| 1. |
- Begum, Shahina, et al.
(författare)
-
A Fusion Based System for Physiological Sensor Signal Classification
- 2014
-
Ingår i: Medicinteknikdagarna 2014 MTD10.
-
Konferensbidrag (refereegranskat)abstract
- Today, usage of physiological sensor signals is essential in medical applications for diagnoses and classification of diseases. Clinicians often rely on information collected from several physiological sensor signals to diagnose a patient. However, sensor signals are mostly non-stationary and noisy, and single sensor signal could easily be contaminated by uncertain noises and interferences that could cause miscalculation of measurements and reduce clinical usefulness. Therefore, an apparent choice is to use multiple sensor signals that could provide more robust and reliable decision. Therefore, a physiological signal classification approach is presented based on sensor signal fusion and case-based reasoning. To classify Stressed and Relaxed individuals from physiological signals, data level and decision level fusion are performed and case-based reasoning is applied as classification algorithm. Five physiological sensor signals i.e., Heart Rate (HR), Finger Temperature (FT), Respiration Rate (RR), Carbon dioxide (CO2) and Oxygen Saturation (SpO2) are collected during the data collection phase. Here, data level fusion is performed using Multivariate Multiscale Entropy (MMSE) and extracted features are then used to build a case- library. Decision level fusion is performed on the features extracted using traditional time and frequency domain analysis. Case-Based Reasoning (CBR) is applied for the classification of the signals. The experimental result shows that the proposed system could classify Stressed or Relaxed individual 87.5% accurately compare to an expert in the domain. So, it shows promising result in the psychophysiological domain and could be possible to adapt this approach to other relevant healthcare systems.
|
|
| 2. |
- Ahmed, Mobyen Uddin, et al.
(författare)
-
A Generic System-level Framework for Self-Serve Health Monitoring System through Internet of Things(IoT)
- 2015
-
Ingår i: Studies in Health Technology and Informatics, Volume 211. - 9781614995159 ; , s. 305-307
-
Konferensbidrag (refereegranskat)abstract
- Sensor data are traveling from sensors to a remote server, data is analysed remotely in a distributed manner, and health status of a user is presented in real-time. This paper presents a generic system-level framework for a self-served health monitoring system through the Internet of Things (IoT) to facilities an efficient sensor data management.
|
|
| 3. |
- Barua, Shaibal, et al.
(författare)
-
Automatic driver sleepiness detection using EEG, EOG and contextual information
- 2019
-
Ingår i: Expert systems with applications. - : Elsevier Ltd. - 0957-4174 .- 1873-6793. ; 115, s. 121-135
-
Tidskriftsartikel (refereegranskat)abstract
- The many vehicle crashes that are caused by driver sleepiness each year advocates the development of automated driver sleepiness detection (ADSD) systems. This study proposes an automatic sleepiness classification scheme designed using data from 30 drivers who repeatedly drove in a high-fidelity driving simulator, both in alert and in sleep deprived conditions. Driver sleepiness classification was performed using four separate classifiers: k-nearest neighbours, support vector machines, case-based reasoning, and random forest, where physiological signals and contextual information were used as sleepiness indicators. The subjective Karolinska sleepiness scale (KSS) was used as target value. An extensive evaluation on multiclass and binary classifications was carried out using 10-fold cross-validation and leave-one-out validation. With 10-fold cross-validation, the support vector machine showed better performance than the other classifiers (79% accuracy for multiclass and 93% accuracy for binary classification). The effect of individual differences was also investigated, showing a 10% increase in accuracy when data from the individual being evaluated was included in the training dataset. Overall, the support vector machine was found to be the most stable classifier. The effect of adding contextual information to the physiological features improved the classification accuracy by 4% in multiclass classification and by and 5% in binary classification.
|
|
| 4. |
- Rahman, Hamidur, et al.
(författare)
-
Driver monitoring in the context of autonomous vehicle
- 2015
-
Ingår i: Frontiers in Artificial Intelligence and Applications. - Amsterdam. - 9781614995883 ; , s. 108-117
-
Konferensbidrag (refereegranskat)abstract
- Today research is going on within different essential functions need to bring automatic vehicles to the roads. However, there will be manual driven vehicles for many years before it is fully automated vehicles on roads. In complex situations, automated vehicles will need human assistance for long time. So, for road safety driver monitoring is even more important in the context of autonomous vehicle to keep the driver alert and awake. But, limited effort has been done in total integration between automatic vehicle and human driver. Therefore, human drivers need to be monitored and able to take over control within short notice. This papers provides an overview on autonomous vehicles and un-obstructive driver monitoring approaches that can be implemented in future autonomous vehicles to monitor driver e.g., to diagnose and predict stress, fatigue etc. in semi-automated vehicles.
|
|
| 5. |
- Ahmed, Mobyen Uddin, Dr, 1976-, et al.
