A5011367557- Muscle activation and electromyography studies
- Radiomics and Machine Learning in Medical Imaging
- Quality and Safety in Healthcare
- AI in cancer detection
- Balance, Gait, and Falls Prevention
- Cardiovascular Health and Disease Prevention
- Phonocardiography and Auscultation Techniques
- Gait Recognition and Analysis
- Cerebrovascular and Carotid Artery Diseases
- Biomedical and Engineering Education
- Advanced Sensor and Energy Harvesting Materials
- Prostate Cancer Diagnosis and Treatment
- Non-Invasive Vital Sign Monitoring
- Medical Image Segmentation Techniques
- Advanced X-ray and CT Imaging
- Hernia repair and management
- Healthcare Technology and Patient Monitoring
- Context-Aware Activity Recognition Systems
- Medical Imaging and Analysis
- Artificial Intelligence in Healthcare
- Colorectal Cancer Surgical Treatments
- Quality Function Deployment in Product Design
- Cerebral Palsy and Movement Disorders
- EEG and Brain-Computer Interfaces
- Muscle Physiology and Disorders
Polytechnic University of Turin
2016-2025
Consorzio di Bioingegneria e Informatica Medica
2022
Cairo University
2015-2017
Università degli Studi della Tuscia
1996
Gait abnormalities can be studied by means of instrumented gait analysis. Foot-switches are useful to study the foot-floor contact and for timing phases in many disorders, provided that a reliable foot-switch signal may collected. Considering long walks allows reducing intra-subject variability, but requires automatic user-independent methods analyze large number cycles. The aim this work is describe validate an algorithm segmentation classification performance was assessed comparing its...
In recent years, there is a growing interest in Human Activity Recognition (HAR) systems applied healthcare. A HAR system essentially made of wearable device equipped with set sensors (like accelerometers, gyroscopes, magnetometers, heart-rate sensors, etc...) and classifier able to recognize the activity performed. this study we focused on choice classifier, since isn't unique consolidated methodology for HAR. The main aim compare performances 5 classifiers, based machine learning....
Human Activity Recognition (HAR) refers to an emerging area of interest for medical, military, and security applications. However, the identification features be used activity classification recognition is still open point. The aim this study was compare two different feature sets HAR. Particularly, we compared a set including time, frequency, time-frequency domain widely in literature (FeatSet_A) with time-domain derived by considering physical meaning acquired signals (FeatSet_B)....
Surface electromyography (sEMG) is the main non-invasive tool used to record electrical activity of muscles during dynamic tasks. In clinical gait analysis, a number techniques have been developed obtain and interpret muscle activation patterns patients showing altered locomotion. However, body knowledge described in these studies very seldom translated into routine practice. The aim this work analyze critically key factors limiting extensive use powerful among clinicians. A thorough...
Preventive maintenance is a core function of clinical engineering, and it essential to guarantee the correct functioning equipment. The management control activities are equally important perform maintenance. As variety medical equipment increases, accordingly size need for better become essential. This paper aims develop new model preventive priority using quality deployment as concept in We developed three-domain framework consisting requirement, function, concept. requirement domain house...
Goal: Artificial intelligence applied to medical image analysis has been extensively used develop non-invasive diagnostic and prognostic signatures. However, these imaging biomarkers should be largely validated on multi-center datasets prove their robustness before they can introduced into clinical practice. The main challenge is represented by the great unavoidable variability which usually addressed using different pre-processing techniques including spatial, intensity feature...
Abstract Radiomics-based systems could improve the management of oncological patients by supporting cancer diagnosis, treatment planning, and response assessment. However, one main limitations these is generalizability reproducibility results when they are applied to images acquired in different hospitals scanners. Normalization has been introduced mitigate this issue, two approaches have proposed: rescales image intensities ( normalization ), other feature distributions for each center )....
Overtaking relies heavily on the driver’s attention and cognitive state, illegal overtaking can lead to accidents, severe injuries, or fatalities. To enhance highway safety, we propose a method for accurately detecting continuous road lanes. We used dashboard-mounted smartphone cameras geolocation data filter analysis areas. state-of-the-art deep learning model You Only Look Once version 8 (YOLOv8) detect yellow When these lanes suggest potential overtaking, apply YOLO Panoptic driving...
