Learning When Agents Can Talk to Drivers Using the INAGT Dataset and Multisensor Fusion

2021 ◽  
Vol 5 (3) ◽  
pp. 1-28
Author(s):  
Tong Wu ◽  
Nikolas Martelaro ◽  
Simon Stent ◽  
Jorge Ortiz ◽  
Wendy Ju
Keyword(s):  
Sensor Fusion ◽  
Deep Neural Network ◽  
False Positives ◽  
Driver Distraction ◽  
Third Party ◽  
Good Time ◽  
Visual Data ◽  
Perceived Safety ◽  
To Receive ◽  
And Training

This paper examines sensor fusion techniques for modeling opportunities for proactive speech-based in-car interfaces. We leverage the Is Now a Good Time (INAGT) dataset, which consists of automotive, physiological, and visual data collected from drivers who self-annotated responses to the question "Is now a good time?," indicating the opportunity to receive non-driving information during a 50-minute drive. We augment this original driver-annotated data with third-party annotations of perceived safety, in order to explore potential driver overconfidence. We show that fusing automotive, physiological, and visual data allows us to predict driver labels of availability, achieving an 0.874 F1-score by extracting statistically relevant features and training with our proposed deep neural network, PazNet. Using the same data and network, we achieve an 0.891 F1-score for predicting third-party labeled safe moments. We train these models to avoid false positives---determinations that it is a good time to interrupt when it is not---since false positives may cause driver distraction or service deactivation by the driver. Our analyses show that conservative models still leave many moments for interaction and show that most inopportune moments are short. This work lays a foundation for using sensor fusion models to predict when proactive speech systems should engage with drivers.

scholarly journals Automatic ICD-10 Coding and Training System: Deep Neural Network Based on Supervised Learning

10.2196/23230 ◽  
2021 ◽  
Vol 9 (8) ◽  
pp. e23230
Author(s):  
Pei-Fu Chen ◽  
Ssu-Ming Wang ◽  
Wei-Chih Liao ◽  
Lu-Cheng Kuo ◽  
Kuan-Chih Chen ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Deep Neural Network ◽  
University Hospital ◽  
Classification Model ◽  
Icd 10 ◽  
And Training

Background The International Classification of Diseases (ICD) code is widely used as the reference in medical system and billing purposes. However, classifying diseases into ICD codes still mainly relies on humans reading a large amount of written material as the basis for coding. Coding is both laborious and time-consuming. Since the conversion of ICD-9 to ICD-10, the coding task became much more complicated, and deep learning– and natural language processing–related approaches have been studied to assist disease coders. Objective This paper aims at constructing a deep learning model for ICD-10 coding, where the model is meant to automatically determine the corresponding diagnosis and procedure codes based solely on free-text medical notes to improve accuracy and reduce human effort. Methods We used diagnosis records of the National Taiwan University Hospital as resources and apply natural language processing techniques, including global vectors, word to vectors, embeddings from language models, bidirectional encoder representations from transformers, and single head attention recurrent neural network, on the deep neural network architecture to implement ICD-10 auto-coding. Besides, we introduced the attention mechanism into the classification model to extract the keywords from diagnoses and visualize the coding reference for training freshmen in ICD-10. Sixty discharge notes were randomly selected to examine the change in the F1-score and the coding time by coders before and after using our model. Results In experiments on the medical data set of National Taiwan University Hospital, our prediction results revealed F1-scores of 0.715 and 0.618 for the ICD-10 Clinical Modification code and Procedure Coding System code, respectively, with a bidirectional encoder representations from transformers embedding approach in the Gated Recurrent Unit classification model. The well-trained models were applied on the ICD-10 web service for coding and training to ICD-10 users. With this service, coders can code with the F1-score significantly increased from a median of 0.832 to 0.922 (P<.05), but not in a reduced interval. Conclusions The proposed model significantly improved the F1-score but did not decrease the time consumed in coding by disease coders.

Skills and Training for the Older Population

2017 ◽  
Author(s):  
Wendy Loretto ◽  
Chris Phillipson ◽  
Sarah Vickerstaff
Keyword(s):  
Older Workers ◽  
Training And Development ◽  
Active Ageing ◽  
Training And Education ◽  
Training Development ◽  
Work Training ◽  
Employment Rates ◽  
The Uk ◽  
To Receive ◽  
And Training

Despite rises in employment rates across many countries, older workers (those aged 50+) are less likely than younger employees to receive workplace training and skills development. Using the UK as its starting focus, this chapter analyses the theoretical and empirical reasons for these gaps. The analysis covers in-work training and development, as well as considering the position of those older people who are unemployed but looking for work. The discussion also embraces the roles of training and education for older workers who may want to delay retirement or retire flexibly, and examines the relationships between training, development and active ageing. Concluding discussions highlight national and international policy initiatives to encourage investment in educating and training for this new work generation.

