scholarly journals Detecting Mistakes in CPR Training with Multimodal Data and Neural Networks

Sensors ◽  
2019 ◽  
Vol 19 (14) ◽  
pp. 3099 ◽  
Author(s):  
Daniele Di Mitri ◽  
Jan Schneider ◽  
Marcus Specht ◽  
Hendrik Drachsler
Keyword(s):  
Neural Networks ◽  
Body Weight ◽  
Performance Indicators ◽  
Research Team ◽  
Body Position ◽  
Microsoft Kinect ◽  
Chest Compressions ◽  
Multimodal Data ◽  
Myo Armband ◽  
Cpr Training

This study investigated to what extent multimodal data can be used to detect mistakes during Cardiopulmonary Resuscitation (CPR) training. We complemented the Laerdal QCPR ResusciAnne manikin with the Multimodal Tutor for CPR, a multi-sensor system consisting of a Microsoft Kinect for tracking body position and a Myo armband for collecting electromyogram information. We collected multimodal data from 11 medical students, each of them performing two sessions of two-minute chest compressions (CCs). We gathered in total 5254 CCs that were all labelled according to five performance indicators, corresponding to common CPR training mistakes. Three out of five indicators, CC rate, CC depth and CC release, were assessed automatically by the ReusciAnne manikin. The remaining two, related to arms and body position, were annotated manually by the research team. We trained five neural networks for classifying each of the five indicators. The results of the experiment show that multimodal data can provide accurate mistake detection as compared to the ResusciAnne manikin baseline. We also show that the Multimodal Tutor for CPR can detect additional CPR training mistakes such as the correct use of arms and body weight. Thus far, these mistakes were identified only by human instructors. Finally, to investigate user feedback in the future implementations of the Multimodal Tutor for CPR, we conducted a questionnaire to collect valuable feedback aspects of CPR training.

scholarly journals Keep Me in the Loop: Real-Time Feedback with Multimodal Data

2021 ◽  
Author(s):  
Daniele Di Mitri ◽  
Jan Schneider ◽  
Hendrik Drachsler
Keyword(s):  
Field Study ◽  
Real Time ◽  
Performance Indicators ◽  
User Study ◽  
Methodological Approach ◽  
Feedback System ◽  
Multimodal Feedback ◽  
Multimodal Data ◽  
Cpr Training ◽  
The Impact

AbstractThis paper describes the CPR Tutor, a real-time multimodal feedback system for cardiopulmonary resuscitation (CPR) training. The CPR Tutor detects training mistakes using recurrent neural networks. The CPR Tutor automatically recognises and assesses the quality of the chest compressions according to five CPR performance indicators. It detects training mistakes in real-time by analysing a multimodal data stream consisting of kinematic and electromyographic data. Based on this assessment, the CPR Tutor provides audio feedback to correct the most critical mistakes and improve the CPR performance. The mistake detection models of the CPR Tutor were trained using a dataset from 10 experts. Hence, we tested the validity of the CPR Tutor and the impact of its feedback functionality in a user study involving additional 10 participants. The CPR Tutor pushes forward the current state of the art of real-time multimodal tutors by providing: (1) an architecture design, (2) a methodological approach for delivering real-time feedback using multimodal data and (3) a field study on real-time feedback for CPR training. This paper details the results of a field study by quantitatively measuring the impact of the CPR Tutor feedback on the performance indicators and qualitatively analysing the participants’ questionnaire answers.

scholarly journals Table Tennis Tutor: Forehand Strokes Classification Based on Multimodal Data and Neural Networks

Sensors ◽  
2021 ◽  
Vol 21 (9) ◽  
pp. 3121
Author(s):  
Khaleel Asyraaf Mat Sanusi ◽  
Daniele Di Mitri ◽  
Bibeg Limbu ◽  
Roland Klemke
Keyword(s):  
Real Time ◽  
Body Position ◽  
Feedback System ◽  
Microsoft Kinect ◽  
Psychomotor Skills ◽  
Table Tennis ◽  
Multimodal Data ◽  
Interview Session ◽  
Expert Interview ◽  
Smartphone Sensors

