A5019913633- Radiomics and Machine Learning in Medical Imaging
- Coronary Interventions and Diagnostics
- Aortic aneurysm repair treatments
- Dental Radiography and Imaging
- Facial Rejuvenation and Surgery Techniques
- Nasal Surgery and Airway Studies
- Non-Invasive Vital Sign Monitoring
- Obstructive Sleep Apnea Research
- Sleep and Wakefulness Research
- Medical Imaging and Analysis
- AI in cancer detection
- Orthodontics and Dentofacial Orthopedics
- Reconstructive Facial Surgery Techniques
- Vascular Procedures and Complications
- Artificial Intelligence in Healthcare and Education
- Cerebrovascular and Carotid Artery Diseases
Politecnico di Milano
2023
University of Milan
2021
Azienda Ospedaliera di Desio e Vimercate
2019
Background and purpose: Accurate instance segmentation of teeth in panoramic dental X-rays is a challenging task due to variations tooth morphology overlapping regions. In this study, we propose new algorithm, for instance, the different X-rays. Methods: An model was trained using architecture Mask Region-based Convolutional Neural Network (Mask-RCNN). The data training, validation, testing were taken from Tuft database (1000 radiographs). number predicted label 52 (20 deciduous 32...
Evidence is rapidly accumulating that multifactorial nocturnal monitoring, through the coupling of wearable devices and deep learning, may be disruptive for early diagnosis assessment sleep disorders. In this work, optical, differential air-pressure acceleration signals, acquired by a chest-worn sensor, are elaborated into five somnographic-like which then used to feed network. This addresses three-fold classification problem predict overall signal quality (normal, corrupted), three...
The actual role of landmarks labeling before three-dimensional (3D) facial acquisition is still debated. In this study, several measurements were compared among textured labeled (TL), unlabeled (NL), and untextured (NTL) 3D models.The face 50 subjects was acquired through stereophotogrammetry. Landmark coordinates extracted from TL, NL, NTL models, 33 linear angular calculated, together with surface area volume. Accuracy models assessed calculation relative technical error measurement...
The aim of this study was to explore the potential intraprocedural benefits Proximal Side Optimization (PSO) technique by Optical Coherence Tomography (OCT).A case series 10 consecutive true bifurcation lesions, with severe long pathology side branch (SB), were randomly assigned be treated standard DK Crush procedure (non-PSO group) as compared in PSO modification (PSO group). data from OCT investigation before crushing SB Drug-Eluting Stent (DES), after crushing, first kissing balloon...
The aim of the study was to evaluate use narrow implants (3.25 mm) replace upper lateral and lower incisors according an established immediate function protocol.A total 49 (Neoss Proactive 3.25 mm implants, Neoss Ltd, Harrogate, UK) in 35 patients were evaluated.Thirty-one placed mandible 18 maxilla.The mean insertion torque measured Ncm.Thirty-six fresh extraction sockets.Implant stability measurements performed at baseline, after 2, 4, 6 weeks 3, months using resonance frequency analysis...
In simple bifurcation lesions provisional single stent strategy remains the standard of care. While complex bifurcations, defined based on 1) Side Branch (SB) lesion length > 10 mm and 2) SB ostial diameter stenosis >70% are approached with a 2-DES upfront. The Crush techniques which composed classic Crush, mini-Crush double kissing (DK-Crush) share core principle protruding DES within Main (MB) to minimize risk restenosis, most prevalent etiology failure during 2-stent approach in...