A5073182931- Radiomics and Machine Learning in Medical Imaging
- Neuroblastoma Research and Treatments
- MRI in cancer diagnosis
- Elasticity and Material Modeling
- Cancer, Hypoxia, and Metabolism
- AI in cancer detection
- Cell Image Analysis Techniques
- Ultrasound Imaging and Elastography
- Cardiac electrophysiology and arrhythmias
- Cellular Mechanics and Interactions
- Endometrial and Cervical Cancer Treatments
- Collagen: Extraction and Characterization
- Ultrasound and Hyperthermia Applications
- Prostate Cancer Diagnosis and Treatment
- Medical Imaging Techniques and Applications
- Ovarian cancer diagnosis and treatment
- Digital Imaging for Blood Diseases
- Advanced Neuroimaging Techniques and Applications
- Advanced MRI Techniques and Applications
- Image Processing Techniques and Applications
- Breast Cancer Treatment Studies
- Cutaneous Melanoma Detection and Management
- Bioinformatics and Genomic Networks
- Diamond and Carbon-based Materials Research
- Cancer Cells and Metastasis
Agostino Gemelli University Polyclinic
2023-2024
Institute of Cancer Research
2017-2024
Istituti di Ricovero e Cura a Carattere Scientifico
2023-2024
Royal Marsden NHS Foundation Trust
2017-2019
PurposeManual recontouring of targets and Organs At Risk (OARs) is a time-consuming operator-dependent task. We explored the potential Generative Adversarial Networks (GAN) to auto-segment rectum, bladder femoral heads on 0.35T MRIs accelerate online MRI-guided-Radiotherapy (MRIgRT) workflow.Methods3D planning from 60 prostate cancer patients treated with MR-Linac were collected. A 3D GAN architecture its equivalent 2D version trained, validated tested 40, 10 respectively. The volumetric...
Increased stiffness in the extracellular matrix (ECM) contributes to tumor progression and metastasis. Therefore, stromal modulating therapies accompanying biomarkers are being developed target ECM stiffness. Magnetic resonance (MR) elastography can noninvasively quantitatively map viscoelastic properties of tumors
High computational cost associated with digital pathology image analysis approaches is a challenge towards their translation in routine clinic. Here, we propose computationally efficient framework (SuperHistopath), designed to map global context features reflecting the rich tumor morphological heterogeneity. SuperHistopath efficiently combines i) segmentation approach using linear iterative clustering (SLIC) superpixels algorithm applied directly on whole-slide images at low resolution (5x...
To use deep learning to improve the image quality of subsampled images (number acquisitions = 1 [NOA1]) reduce whole-body diffusion-weighted MRI (WBDWI) acquisition times.Both retrospective and prospective patient groups were used develop a learning-based denoising filter (DNIF) model. For initial model training validation, 17 patients with metastatic prostate cancer acquired WBDWI NOA1 NOA9 (acquisition period, 2015-2017) retrospectively included. An additional 22 advanced cancer, myeloma,...
Computational pathology-based cell classification algorithms are revolutionizing the study of tumor microenvironment and can provide novel predictive/prognosis biomarkers crucial for delivery precision oncology. Current used on hematoxylin eosin slides based individual nuclei morphology with limited local context features. Here, we propose a multi-resolution hierarchical framework (SuperCRF) inspired by way pathologists perceive regional tissue architecture to improve demonstrate its...
Abstract Noninvasive early indicators of treatment response are crucial to the successful delivery precision medicine in children with cancer. Neuroblastoma is a common solid tumor young that arises from anomalies neural crest development. Therapeutic approaches aiming destabilize MYCN protein, such as small-molecule inhibitors Aurora A and mTOR, currently being evaluated phase clinical trials high-risk MYCN-driven disease, limited ability evaluate conventional pharmacodynamic biomarkers...
Hyaluronan (HA) is a key component of the dense extracellular matrix in breast cancer, and its accumulation associated with poor prognosis metastasis. Pegvorhyaluronidase alfa (PEGPH20) enzymatically degrades HA can enhance drug delivery treatment response preclinical tumour models. Clinical development stromal-targeted therapies would be accelerated by imaging biomarkers that inform on therapeutic efficacy vivo. Here, PEGPH20 was assessed multiparametric magnetic resonance (MRI) three...
