surgeons, oncologists, pathologists, radiologists, MR technologists, physicists, and computational engineers. QACCI’s mission is to advance the clinical adoption of AI-driven tools, empowering healthcare providers to make more confident and individualized decisions in cancer care.
- Design and development of multi-parametric MRI protocols for cancer diagnosis, treatment monitoring, and planning
- Patient-specific imaging solutions for brain tumors, head and neck, breast, prostate, liver, and gynecological cancers
- Quantitative analysis using AI-driven techniques, integrating both big and small data for precision medicine
Glioma Brain Tumors:
- Development of a novel DWI/DSC-MRI fusion technique for accurate quantification of glioma subregions
- Differentiation of tumorous regions in glioblastoma multiforme using multi-parametric MRI (DWI/PWI/T2/MRS) and multivariate classification methods
- Determination of target volumes in glioma for conformal radiotherapy and IMRT using Diffusion Tensor Imaging (DTI)
- Quantitative analysis of glioma response to chemoradiotherapy using functional diffusion MRI biomarkers
- Assessment of emerging MRI techniques (T2-relaxometry, IVIM) for biopsy-validated tumor differentiation in glioma
Breast Cancer:
- Segmentation and classification of breast tumors using quantitative DCE-MRI
- Clinical methodology for multi-parametric MRI quantification in breast cancer
- Comparative study of DWI-MRI vs. PET/CT for detecting bone marrow metastasis in pelvic breast cancer
Ovarian and Gynecological Imaging:
- Non-rigid registration algorithm for DCE-MRI in the ovary region based on pharmacokinetic parameters (listed twice – kept once)
- Investigation of the diagnostic role of quantitative DWI and T2-weighted MRI in complex ovarian masses
Prostate and Pelvic Imaging:
- Multiparametric MRI-based grading of patients suspected of prostate cancer, with ROI quantification and histopathological correlation
- Comparative evaluation of prostate mpMRI with pathological outcomes in a clinical imaging center in Iran
Head & Neck Imaging:
- Enhanced detection and differentiation of parotid tumors using multiparametric analysis of perfusion and diffusion MRI