Beginner table-tennis players require constant real-time feedback while learning the fundamental techniques. However, due to various constraints such as the mentor’s inability to be around all the time, expensive sensors and equipment for sports training, beginners are unable to get the immediate real-time feedback they need during training. Sensors have been widely used to train beginners and novices for various skills development, including psychomotor skills. Sensors enable the collection of multimodal data which can be utilised with machine learning to classify training mistakes, give feedback, and further improve the learning outcomes. In this paper, we introduce the Table Tennis Tutor (T3), a multi-sensor system consisting of a smartphone device with its built-in sensors for collecting motion data and a Microsoft Kinect for tracking body position. We focused on the forehand stroke mistake detection. We collected a dataset recording an experienced table tennis player performing 260 short forehand strokes (correct) and mimicking 250 long forehand strokes (mistake). We analysed and annotated the multimodal data for training a recurrent neural network that classifies correct and incorrect strokes. To investigate the accuracy level of the aforementioned sensors, three combinations were validated in this study: smartphone sensors only, the Kinect only, and both devices combined. The results of the study show that smartphone sensors alone perform sub-par than the Kinect, but similar with better precision together with the Kinect. To further strengthen T3’s potential for training, an expert interview session was held virtually with a table tennis coach to investigate the coach’s perception of having a real-time feedback system to assist beginners during training sessions. The outcome of the interview shows positive expectations and provided more inputs that can be beneficial for the future implementations of the T3.

Predicting carcass energy content and composition in broilers using the group method of data handling-type neural networks

2010 ◽  
Vol 149 (2) ◽  
pp. 249-254 ◽  
Author(s):  
A. FARIDI ◽  
M. MOTTAGHITALAB ◽  
H. DARMANI-KUHI ◽  
J. FRANCE ◽  
H. AHMADI
Keyword(s):  
Neural Networks ◽  
Body Weight ◽  
Goodness Of Fit ◽  
Carcass Composition ◽  
Poultry Meat ◽  
Alternative Methods ◽  
Group Method ◽  
Meat Production ◽  
Data Handling ◽  
Input Variables

SUMMARYThe success of poultry meat production has been strongly related to improvements in growth and carcass yield, mainly by increasing breast proportion and reducing carcass fat. Conventional laboratory techniques for determining carcass composition are expensive, cumbersome and time consuming. These disadvantages have prompted a search for alternative methods. In this respect, the potential benefits from modelling growth are considerable. Neural networks (NNs) are a relatively new option for modelling growth in animal production systems. One self-organizing sub-model of artificial NN is the group method of data handling-type NN (GMDH-type NN). The present study aimed at applying the GMDH-type NNs to data from two studies with broilers in order to predict carcass energy (CEn, MJ/g) content and relative growth (g/g of body weight) of carcass components (carcass protein, breast muscle, leg and thigh muscles, carcass fat, abdominal fat, skin fat and visceral fat). The effective input variables involved in the prediction of CEn and carcass fat content using data from the first study were dietary metabolizable energy (ME, kJ/kg), crude protein (CP, g/kg of diet), fat (g/kg of diet) and crude fibre (CF, g/kg of diet). For data from the second study, the effective input variables involved in the prediction of carcass components were dietary ME (MJ/kg), CP (g/kg of diet), methionine (g/kg of diet), lysine (g/kg of diet) and body weight (kg). Quantitative examination of the goodness of fit, using R2 and error measurement indices, for the predictive models proposed by the GMDH-type NN revealed close agreement between observed and predicted values of CEn and carcass components.

The Effect of Body Position and Number of Supports on Wall Reaction Forces in Rock Climbing

1997 ◽  
Vol 13 (1) ◽  
pp. 14-23 ◽  
Author(s):  
Franck Quaine ◽  
Luc Martin ◽  
Jean-Pierre Blanchi
Keyword(s):  
Body Weight ◽  
Contact Force ◽  
Body Position ◽  
Three Dimensional ◽  
Rock Climbing ◽  
Contact Forces ◽  
Vertical Posture ◽  
Reaction Forces ◽  
Force Data ◽  
Lower Contact

This manuscript describes three-dimensional force data collected during postural shifts performed by individuals simulating rock-climbing skills. Starting from a quadrupedal vertical posture, 6 expert climbers had to release their right-hand holds and maintain the tripedal posture for a few seconds. The vertical and contact forces (lateral and anteroposterior forces) applied on the holds were analyzed in two positions: an “imposed” position (the trunk far from the supporting wall) and an “optimized” position (the trunk close to the wall and lower contact forces at the holds). The tripedal postures performed in the two positions were achieved by the same pattern of vertical and contact forces exerted by the limbs on the holds. In the optimized position, the transfer of the forces was less extensive than in the imposed position, so that the forces were exerted primarily on the ipsilateral hold. Moreover, a link between the contact force values and the couple due to body weight with respect to the feet was shown.

scholarly journals A Novel Method for Performance Measurement of Public Educational Institutions Using Machine Learning Models

