How to ensure safety of a moving target: an industry perspective on algorithm change protocols for learning medical AI Algorithms (Preprint)

2021 ◽  
Author(s):  
Stephen Gilbert ◽  
Matthew Fenech ◽  
Martin Hirsch ◽  
Shubhanan Upadhyay ◽  
Andrea Biasiucci ◽  
...  
Keyword(s):  
Real Time ◽  
Quality Management Systems ◽  
Learning From Data ◽  
Balanced Approach ◽  
Regulatory Frameworks ◽  
International Perspectives ◽  
Automated Learning ◽  
The Status ◽  
Recent Developments ◽  
Innovative Approaches

UNSTRUCTURED One of the greatest strengths of artificial intelligence and machine learning (AI/ML) approaches in healthcare is that their performance can be continually improved based on updates from automated learning from data. However, healthcare AI/ML models are currently essentially regulated under provisions that were developed for an earlier age of slowly updated medical devices - requiring major documentation reshape and re-validation with every major update of the model generated by the ML algorithm. This creates minor problems for models that will be re-trained and updated only occasionally, but major problems for models that will learn from data in real-time or near real-time. Regulators have announced action plans for fundamental changes in regulatory approaches. Here, we examine the current regulatory frameworks and the developments in this domain. The status quo and recent developments are reviewed, and we argue that these innovative approaches to healthcare need these matching innovative approaches to regulation and that these approaches will bring benefits for patients. International perspectives from the WHO, and the FDA’s proposed approach, based around oversight of tool developers’ quality management systems and defined algorithm change protocols, offer a much-needed paradigm shift, and strive for a balanced approach to enabling rapid improvements in healthcare through AI innovation, whilst simultaneously ensuring patient safety. The draft EU regulatory framework indicates similar approaches, but no detail has yet been provided on how algorithm change protocols will be implemented in the EU, and this is required for the full benefits of AI/ML-based innovation for patients and for EU healthcare systems to be realised.

Recent Developments of Ambulatory Assessment Methods

2014 ◽  
Vol 25 (4) ◽  
pp. 279-287 ◽  
Author(s):  
Stefan Hey ◽  
Panagiota Anastasopoulou ◽  
André Bideaux ◽  
Wilhelm Stork
Keyword(s):  
Real Time ◽  
Information And Communication Technologies ◽  
New Technologies ◽  
Psychological Research ◽  
Ambulatory Assessment ◽  
Emotional States ◽  
Recent Developments ◽  
Potential Applications ◽  
Information And Communication ◽  
Physiological Sensors

Ambulatory assessment of emotional states as well as psychophysiological, cognitive and behavioral reactions constitutes an approach, which is increasingly being used in psychological research. Due to new developments in the field of information and communication technologies and an improved application of mobile physiological sensors, various new systems have been introduced. Methods of experience sampling allow to assess dynamic changes of subjective evaluations in real time and new sensor technologies permit a measurement of physiological responses. In addition, new technologies facilitate the interactive assessment of subjective, physiological, and behavioral data in real-time. Here, we describe these recent developments from the perspective of engineering science and discuss potential applications in the field of neuropsychology.

Recycling and Reuse in the Roman Economy

2020 ◽  
Keyword(s):  
Building Materials ◽  
Computational Modelling ◽  
Late Antique ◽  
Roman Economy ◽  
Wide Range ◽  
The Status ◽  
Recent Developments ◽  
Methodological Approaches ◽  
First Time ◽  
Site Types

The recycling and reuse of materials and objects were extensive in the past, but have rarely been embedded into models of the economy; even more rarely has any attempt been made to assess the scale of these practices. Recent developments, including the use of large datasets, computational modelling, and high-resolution analytical chemistry, are increasingly offering the means to reconstruct recycling and reuse, and even to approach the thorny matter of quantification. Growing scholarly interest in the topic has also led to an increasing recognition of these practices from those employing more traditional methodological approaches, which are sometimes coupled with innovative archaeological theory. Thanks to these efforts, it has been possible for the first time in this volume to draw together archaeological case studies on the recycling and reuse of a wide range of materials, from papyri and textiles, to amphorae, metals and glass, building materials and statuary. Recycling and reuse occur at a range of site types, and often in contexts which cross-cut material categories, or move from one object category to another. The volume focuses principally on the Roman Imperial and late antique world, over a broad geographical span ranging from Britain to North Africa and the East Mediterranean. Last, but not least, the volume is unique in focusing upon these activities as a part of the status quo, and not just as a response to crisis.

scholarly journals Nanozymes for Environmental Pollutant Monitoring and Remediation