A Similarity-Based Object Classification Using Deep Neural Networks

2019 ◽  
pp. 197-219
Author(s):  
Parvathi R. ◽  
Pattabiraman V.
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Deep Neural Network ◽  
Search Algorithm ◽  
Random Projection ◽  
Learning Networks ◽  
Learning Models ◽  
External Conditions ◽  
And Training

This chapter proposes a hybrid method for classification of the objects based on deep neural network and a similarity-based search algorithm. The objects are pre-processed with external conditions. After pre-processing and training different deep learning networks with the object dataset, the authors compare the results to find the best model to improve the accuracy of the results based on the features of object images extracted from the feature vector layer of a neural network. RPFOREST (random projection forest) model is used to predict the approximate nearest images. ResNet50, InceptionV3, InceptionV4, and DenseNet169 models are trained with this dataset. A proposal for adaptive finetuning of the deep learning models by determining the number of layers required for finetuning with the help of the RPForest model is given, and this experiment is conducted using the Xception model.

Current Status of Vascularized Composite Allotransplantation

2019 ◽  
Vol 85 (6) ◽  
pp. 631-637 ◽  
Author(s):  
Christina L. Kaufman ◽  
Neal Bhutiani ◽  
Allan Ramirez ◽  
Huey Y. Tien ◽  
Michelle D. Palazzo ◽  
...  
Keyword(s):  
Standard Of Care ◽  
The United States ◽  
Third Party ◽  
Tissue Loss ◽  
Current Status ◽  
Vascularized Composite Allotransplantation ◽  
20Th Anniversary ◽  
Treatment Alternatives ◽  
To Receive

The field of vascularized composite allotransplantation (VCA) has moved from a highly experimental procedure to, at least for some patients, one of the best treatment alternatives for catastrophic tissue loss or dysfunction. Although the worldwide experience is still limited, progress has been made in translation to the clinic, and hand transplantation was recently designated standard of care and is now covered in full by the British Health System. This progress is tempered by the long-term challenges of systemic immunosuppression, and the rapidly evolving indications for VCA such as urogenital transplantation. This update will cover the state of and recent changes in the field, and an update of the Louisville VCA program as our initial recipient, the first person to receive a hand transplant in the United States celebrates the 20th anniversary of his transplant. The achievements and complications encountered over the last two decades will be reviewed. In addition, potential directions for research and collaboration as well as practical issues of how third party payers and funding are affecting growth of the field are presented.

scholarly journals Experience With Using Code Case 2564 to Design and Certify an Impulsively Loaded Vessel

2013 ◽  
Author(s):  
Brent Haroldsen ◽  
Jerome Stofleth ◽  
Mien Yip ◽  
Allan Caplan
Keyword(s):  
Lessons Learned ◽  
Regulatory Approval ◽  
Department Of Energy ◽  
Third Party ◽  
Us Army ◽  
The Us ◽  
User Design ◽  
Qualification Testing ◽  
To Receive ◽  
Explosive Destruction

Code Case 2564 for the design of impulsively loaded vessels was approved in January 2008. In 2010 the US Army Non-Stockpile Chemical Materiel Program, with support from Sandia National Laboratories, procured a vessel per this Code Case for use on the Explosive Destruction System (EDS). The vessel was delivered to the Army in August of 2010 and approved for use by the DoD Explosives Safety Board in 2012. Although others have used the methodology and design limits of the Code Case to analyze vessels, to our knowledge, this was the first vessel to receive an ASME explosive rating with a U3 stamp. This paper discusses lessons learned in the process. Of particular interest were issues related to defining the design basis in the User Design Specification and explosive qualification testing required for regulatory approval. Specifying and testing an impulsively loaded vessel is more complicated than a static pressure vessel because the loads depend on the size, shape, and location of the explosive charges in the vessel and on the kind of explosives used and the point of detonation. Historically the US Department of Defense and Department of Energy have required an explosive test. Currently the Code Case does not address testing requirements, but it would be beneficial if it did since having vetted, third party standards for explosive qualification testing would simplify the process for regulatory approval.

Assessing patient satisfaction after comprehensive medication review including oral chemotherapy adherence with an oncology pharmacist at the Southcoast Centers for Cancer Care (SCCC).

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. e19241-e19241
Author(s):  
Patrick John Skeffington ◽  
Laura Haynes ◽  
Donna Raymond ◽  
Heather McCarthy
Keyword(s):  
Patient Satisfaction ◽  
Community Pharmacists ◽  
Oral Chemotherapy ◽  
Third Party ◽  
Effective Counseling ◽  
Open Discussion ◽  
Specialty Pharmacy ◽  
Strongly Agree ◽  
Pharmacist Services ◽  
To Receive