2021 ◽  
Vol 11 (19) ◽  
pp. 9296
Author(s):  
Talha Mahboob Alam ◽  
Mubbashar Mushtaq ◽  
Kamran Shaukat ◽  
Ibrahim A. Hameed ◽  
Muhammad Umer Sarwar ◽  
...  
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Feature Selection ◽  
Artificial Neural Networks ◽  
Performance Indicators ◽  
Key Performance Indicators ◽  
Educational Institutions ◽  
Institutional Performance ◽  
Learning Models ◽  
Machine Learning Models

Lack of education is a major concern in underdeveloped countries because it leads to poor human and economic development. The level of education in public institutions varies across all regions around the globe. Current disparities in access to education worldwide are mostly due to systemic regional differences and the distribution of resources. Previous research focused on evaluating students’ academic performance, but less has been done to measure the performance of educational institutions. Key performance indicators for the evaluation of institutional performance differ from student performance indicators. There is a dire need to evaluate educational institutions’ performance based on their disparities and academic results on a large scale. This study proposes a model to measure institutional performance based on key performance indicators through data mining techniques. Various feature selection methods were used to extract the key performance indicators. Several machine learning models, namely, J48 decision tree, support vector machines, random forest, rotation forest, and artificial neural networks were employed to build an efficient model. The results of the study were based on different factors, i.e., the number of schools in a specific region, teachers, school locations, enrolment, and availability of necessary facilities that contribute to school performance. It was also observed that urban regions performed well compared to rural regions due to the improved availability of educational facilities and resources. The results showed that artificial neural networks outperformed other models and achieved an accuracy of 82.9% when the relief-F based feature selection method was used. This study will help support efforts in governance for performance monitoring, policy formulation, target-setting, evaluation, and reform to address the issues and challenges in education worldwide.

Abstract 2019: Untrained Volunteers Perform High Quality CPR When using an Automatic External Defibrillator with a CPR Voice Prompting Algorithm

Circulation ◽  
2007 ◽  
Vol 116 (suppl_16) ◽  
Author(s):  
Benjamin S Abella ◽  
Salem Kim ◽  
Alexandra Colombus ◽  
Cheryl L Shea ◽  
Lance B Becker
Keyword(s):  
Real Time ◽  
Patient Outcomes ◽  
Chest Compression ◽  
Automatic External Defibrillator ◽  
Chest Compressions ◽  
High Quality ◽  
Cpr Training ◽  
Adult Volunteers ◽  
Improve Patient ◽  
External Defibrillator

Background: Recent investigations have demonstrated that CPR performance among trained providers can be improved by audiovisual prompting and real-time feedback, and higher quality CPR before defibrillation can improve shock success and has the potential to improve patient outcomes. Objective: We hypothesized that simplified voice prompts incorporated into an automatic external defibrillator (AED) can lead to improvements in CPR performance by untrained lay rescuers. Methods: Adult volunteers with no prior CPR training were assessed in their use of an AED with chest compression voice instructions and metronome prompts on a CPR-recording manikin. Volunteers were given minimal instructions regarding use of the device and were given no instructions regarding CPR performance. The AED was designed to prompt five cycles of 30 chest compressions between defibrillatory attempts. Chest compression rates and depths were measured via review of videotape and manikin recording data, respectively. Results: A total of 60 adults were assessed in their use of the AED, with a mean age of 33.6±12.8; 36/63 (57%) were female. Mean chest compression rate was 103±12 and mean depth was 37±14 mm. Furthermore, minimal decay in chest compression rates occurred over 5 cycles of chest compressions, with mean rate of 101±19 during the first cycle and 104±10 during the 5 th cycle. No volunteers were unable to use the AED or complete 5 cycles of chest compressions. Conclusions: Our work demonstrates that with appropriate real-time prompts delivered even in the absence of training or human coaching, laypersons can perform CPR that has a quality often similar to trained providers. This finding has important implications for AED design especially in light of the renewed importance of both CPR and the interaction of quality chest compressions and defibrillatory success.

scholarly journals Performance indicators of broilers fed propolis and bee pollen additive

Poljoprivreda ◽  
2019 ◽  
Vol 25 (1) ◽  
pp. 69-75 ◽  
Author(s):  
Ivana Prakatur ◽  
Matija Domaćinović ◽  
Božica Lachner ◽  
Zvonimir Steiner ◽  
Dalida Galović ◽  
...  
Keyword(s):  
Experimental Study ◽  
Body Weight ◽  
Weight Gain ◽  
Performance Indicators ◽  
Control Group ◽  
Significant Positive Effect ◽  
Feed Mixture ◽  
Bee Pollen ◽  
Positive Effect ◽  
Diet Supplementation

