检索结果-内蒙古大学图书馆

IEEE International Conference on Consumer Electronics (ICCE)

作者： Andrea Mattia Garavagno Edoardo Ragusa Antonio Frisoli Paolo Gastaldo Department of Electrical Electronic Telecommunication Engineering and Naval Architecture DITEN University of Genoa Genoa Italy Department of Excellence in Robotics & AI Institute of Mechanical Intelligence Scuola Superiore Sant’Anna Pisa Italy

This document proposes a novel approach to hardware-aware neural architecture search (HW NAS) that considers the resources available on the computing platform running it, enabling its execution on various embedded devices. The presented HW NAS produces tiny convolutional neural networks (CNNs) targeting low-end microcontroller units (MCUs), typically involved in the Internet of Things (IoT) or wearable robotics, opening new use cases. A gateway could run it to tailor CNNs’ architecture on the acquired data without using external servers, ensuring privacy. The proposed technique achieves state-of-the-art results in the human-recognition tasks on the Visual Wake Word dataset, a standard TinyML benchmark, on several embedded devices.

关键词：

来源：评论

学校读者我要写书评

暂无评论

On the Use of Contrastive Learning for Standard-Plane Classification in Fetal Ultrasound Imaging

SSRN

引用

SSRN 2023年

作者： Migliorelli, Giovanna Fiorentino, Maria Chiara Cosmo, Mariachiara Di Villani, Francesca Pia Mancini, Adriano Moccia, Sara Department of Law Università degli Studi di Macerata Macerata Italy Department of Information Engineering Università Politecnica delle Marche Ancona Italy Department of Humanities Università degli Studi di Macerata Macerata Italy The BioRobotics Institute Department of Excellence in Robotics and AI Scuola Superiore Sant’Anna Pisa Italy

Background and Objective: To investigate the effectiveness of contrastive learning, in particular SimClr, in reducing the need for large annotated ultrasound (US) image datasets for fetal standard plane identification. Methods: We explore SimClr advantage in the cases of both low and high inter-class variability, considering at the same time how classification performance varies according to different amounts of labels used. This evaluation is performed by exploiting contrastive learn- ing through different training strategies. We apply both quantitative and qualitative analyses, us- ing standard metrics (F1-score, sensitivity, and precision), Class Activation Mapping (CAM), and t-Distributed Stochastic Neighbor Embedding (t-SNE). Results: When dealing with high inter-class variability classification tasks, contrastive learning does not bring a significant advantage;whereas it results to be relevant for low inter-class variability classification, specifically when initialized with ImageNet weights. Conclusions: Contrastive learning approaches are typically used when a large number of unlabeled data is available, which is not representative of US datasets. We proved that SimClr either as pre-training with backbone initialized via ImageNet weights or used in an end-to-end dual-task may impact positively the performance over standard transfer learning approaches, under a scenario in which the dataset is small and characterized by low inter-class variability. © 2023, The Authors. All rights reserved.

关键词： Large dataset

来源：评论

学校读者我要写书评

暂无评论

Mimicking Human Laryngeal Tissues for a Soft Artificial Prosthesis

Mimicking Human Laryngeal Tissues for a Soft Artificial Pros...

引用

Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

作者： Jasmine Pozzi Arianna Conte Lucrezia Lorenzon Martina Maselli Maria Raffaella Marchese Andrea Nacci Matteo Cianchetti Department of Excellence in Robotics & AI The BioRobotics Institute Scuola Superiore Sant’Anna Pisa Italy Dipartimento di Neuroscienze Unità Operativa Complessa di Otorinolaringoiatria Organi di Senso e Torace Fondazione Policlinico Universitario A. Gemelli IRCSS Rome Italy ENT Audiology and Phoniatrics Unit University Hospital of Pisa Pisa Italy