Abstract Childhood neuroblastoma is a hypervascular tumor of neural origin, for which antiangiogenic drugs are currently being evaluated; however, predictive biomarkers treatment response, crucial successful delivery precision therapeutics, lacking. We describe an MRI-pathologic cross-correlative approach using intrinsic susceptibility (IS) and contrast (SC) MRI to noninvasively map the vascular phenotype in Th-MYCN transgenic mice treated with endothelial growth factor receptor inhibitor...
Diffusion-weighted magnetic resonance imaging (DWI) enables non-invasive, quantitative staging of prostate cancer via measurement the apparent diffusion coefficient (ADC) water within tissues. In cancer, more advanced disease is often characterized by higher cellular density (cellularity), which generally accepted to correspond a lower measured ADC. A relationship between tissue structure and in vivo measurements ADC has yet be determined for cancer. this study, we establish theoretical...
Motivation: Whole-body diffusion-weighted imaging in oncology assumes isotropic diffusion therefore that data probing the directional dependence of body cancer applications is scarce. Goal(s): To report distribution fractional anisotropy (FA) values metastatic bone disease for a cohort patients who underwent multi-directional weighted imaging. Approach: Deep learning was used to auto-segment lesions; FA are calculated each segmented region and distributions reported. Results: from 85 with...
Motivation: This study seeks to refine diagnostic precision for endometrial cancer, addressing the need non-invasive biomarkers improve preoperative assessment and risk stratification. Goal(s): To evaluate feasibility of synthetic MRI using T1-, T2- proton density(PD) mapping characterization, reflecting underlying tumor biology. Approach: A pilot with twenty patients a 2D fast-spin-echo multi-saturation-delay multi-echo sequence acquire data, analysis T1, T2, PD metrics compared...
To use deep learning to calculate the uncertainty in apparent diffusion coefficient (σADC) voxel-wise measurements clinically impact monitoring of treatment response and improve quality ADC maps.We a uniquely designed diffusion-weighted imaging (DWI) acquisition protocol that provides gold-standard σADC train model on two separate cohorts: 16 patients with prostate cancer 28 mesothelioma. Our network was trained novel cost function, which incorporates perception metric b-value regularisation...
<p>Table S1: MRI acquisition parameters, Supplementary results including Table S2 summarising the subregional analysis, S3: confusion matrix for cell classification, Figure S1 showing tumor volumes at time of enrolment, individual therapy-induced changes in volume, S3 validation our algorithm automatic classification apoptotic cells from HE-stained slides digitised images with cleaved caspase-3 imunohistochemistry, S4 imunohistochemistry.</p>
<div>Abstract<p>Noninvasive early indicators of treatment response are crucial to the successful delivery precision medicine in children with cancer. Neuroblastoma is a common solid tumor young that arises from anomalies neural crest development. Therapeutic approaches aiming destabilize <i>MYCN</i> protein, such as small-molecule inhibitors Aurora A and mTOR, currently being evaluated phase clinical trials high-risk <i>MYCN</i>-driven disease, limited...
<div>Abstract<p>Increased stiffness in the extracellular matrix (ECM) contributes to tumor progression and metastasis. Therefore, stromal modulating therapies accompanying biomarkers are being developed target ECM stiffness. Magnetic resonance (MR) elastography can noninvasively quantitatively map viscoelastic properties of tumors <i>in vivo</i> thus has clear clinical applications. Herein, we used MR elastography, coupled with computational histopathology,...
<div>Abstract<p>Noninvasive early indicators of treatment response are crucial to the successful delivery precision medicine in children with cancer. Neuroblastoma is a common solid tumor young that arises from anomalies neural crest development. Therapeutic approaches aiming destabilize <i>MYCN</i> protein, such as small-molecule inhibitors Aurora A and mTOR, currently being evaluated phase clinical trials high-risk <i>MYCN</i>-driven disease, limited...