Head-worn inertial sensors represent a valuable option to characterize gait in real-world conditions, thanks the integration with glasses and hearing aids. Few methods based on head-worn allow for stride-by-stride speed estimation, but none has been developed data collected settings. This study aimed at validating two-steps machine learning method estimate initial contacts using single sensor attached temporal region. A convolutional network is used detect strides. Then, inferred from...
Abstract The aim of our study was to develop and validate a machine learning algorithm predict response individual HER2‐amplified colorectal cancer liver metastases (lmCRC) undergoing dual HER2‐targeted therapy. Twenty‐four radiomics features were extracted after 3D manual segmentation 141 lmCRC on pretreatment portal CT scans cohort including 38 patients; feature selection then performed using genetic algorithms. classified as nonresponders (R−), if their largest diameter increased more...
The home monitoring of patients affected by chronic heart failure (CHF) is key importance in preventing acute episodes. Nevertheless, no wearable technological solution exists to date. A possibility could be offered Cardiac Time Intervals extracted from simultaneous recordings electrocardiographic (ECG) and phonocardiographic (PCG) signals. the recording a good-quality PCG signal requires accurate positioning stethoscope over chest, which unfeasible for naïve user as patient. In this work,...
ABSTRACT Noninvasive Artificial Intelligence (AI) techniques have shown great potential in assisting clinicians through the analysis of medical images. However, significant challenges remain integrating these systems into clinical practice due to variability data across multi‐center databases and lack clear implementation guidelines. These issues hinder ability achieve robust, reproducible, statistically results. This study thoroughly analyzes several decision‐making steps involved managing...
Gait asymmetry is typically evaluated using spatio-temporal or joint kinematics parameters. Only a few studies addressed the problem of defining an index directly based on muscle activity, extracting parameters from surface electromyography (sEMG) signals. Moreover, no used extraction principal activations (activations that are necessary for accomplishing specific motor task) as base to construct index, less affected by variability sEMG patterns. The aim this paper define robust...
The effects of epoch duration, time windowing, and an algorithm upon the estimates spectral parameters surface myoelectric signals are discussed. Simulation with computer-generated shows that: (a) rectangular window linear interpolation between contiguous values acceptable for durations as short 0.25 s; (b) raw periodogram 1-2 s, coefficients variation near 3-4% f/sub mean/ 4-6% med/ should be expected; (c) epochs s these double, shorter become meaningless because high variance. These...
Considers muscle activation and fatigue during propulsion of a racing wheelchair. The authors discuss single case study on the modality To investigate coordination, fatigue, strategies that subject exploits to counteract effects keep high level performance, first studied intervals seven trunk upper limb muscles then evolution in time their spectral content by means two different approaches, according statistical properties myoelectric signals. Angle, angular velocity, acceleration elbow...
The aim of the study is to present a new Convolutional Neural Network (CNN) based system for automatic segmentation colorectal cancer. algorithm implemented consists several steps: pre-processing normalize and highlights tumoral area, classification on CNNs, post-processing aimed at reducing false positive elements. performed using three CNNs: each them classifies same regions interest acquired from different MR sequences. final mask obtained by majority voting. Performances were evaluated...
Discretization is an important step introduced in the field of Knowledge Discovery Databases to better represent knowledge domain and increase learning speed performance Data Reduction Mining algorithms. However, no studies evaluated benefits introducing a discretization into more complex system. In this study we seek evaluate how ChiMerge method could improve CAD system for automatic detection prostate cancer (PCa) based on multi-parametric Magnetic Resonance (mp-MR) imaging. 16...
The aim of this study is to present a personalized predictive model (PPM) with machine learning (ML) system that able identify and classify patients suspected prostate cancer (PCa) following mpMRI. We extracted all the who underwent fusion biopsy (FB) from March 2014 December 2019, while August 2020 April 2021 were included as validation set. proposed was based on four ML methods: fuzzy inference (FIS), support vector (SVM), k-nearest neighbors (KNN), self-organizing maps (SOMs). Then, logic...
Aim of this paper is to develop an automated system for the classification and characterization carotid wall status a robust based on local texture descriptors. A database 200 longitudinal ultrasound images artery used. One-hundred with Intima-Media Thickness (IMT) value higher than 0.8[Formula: see text]mm are considered as high risk. Six different rectangular pixel neighborhoods were considered: four areas centered selected element, sizes [Formula: text], text] pixels, two noncentered...