ISBN: (数字)9798350371499

ISBN: (纸本)9798350371505

The larynx is a vital organ, with three main functions: breathing, swallowing, and phonation. Nevertheless, in the case of advanced laryngeal carcinoma, the total laryngectomy is still the gold standard. This clinical practice causes several physical and social problems to the patient. An implantable artificial larynx, able to reproduce the natural organ functions, is still missing. In a previous work, we have identified the basic structures and movements of the larynx, obtaining a biomimetic design, composed of cartilages, and a thin membrane (the elastic cone) that integrates artificial vocal folds. In this work, we have reproduced the anatomical structures with materials whose mechanical properties match the ones of the natural counterparts, creating a full-scale benchtop prototype of the artificial larynx. Since quantitative data on the mechanical characteristics of the laryngeal cartilages and the elastic cone have never been reported before, this work contributes to the field of artificial organs, with the creation of artificial airway cartilages and mucous-associated membranes. Moreover, it represents a fundamental step towards the creation of an implantable total artificial larynx that will improve the physical and psychological health of laryngectomee patients.

关键词： Structural rings Young's modulus Prototypes Psychology Biological systems Larynx Artificial biological organs Biomembranes Safety Robots

来源：评论

学校读者我要写书评

暂无评论

A virtual suturing task: proof of concept for awareness in autonomous camera motion

A virtual suturing task: proof of concept for awareness in a...

引用

IEEE International Conference on Robotic Computing (IRC)

作者： Nicolò Pasini Andrea Mariani Adnan Munawar Elena De Momi Peter Kazanzides Department of Electronics Information and Bioengineering Politecnico di Milano Milano Italy The BioRobotics Institute and Department of Excellence in Robotics & AI Scuola Superiore Sant’Anna Pisa Italy Department of Computer Science Johns Hopkins University Baltimore Maryland

ISBN: (纸本)9781665472616

Robot-assisted Minimally Invasive Surgery (MIS) requires the surgeon to alternatively control both the surgical instruments and the endoscopic camera, or to leave this burden to an assistant. This increases the cognitive load and interrupts the workflow of the operation. Camera motion automation has been examined in the literature to mitigate these aspects, but still lacks situation awareness, a key factor for camera navigation enhancement. This paper presents the development of a phase-specific camera motion automation, implemented in Virtual Reality (VR) during a suturing task. A user study involving 10 users was carried out using the master console of the da Vinci Research Kit. Each subject performed the suturing task undergoing both the proposed autonomous camera motion and the traditional manual camera control. Results show that the proposed system can reduce operational time, decreasing both the user's mental and physical demand. Situational awareness is shown to be fundamental in exploiting the benefits introduced by camera motion automation.

关键词： Automation Minimally invasive surgery Navigation Instruments Robot vision systems Virtual reality Manuals

来源：评论

学校读者我要写书评

暂无评论

Biomechanics in"Sino-Italian Joint"

引用

Acta Mechanica Sinica 2021年第2期37卷 169-172页

作者： Jizeng Wang Xiqiao Feng Paolo Bisegna Antonio DeSimone Key Laboratory of Mechanics on Disaster and Environment in Western China The Ministry of EducationCollege of Civil Engineering and MechanicsLanzhou UniversityLanzhou 730000China Institute of Biomechanics and Medical Engineering Department of Engineering MechanicsTsinghua UniversityBeijing 100084China Department of Civil Engineering and Computer Science University of Rome Tor Vergata00133 RomeItaly Department of Excellence in Robotics and AI The Biorobotics InstituteScuola Superiore Sant'Anna56127 PisaItaly SISSA-Scuola Internazionale Superiore di Studi Avanzati 34136 TriesteItaly

As originally proposed by professor Alberto Corigliano of Politecnico di Milano and professor Jizeng Wang of Lanzhou University,the Chinese Society of Theoretical and Applied Mechanics(CSTAM)and the Italian Association for Theoretical and Applied Mechanics(aiMETA)signed a memorandum of understanding on strengthening exchange and cooperation between mechanics scholars of the two countries in January,2018,officially opening the activities of bilateral academic exchange.

关键词： Italian strengthening CSTA

来源：评论

学校读者我要写书评

暂无评论

Time-Predictable Acceleration of Deep Neural Networks on FPGA SoC Platforms 42

Time-Predictable Acceleration of Deep Neural Networks on FPG...