Sensors ◽  
2021 ◽  
Vol 21 (2) ◽  
pp. 408
Author(s):  
Elicia L. S. Wong ◽  
Khuong Q. Vuong ◽  
Edith Chow
Keyword(s):  
Real Time ◽  
Organic Pollutants ◽  
Architectural Design ◽  
Electrochemical Techniques ◽  
Environmental Pollutant ◽  
Field Application ◽  
Optical Methods ◽  
Enzyme Mimicking ◽  
Active Metal ◽  
Recent Developments

Nanozymes are advanced nanomaterials which mimic natural enzymes by exhibiting enzyme-like properties. As nanozymes offer better structural stability over their respective natural enzymes, they are ideal candidates for real-time and/or remote environmental pollutant monitoring and remediation. In this review, we classify nanozymes into four types depending on their enzyme-mimicking behaviour (active metal centre mimic, functional mimic, nanocomposite or 3D structural mimic) and offer mechanistic insights into the nature of their catalytic activity. Following this, we discuss the current environmental translation of nanozymes into a powerful sensing or remediation tool through inventive nano-architectural design of nanozymes and their transduction methodologies. Here, we focus on recent developments in nanozymes for the detection of heavy metal ions, pesticides and other organic pollutants, emphasising optical methods and a few electrochemical techniques. Strategies to remediate persistent organic pollutants such as pesticides, phenols, antibiotics and textile dyes are included. We conclude with a discussion on the practical deployment of these nanozymes in terms of their effectiveness, reusability, real-time in-field application, commercial production and regulatory considerations.

Study and Comparison of Vibration-Based Intrusive and Non-Intrusive Sand Monitoring System

2013 ◽  
Vol 701 ◽  
pp. 440-444
Author(s):  
Gang Liu ◽  
Peng Tao Liu ◽  
Bao Sheng He
Keyword(s):  
Real Time ◽  
Monitoring System ◽  
Oil Production ◽  
Oil Wells ◽  
Oil Field ◽  
Sand Production ◽  
The Status

Sand production is a serious problem during the exploitation of oil wells, and people put forward the concept of limited sand to alleviate this problem. Oil production with limited sanding is an efficient mod of production. In order to complete limited sand exploitation, improve the productivity of oil wells, a real-time sand monitoring system is needed to monitor the status of wells production. Besides acoustic sand monitoring and erosion-based sand monitoring, a vibration-based sand monitoring system with two installing styles is proposed recently. The paper points out the relationships between sand monitoring signals collected under intrusive and non-intrusive installing styles and sanding parameters, which lays a good foundation for further study and actual sand monitoring in oil field.

scholarly journals Distributed X-ray photon correlation spectroscopy data reduction using Hadoop MapReduce

2018 ◽  
Vol 25 (4) ◽  
pp. 1135-1143 ◽  
Author(s):  
Faisal Khan ◽  
Suresh Narayanan ◽  
Roger Sersted ◽  
Nicholas Schwarz ◽  
Alec Sandy
Keyword(s):  
Real Time ◽  
Photon Correlation Spectroscopy ◽  
Correlation Spectroscopy ◽  
Frame Rate ◽  
Computational Techniques ◽  
Complex Materials ◽  
Photon Correlation ◽  
X Ray ◽  
Wide Range ◽  
Recent Developments

Multi-speckle X-ray photon correlation spectroscopy (XPCS) is a powerful technique for characterizing the dynamic nature of complex materials over a range of time scales. XPCS has been successfully applied to study a wide range of systems. Recent developments in higher-frame-rate detectors, while aiding in the study of faster dynamical processes, creates large amounts of data that require parallel computational techniques to process in near real-time. Here, an implementation of the multi-tau and two-time autocorrelation algorithms using the Hadoop MapReduce framework for distributed computing is presented. The system scales well with regard to the increase in the data size, and has been serving the users of beamline 8-ID-I at the Advanced Photon Source for near real-time autocorrelations for the past five years.

Recent Developments

2004 ◽  
Vol 48 (2) ◽  
pp. 268-270 ◽  
Author(s):  
Jon C. Lovett
Keyword(s):  
Biological Diversity ◽  
National Policy ◽  
Plant Conservation ◽  
Global Strategy ◽  
Convention On Biological Diversity ◽  
International Botanical ◽  
International Botanical Congress ◽  
The Status ◽  
Recent Developments ◽  
Essential Resource

At the 1999 International Botanical Congress held in St Louis, Missouri, the President of the Congress, Dr Peter Raven, presented a keynote address emphasizing the importance of plants to human existence and underlined concerns that the Earth is undergoing a human-induced extinction crisis. One of the resolutions of the congress was to establish a new co-ordinating body associated with the United Nations to monitor the status of plants throughout the world and take steps to conserve them. The resolutions were followed up with a meeting in Gran Canaria on 3–4 April, 2000 when leading botanists met to formulate a declaration which could be taken forward to the fifth Conference of the Parties (CoP5) to the Convention on Biological Diversity held in Nairobi in May 2000. CoP5 recognized that plant diversity is a common concern of humankind and an essential resource for the planet, with as many as two-thirds of the world's plant species in danger of extinction, and proposed that at the sixth meeting of the Conference of the Parties (CoP6) the establishment of a global strategy for plant conservation should be considered. A strategy with 16 targets was presented and adopted at the CoP6 meeting held in The Hague in April 2002. These targets differ from the normal approach adopted in the Convention on Biological Diversity of using general principles that can be interpreted by national policy, in that they are quantified.

