A5022772949- Brain Tumor Detection and Classification
- Advanced MRI Techniques and Applications
- Traumatic Brain Injury and Neurovascular Disturbances
- Domain Adaptation and Few-Shot Learning
- Neonatal and fetal brain pathology
- Glioma Diagnosis and Treatment
- Medical Image Segmentation Techniques
- Advanced Neural Network Applications
- MRI in cancer diagnosis
- S100 Proteins and Annexins
- Medical Imaging Techniques and Applications
- Cell Image Analysis Techniques
- Machine Learning in Materials Science
- Virus-based gene therapy research
- Prostate Cancer Treatment and Research
- Epilepsy research and treatment
- Intracranial Aneurysms: Treatment and Complications
- Molecular Biology Techniques and Applications
- Machine Learning and ELM
- Prostate Cancer Diagnosis and Treatment
- Cerebrovascular and Carotid Artery Diseases
- Image Processing Techniques and Applications
- Traumatic Brain Injury Research
- Advanced Neuroimaging Techniques and Applications
- Fetal and Pediatric Neurological Disorders
University of Eastern Finland
2019-2022
Sapienza University of Rome
2018-2022
Enrico Fermi Center for Study and Research
2018-2021
Finland University
2019
Accurate detection and quantification of unruptured intracranial aneurysms (UIAs) is important for rupture risk assessment to allow an informed treatment decision be made. Currently, 2D manual measures used assess UIAs on Time-of-Flight magnetic resonance angiographies (TOF-MRAs) lack 3D information there substantial inter-observer variability both aneurysm size growth. could helpful improve but are time-consuming would therefore benefit from a reliable automatic UIA segmentation method. The...
Skull-stripping and region segmentation are fundamental steps in preclinical magnetic resonance imaging (MRI) studies, these common procedures usually performed manually. We present Multi-task U-Net (MU-Net), a convolutional neural network designed to accomplish both tasks simultaneously. MU-Net achieved higher accuracy than state-of-the-art multi-atlas methods with an inference time of 0.35 s no pre-processing requirements. trained validated on 128 T2-weighted mouse MRI volumes as well the...
We present a fully convolutional neural network (ConvNet), named RatLesNetv2, for segmenting lesions in rodent magnetic resonance (MR) brain images. RatLesNetv2 architecture resembles an autoencoder and it incorporates residual blocks that facilitate its optimization. is trained end to on three-dimensional images requires no preprocessing. evaluated exceptionally large dataset composed of 916 T2-weighted rat MRI scans 671 rats at nine different lesion stages were used study focal cerebral...
Registration-based methods are commonly used in the automatic segmentation of magnetic resonance (MR) brain images. However, these not robust to presence gross pathologies that can alter anatomy and affect alignment atlas image with target image. In this work, we develop a algorithm, MU-Net-R, for normal injured rat hippocampus based on an ensemble U-net-like Convolutional Neural Networks (CNNs). MU-Net-R was trained manually segmented MR images sham-operated rats traumatic injury (TBI) by...
Abstract We present MedicDeepLabv3+, a convolutional neural network that is the first completely automatic method to segment cerebral hemispheres in magnetic resonance (MR) volumes of rats with ischemic lesions. MedicDeepLabv3+ improves state-of-the-art DeepLabv3+ an advanced decoder, incorporating spatial attention layers and additional skip connections that, as we show our experiments, lead more precise segmentations. requires no MR image preprocessing, such bias-field correction or...
Diffusion neuro-MRI has benefited significantly from sophisticated pre-processing procedures aimed at improving image quality and diagnostic. In this work, diffusion-weighted imaging (DWI) was used with artifact correction the apparent diffusion coefficient (ADC) quantified to investigate fetal brain development. The DWI protocol designed in order limit acquisition time estimate ADC without perfusion bias. normal brains compared cases isolated ventriculomegaly (VM), a common disease whose...
It is necessary to develop reliable biomarkers for epileptogenesis and cognitive impairment after traumatic brain injury when searching novel antiepileptogenic cognition-enhancing treatments. We hypothesized that a multiparametric magnetic resonance imaging (MRI) analysis along the septotemporal hippocampal axis could predict development of post-traumatic epilepsy impairment. performed quantitative T2 T2* MRIs at 2, 7 21 days, diffusion tensor days lateral fluid-percussion in male rats....
This study was undertaken to identify prognostic biomarkers for posttraumatic epileptogenesis derived from parameters related the hippocampal position and orientation.Data were two preclinical magnetic resonance imaging (MRI) follow-up studies: EPITARGET (156 rats) Epilepsy Bioinformatics Study Antiepileptogenic Therapy (EpiBioS4Rx; University of Eastern Finland cohort, 43 rats). Epileptogenesis induced with lateral fluid percussion-induced traumatic brain injury (TBI) in adult male Sprague...
Regular monitoring of the primary particles and purity profiles a drug product during development manufacturing processes is essential for manufacturers to avoid variability contamination. Transmission electron microscopy (TEM) imaging helps predict how changes affect particle characteristics virus-based gene therapy vector products intermediates. Since intact can characterize efficacious products, it beneficial automate detection adenovirus against non-intact-viral background mixed with...
Manual segmentation of rodent brain lesions from magnetic resonance images (MRIs) is an arduous, time-consuming and subjective task that highly important in pre-clinical research. Several automatic methods have been developed for different human MRI segmentation, but little research has targeted lesion segmentation. The existing tools performing rodents are constrained by strict assumptions about the data. Deep learning successfully used medical image However, there not any deep approach...
Abstract Skull-stripping and region segmentation are fundamental steps in preclinical magnetic resonance imaging (MRI) studies, these common procedures usually performed manually. We present Multi-task U-Net (MU-Net), a convolutional neural network designed to accomplish both tasks simultaneously. MU-Net achieved higher accuracy than state-of-the-art multi-atlas methods with an inference time of 0.35 seconds no pre-processing requirements. evaluated the performance our presence skip...
A bstract Registration-based methods are commonly used in the anatomical segmentation of magnetic resonance (MR) brain images. However, they sensitive to presence deforming pathologies that may interfere with alignment atlas image target image. Our goal was develop an algorithm for automated normal and injured rat hippocampus. We implemented using a U-Net-like Convolutional Neural Network (CNN). sham-operated experimental controls rats lateral-fluid-percussion induced traumatic injury (TBI)...
We present MedicDeepLabv3+, a convolutional neural network that is the first completely automatic method to segment cerebral hemispheres in magnetic resonance (MR) volumes of rats with lesions. MedicDeepLabv3+ improves state-of-the-art DeepLabv3+ an advanced decoder, incorporating spatial attention layers and additional skip connections that, as we show our experiments, lead more precise segmentations. requires no MR image preprocessing, such bias-field correction or registration template,...