引用

42nd IEEE Real-Time Systems Symposium, RTSS 2021

作者： Restuccia, Francesco Biondi, Alessandro TeCIP Institute Scuola Superiore Sant’Anna Pisa Italy Department of Excellence in Robotics and AI Scuola Superiore Sant’Anna Pisa Italy

ISBN: (纸本)9781665428026

This work focuses on the time-predictable execution of Deep Neural Networks (DNNs) accelerated on FPGA System-on-Chips (SoCs). The modern DPU accelerator by Xilinx is considered. An extensive profiling campaign targeting the Zynq Ultrascale+ platform has been performed to study the execution behavior of the DPU when accelerating a set of state-of-the-art DNNs for Advanced Driver Assistance Systems (ADAS). Based on the profiling, an execution model is proposed and then used to derive a response-time analysis. A custom FPGA module named DICTAT is also proposed to improve the predictability of the acceleration of DNNs and tighten the analytical bounds. A rich set of experimental results based on both analytical bounds and measurements from the target platform is finally presented to assess the effectiveness and the performance of the proposed approach on ADAS applications. ©2021 IEEE

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Low-Frequency Ultrasound Triggers Drug Release from Perfluorocarbon Droplets at Low Intensities

SSRN

引用

SSRN 2025年

作者： Sirolli, Sofia Boroujeni, Faraz Amini Guachi-Guachi, Lorena Prentice, Paul Ricotti, Leonardo Cafarelli, Andrea The BioRobotics Institute Scuola Superiore Sant’Anna Piazza Martiri della Libertà 33 Pisa56127 Italy Department of Excellence in Robotics & AI Scuola Superiore Sant’Anna Piazza Martiri della Libertà 33 Pisa56127 Italy James Watt School of Engineering University of Glasgow G12 8QQ United Kingdom

Perfluorocarbon droplets represent a promising platform for ultrasound-triggered drug delivery. Their liquid core can vaporize upon the application of a pressure wave such as ultrasound, resulting in the controlled and targeted release of their drug cargo. These carriers are highly tunable, exhibit a sharp on-off behavior and have longer lifetimes compared to ultrasound-responsive microbubbles. However, despite increasing efforts in the field, achieving an acceptable balance between stability and a safe activation threshold remains challenging. Indeed, stable droplet fomulations are usually activated at very high ultrasound pressures, exceeding 1 MPa, raising safety concerns. In this study, we investigated the response of stable surfactant-shelled perfluoropentane droplets to low-intensity pulsed ultrasound. We demonstrated that a significant drug release can be triggered at a pressure as low as 100 kPa using a very low frequency (i.e., 38 kHz). Therefore, we focused on characterizing the behavior of this droplet formulation under 38 kHz ultrasound stimulation, optimizing a protocol able to achieve efficient drug release while adhering to safety guidelines, thus facilitating a future in vivo translation. We systematically examined how drug delivery efficiency varies with different stimulation parameters, including pressure, pulse repetition frequency and duty cycle. Additionally, high-speed camera imaging and ultrasound imaging were performed to further elucidate the response mechanisms of perfluorocarbon droplets to low-frequency ultrasound. © 2025, The Authors. All rights reserved.

关键词： High speed cameras

来源：评论

学校读者我要写书评

暂无评论

A non-invasive device for skin cancer diagnosis: first clinical evidence with spectroscopic data enhanced by machine learning algorithms

A non-invasive device for skin cancer diagnosis: first clini...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： V. Mainardi L. Carletti D. Tsiakmakis M. Dal Canto T. Melillo S. Noferi G. Bagnoni P. Rubegni G. Ciuti The BioRobotics Institute Scuola Superiore Sant’Anna Pisa Italy Department of Excellence in Robotics & AI Scuola Superiore Sant’Anna Pisa Italy Noze S.r.l. Pisa Italy Dermatological Department Spedali Riuniti Leghorn Italy Dermatological Department Senese Hospital Siena Italy Health Science Interdisciplinary Center Scuola Superiore Sant’Anna Pisa Italy