scholarly journals Deep Learning-Based Cattle Vocal Classification Model and Real-Time Livestock Monitoring System with Noise Filtering

Animals ◽  
2021 ◽  
Vol 11 (2) ◽  
pp. 357
Author(s):  
Dae-Hyun Jung ◽  
Na Yeon Kim ◽  
Sang Ho Moon ◽  
Changho Jhin ◽  
Hak-Jin Kim ◽  
...  
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Model Development ◽  
Classification Model ◽  
Intelligence Analysis ◽  
Noise Filtering ◽  
Meat Industry ◽  
Mel Frequency Cepstral Coefficients ◽  
Web Based ◽  
The Status

The priority placed on animal welfare in the meat industry is increasing the importance of understanding livestock behavior. In this study, we developed a web-based monitoring and recording system based on artificial intelligence analysis for the classification of cattle sounds. The deep learning classification model of the system is a convolutional neural network (CNN) model that takes voice information converted to Mel-frequency cepstral coefficients (MFCCs) as input. The CNN model first achieved an accuracy of 91.38% in recognizing cattle sounds. Further, short-time Fourier transform-based noise filtering was applied to remove background noise, improving the classification model accuracy to 94.18%. Categorized cattle voices were then classified into four classes, and a total of 897 classification records were acquired for the classification model development. A final accuracy of 81.96% was obtained for the model. Our proposed web-based platform that provides information obtained from a total of 12 sound sensors provides cattle vocalization monitoring in real time, enabling farm owners to determine the status of their cattle.

Past participles of strong verbs in Jutland Danish: A real-time study of regionalization and standardization

2012 ◽  
Vol 35 (2) ◽  
pp. 169-195 ◽  
Author(s):  
Torben Juel Jensen ◽  
Marie Maegaard
Keyword(s):  
Real Time ◽  
Geographical Location ◽  
Past Participle ◽  
Time Study ◽  
The Past ◽  
Time Period ◽  
The Status ◽  
Standardization Process ◽  
Large Corpus ◽  
Strong Verbs

The article presents a real-time study of standardization and regionalization processes with respect to the use of past participles of strong verbs in the western part of Denmark. Analyses of a large corpus of recordings of informants from two localities show that the use of the dialectalenform of the past participle suffix has been in decline during the last 30 years. Theenforms are replaced by three other forms, one of which is (partly) dialectal, one regional and one standard Danish. The study indicates that a regionalization process has taken place prior to the time period studied, but that it has now been overtaken by a Copenhagen-based standardization process. The study also shows interesting differences between the two localities, arguably due to the geographical location and size, and to the status of the different participle forms in the traditional local dialects.

scholarly journals High-Quality Transmission of Cardiotocogram and Fetal Information Using a 5G System: Pilot Experiment (Preprint)

2020 ◽  
Author(s):  
Katsuhiko Naruse ◽  
Tomoya Yamashita ◽  
Yukari Onishi ◽  
Yuhi Niitaka ◽  
Fumikage Uchida ◽  
...  
Keyword(s):  
Real Time ◽  
Video Camera ◽  
Ultrasound Images ◽  
Radio Waves ◽  
High Definition ◽  
Waveform Data ◽  
High Definition Video ◽  
Other Information ◽  
Emergency Transport ◽  
The Status

BACKGROUND A cardiotocogram (CTG) is a device used to perceive the status of a fetus in utero in real time. There are a few reports of its use at home or during emergency transport. OBJECTIVE The aim of this study was to test whether CTG and other perinatal information can be transmitted accurately using an experimental station with a 5G transmission system. METHODS In the research institute, real-time fetal heart rate waveform data from the CTG device, high-definition video ultrasound images of the fetus, and high-definition video taken with a video camera on a single line were transmitted by 5G radio waves from the transmitting station to the receiving station. RESULTS All data were proven to be transmitted with a minimum delay of less than 1 second. The CTG waveform image quality was not inferior, and there was no interruption in transmission. Images of the transmitted ultrasound examination and video movie were fine and smooth. CONCLUSIONS CTG and other information about the fetuses and pregnant women were successfully transmitted by a 5G system. This finding will lead to prompt and accurate medical treatment and improve the prognosis of newborns.

Sign in / Sign up

Export Citation Format

Share Document