e19241 Background: Use of oral chemotherapy has increased dramatically over the past few years. Patient often are required to obtain their oral chemotherapy from a third party specialty pharmacy while continuing to receive their other medications from other pharmacies. Many community pharmacists lack knowledge about oral chemotherapy, safe practices, or effective counseling of these medications. Objective: To assess patient perception and satisfaction of a pharmacist 1 on 1 appointment when starting oral chemotherapy. Methods: A program was designed at SCCC whereby all patients starting oral chemotherapy are scheduled for an appointment with a clinical oncology pharmacist to update medication lists, evaluate adherence, and conduct a "brown bag" visit where patients are allowed to voice concerns and ask questions. After each appointment patients were asked to fill out a short survey, The Patient Satisfaction with Pharmacist Services Questionnaire (PSPSQ 2.0). Results: PSPSQ 2.0 uses a Likert scale ranging from 1 to 4. From October 2016 to June 2019, 174 patients had appointments and 55 returned their surveys yielding a 30% response rate. Average scores hovered around 1 (strongly agree) for each question except question 11 (the only negatively worded question). Question 11 averaged 3.1; Disagree. Conclusions: Patients who were seen by an oncology clinical pharmacist to evaluate adherence, participate in a "brown bag" clinic and open discussion, found the appointment worthy of their time.

Study on the Application of Virtual Prototyping Technology to UAV Fault Diagnosis Training

2013 ◽  
Vol 347-350 ◽  
pp. 396-400
Author(s):  
Yong Kang Jiao ◽  
Xiao Min Li
Keyword(s):  
Fault Diagnosis ◽  
Design Thinking ◽  
Virtual Prototyping ◽  
Third Party ◽  
Appearance Model ◽  
Mechanism Model ◽  
Modeling Software ◽  
Virtual Prototyping Technology ◽  
Performance Capability ◽  
And Training

In accordance with the characteristic that UAV equipment is hard to realize the real fault diagnosis training, a design scheme of fault diagnosis training based on virtual prototyping is presented. The training-oriented virtual prototyping model is decomposed into two parts, appearance model and mechanism model. The design thinking and training flow of fault diagnosis training are given, and two different kinds of training modes are used to make the training much more targeted. In the end, Virtools and third-party modeling software are adopted to build the virtual environment, the verification of UAV ground control station achieves the purpose of enhancing the performance capability of maintenance man.

scholarly journals FPGA Realization of Deep Neural Network for Hardware Trojan Detection

2020 ◽  
Vol 9 (3) ◽  
pp. 764
Author(s):  
Varun Reddy ◽  
Nirmala Devi M
Keyword(s):  
Neural Network ◽  
Integrated Circuits ◽  
Power Consumption ◽  
Deep Neural Network ◽  
Random Forest Classifier ◽  
Third Party ◽  
Hardware Trojan ◽  
Hardware Trojan Detection ◽  
Trojan Detection ◽  
Field Programmable

With the increase in outsourcing design and fabrication, malicious third-party vendors often insert hardware Trojan (HT) in the integrated Circuits(IC). It is difficult to identify these Trojans since the nature and characteristics of each Trojan differ significantly. Any method developed for HT detection is limited by its capacity on dealing with varied types of Trojans. The main purpose of this study is to show using deep learning (DL), this problem can be dealt with some extent and the effect of deep neural network (DNN) when it is realized on field programmable gate array (FPGA). In this paper, we propose a comparison of accuracy in finding faults on ISCAS’85 benchmark circuits between random forest classifier and DNN. Further for the faster processing time and less power consumption, the network is implemented on FPGA. The results show the performance of deep neural network gets better when a large number of nets are used and faster in the execution of the algorithm. Also, the speedup of the neuron is 100x times better when implemented on FPGA with 15.32% of resource utilization and provides less power consumption than GPU.

scholarly journals The Study of Teachers' and Parents' Needs for Psychological Consultation from School Psychologist

2017 ◽  
Vol 9 (4) ◽  
pp. 1-11 ◽  
Author(s):  
E.A. Savina ◽  
I.A. Savenkova ◽  
A.E. Esterle ◽  
E.A. Ovsyanikova ◽  
M.Y. Khudaeva
Keyword(s):  
Teaching Methods ◽  
School Psychologist ◽  
Interpersonal Problems ◽  
Learning Problems ◽  
The Past ◽  
Psychological Consultation ◽  
Parent Relationships ◽  
Teachers And Parents ◽  
To Receive ◽  
And Training

The present study was aimed at the investigation of teachers’ and parents’ needs in consultation with a school psychologist. Participants were 159 teachers and 292 parents from three cities in Russia. Two surveys were designed to measure teachers’ and parents’ desire to receive psychological consultation regarding behavioral, emotional, learning and interpersonal problems of students; teaching methods and relationships with colleagues (for teachers); and child-parent relationships. In addition, the participants were asked to indicate whether they received a consultation from a school psychologist in the past and their satisfaction from the consultation. The results indicated that, in general, both teachers and parents are satisfied with the consultation; however, fewer parents received such a consultation compared to teachers. Both teachers and parents are more willing to receive consultation regarding children’s behavioral and emotional problems and relationships with peers. Teachers are less motivated to receive consultation about teaching methods, students’ learning problems, and teachers’ relationships with colleagues. Parents were less interested to receive consultation about child-parent relationships. The results of this study are interpreted in terms of their alignment with standards, which regulate the school psychology profession and training.

Sign in / Sign up

Export Citation Format

Share Document