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

Skin cancer represents a significant global health concern, with melanoma alone accounting for thousands of deaths annually. Early diagnosis is crucial for improving survival rates and reducing healthcare costs. While traditional diagnostic approaches involve visual inspection followed by biopsy, emerging technologies offer less invasive options with improved precision. In this study, a novel non-invasive device was designed, developed, and validated to employ near-infrared reflectance spectroscopy for skin lesion analysis. Furthermore, this work presents a machine learning approach aimed at classifying different types of skin lesions, as well as a new sequential approach to distinguish benign from malignant lesions based on spectral data and exploring the impact of anamnestic features. The device was used in two independent hospitals in Italy to collect data from 69 patients in total, including various types of skin lesions, all of whom followed the standard protocol for screening and diagnosis intervention. The implemented model achieved a recall of 93.8% and an accuracy of 75% for melanoma and benign classification, and a recall of 100% and an accuracy of 98.6% in distinguishing non-melanoma cancer from benign lesions, demonstrating promising results for skin cancer diagnosis utilizing spectral and anamnestic data. In summary, this study contributes to the development of allied non-invasive diagnostic tools and underscores the potential of machine learning in dermatology using spectroscopic data.

关键词： Representation learning Visualization Spectroscopy Accuracy Dermatology Melanoma Skin Lesions Discrete cosine transforms Standards

来源：评论

学校读者我要写书评

暂无评论

On the QNX IPC: Assessing Predictability for Local and Distributed Real-Time Systems

On the QNX IPC: Assessing Predictability for Local and Distr...

引用

IEEE Symposium on Real-Time and Embedded Technology and Applications

作者： Matthias Becker Dakshina Dasari Daniel Casini KTH Royal Institute of Technology Stockholm Sweden Robert Bosch GmbH Germany Department of Excellence in Robotics & AI Scuola Superiore Sant’Anna TeCIP Institute Pisa Italy

With the advent of massively distributed applications such as those required by the IoT-to-Edge-to-Cloud compute continuum (i.e., automotive, smart agriculture, smart manufacturing, and more), real-time communication mechanisms allowing physically distributed nodes to seamlessly communicate as if they were running on the same host acquired noteworthy importance. To this end, the synchronous inter-process communication (IPC) mechanism provided by the QNX operating system (OS) is a promising candidate, as it allows using the application programming interface for communicating both on a single- and multi-node setting. Furthermore, it provides priority and partition inheritance mechanisms to improve predictability when working with the Adaptive Partitioning Scheduler (APS), a reservationbased scheduler provided by the QNX OS. This paper explores the behavior of the QNX synchronous message-passing (SyncMP) IPC with an extensive set of experiments, using them to formalize its behavior and model it from a real-time perspective. Then, it provides a response-time analysis for client-server applications based on the QNX SyncMP building upon self-suspending task theory. Finally, we evaluate the analysis on an application based on the WATERS 2019 Challenge by Bosch.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Actuator Capabilities Aware Limitation for TDPA Passivity Controller Action

Actuator Capabilities Aware Limitation for TDPA Passivity Co...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Francesco Porcini Alessandro Filippeschi Massimiliano Solazzi Carlo Alberto Avizzano Antonio Frisoli PERCRO Laboratory Scuola Superiore Sant'Anna IIM Institute Pisa Italy Department of Excellence in Robotics and AI Scuola Superiore Sant'Anna Pisa Italy

Haptic interaction often requires stabilizing controllers for safety. The Time-Domain Passivity Approach guarantees passivity (then stability) by observing and dissipating energy generated from active elements in a network. The dissipating action is performed by a Passivity Controller, whose action is commanded to the physically limited robot actuators. Thus, the controller stabilizing action should be in turn limited in order to command displayable references to the actuators. This problem is rarely taken into account in the literature and when it is, the limitation is neither directly related to the actuator power limits, nor to the robot's current configuration. The limits of the currently adopted strategies leave room for improvement. In this paper, a new strategy to limit the Passivity Controller action is proposed taking into account both the physical limits of the actuators and the robot configuration. This new strategy is experimentally tested against the classical one based on the sampling time. In the experiment, a human interacts with a virtual wall in a Virtual Environment through a haptic interface. The wall induces an unstable behavior passivated with the two limitation strategies. The results clearly state the benefits introduced by the proposed strategy in two relevant cases.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：