The aim of this study was to determine the effect of diet supplementation with propolis and/or bee pollen on the performance indicators of broilers. This experimental study was conducted on 200 Ross 308 broilers equally distributed by sex and divided into five groups. Throughout the whole study the control group of chickens was fed feed mixture. Feed mixture fed to the experimental groups of chickens contained additives (propolis and/or bee pollen, each supplement separately or in combination in a certain proportion). The average values of broilers body weight were significantly higher on 7th (p=0.001), 14th, 21st, 28th, 35th (p<0.001) and 42nd (p=0.002) day of feeding in the experimental groups of broilers compared to the control group. The average values of broilers weight gain were significantly higher on 1st (p<0.001), 2nd (p=0.002), 3rd (p<0.001), 4th (p=0.029) and 5th (p=0.009) week of feeding in the experimental groups of broilers compared to the control group. This study has undoubtedly shown that propolis and bee pollen have significant positive effect on performance indicators of broilers.

Effects of exercise on fluid exchange and body composition in man during 14-day bed rest

1977 ◽  
Vol 43 (1) ◽  
pp. 126-132 ◽  
Author(s):  
J. E. Greenleaf ◽  
E. M. Bernauer ◽  
L. T. Juhos ◽  
H. L. Young ◽  
J. T. Morse ◽  
...  
Keyword(s):  
Body Weight ◽  
Metabolic Rate ◽  
Lean Body Mass ◽  
Body Mass ◽  
Body Position ◽  
Isometric Exercise ◽  
Bed Rest ◽  
The Body ◽  
Fluid Exchange ◽  
Isotonic Exercise

To determine the cause of the body weight loss during bed rest (BR), fluid balance and anthropometric measurements were taken from seven men (19–21 yr) during three 2-wk BR periods which were separated by 3-wk ambulatory recovery periods. Caloric intake was 3,073 +/- 155 (SD) kcal/day. During two of the three BR periods they performed supine isotonic exercise at 68% of VO2max on the ergometer for 1 h/day; or supine isometric exercise at 21% of maximal leg extension force for 1 min followed by a 1-min rest for 1 h/day. No prescribed exercise was given during the other BR period. During BR, body weight decreased slightly with no exercise (-0.43 kg, NS), but decreased significantly (P less than 0.05) by -0.91 kg with isometric and by -1.77 kg with isotonic exercise. About one-third of the weight reduction with isotonic exercise was due to fat loss (-0.69 kg) and, the remainder, to loss of lean body mass (-0.98 kg). It is concluded that the reduction in body weight during bed rest has two major components: First, a loss of lean body mass caused by assumption of the horizontal body position that is independent of the metabolic rate. Second, a loss of body fat content that is proportional to the metabolic rate.

scholarly journals A hybrid system of neural networks and rough sets for road safety performance indicators

2009 ◽  
Vol 14 (12) ◽  
pp. 1255-1263 ◽  
Author(s):  
Yongjun Shen ◽  
Tianrui Li ◽  
Elke Hermans ◽  
Da Ruan ◽  
Geert Wets ◽  
...  
Keyword(s):  
Neural Networks ◽  
Hybrid System ◽  
Performance Indicators ◽  
Rough Sets ◽  
Road Safety ◽  
Safety Performance

Hybrid Electric Vehicle Emissions Monitoring and Estimation using Artificial Neural Networks: Technical Note

2019 ◽  
Vol 11 (3) ◽  
Author(s):  
K. Sujatha ◽  
V. Karthikeyan ◽  
V. Balaji ◽  
N.P.G. Bhavani ◽  
V. Srividhya ◽  
...  
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Performance Indicators ◽  
Hybrid Electric Vehicle ◽  
Back Propagation ◽  
Technical Note ◽  
Back Propagation Algorithm ◽  
Highway Traffic ◽  
Artificial Neural ◽  
Optimal Control Action

Power is utilized as the prime fuel for hybrid and module electric vehicles in order to build the productivity of commercial vehicles. This paper forecasts the emission factors utilizing discrete Fourier transform, artificial neural networks and hybridization of back propagation algorithm. The DFT facilitates the extraction of the performance indicators which are otherwise called the features. The coefficients of the power spectrum denote the performance indicators. The ANN learns the pattern for emissions from HEVs using these performance indicators. This ANN based strategy offers an optimal control action to detect and reduce the exhaust gas emissions which are hazardous. These vehicles are provided with automated highway traffic Jam assist. Hence the forecast of these emissions offers increased efficiency of 90% to 100% thereby ensuring optimal operating condition for the hybrid vehicles.

Sign in / Sign up

Export Citation Format

Share Document