(författare)
-
A Machine Learning Approach to Classify Pedestrians’ Event based on IMU and GPS
- 2019
-
Ingår i: International Conference on Modern Intelligent Systems Concepts MISC'18. - : CESER Publications. ; 17:2, s. 154-167
-
Konferensbidrag (refereegranskat)abstract
- This paper investigates and implements six Machine Learning (ML) algorithms, i.e. Artificial Neural Network (ANN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), Extra Tree (ET), and Gradient Boosted Trees (GBT) to classify different Pedestrians’ events based on Inertial Measurement Unit (IMU) and Global Positioning System (GPS) signals. Pedestrians’ events are pedestrian movements as the first step of H2020 project called SimuSafe1 with a goal to reduce traffic fatalities by doing risk assessments of the pedestrians. The movements the MLs’ models are attempting to classify are standing, walking, and running. Data, i.e. IMU, GPS sensor signals and other contextual information are collected by a smartphone through a controlled procedure. The smartphone is placed in five different positions onto the body of participants, i.e. arm, chest, ear, hand and pocket. The recordings are filtered, trimmed, and labeled. Next, samples are generated from small overlapping sections from which time and frequency domain features are extracted. Three different experiments are conducted to evaluate the performances in term of accuracy of the MLs’ models in different circumstances. The best performing MLs’ models determined by the average accuracy across all experiments is Extra Tree (ET) with a classification accuracy of 91%.
|
|
| 6. |
- Ahmed, Mobyen Uddin, Dr, 1976-, et al.
(författare)
-
A vision-based indoor navigation system for individuals with visual impairment
- 2019
-
Ingår i: International Journal of Artificial Intelligence. - : CESER Publications. - 0974-0635. ; 17:2, s. 188-201
-
Tidskriftsartikel (refereegranskat)abstract
- Navigation and orientation in an indoor environment are a challenging task for visually impaired people. This paper proposes a portable vision-based system to provide support for visually impaired persons in their daily activities. Here, machine learning algorithms are used for obstacle avoidance and object recognition. The system is intended to be used independently, easily and comfortably without taking human help. The system assists in obstacle avoidance using cameras and gives voice message feedback by using a pre-trained YOLO Neural Network for object recognition. In other parts of the system, a floor plane estimation algorithm is proposed for obstacle avoidance and fuzzy logic is used to prioritize the detected objects in a frame and generate alert to the user about possible risks. The system is implemented using the Robot Operating System (ROS) for communication on a Nvidia Jetson TX2 with a ZED stereo camera for depth calculations and headphones for user feedback, with the capability to accommodate different setup of hardware components. The parts of the system give varying results when evaluated and thus in future a large-scale evaluation is needed to implement the system and get it as a commercialized product in this area.
|
|
| 7. |
- Ahmed, Mobyen Uddin, 1976-, et al.
(författare)
-
A case-based patient identification system using pulseoximeter and a personalized health profile
- 2012
-
Ingår i: Proceedings of the ICCBR 2012 Workshops. - Lyon, France. ; , s. 117-128
-
Konferensbidrag (refereegranskat)abstract
- This paper proposes a case-based system framework in order to identify patient using their health parameters taken with physiological sensors. It combines a personalized health profiling protocol with a Case-Based Reasoning (CBR) approach. The personalized health profiling helps to determine a number of individual parameters which are important inputs for a clinician to make the final diagnosis and treatment plan. The proposed system uses a pulse oximeter that measures pulse rate and blood oxygen saturation. The measurements are taken through an android application in a smart phone which is connected with the pulseoximeter and bluetooth communication. The CBR approach helps clinicians to make a diagnosis, classification and treatment plan by retrieving the most similar previous case. The case may also be used to follow the treatment progress. Here, the cases are formulated with person’s contextual information and extracted features from sensor signal measurements. The features are extracted considering three domain analysis:1) time domain features using statistical measurement, 2) frequency domain features applying Fast Fourier Transform (FFT), and 3) time-frequency domain features applying Discrete Wavelet Transform (DWT). The initial result is acceptable that shows the advancement of the system while combining the personalized health profiling together with CBR.
|
|
| 8. |
- Ahmed, Mobyen Uddin
(författare)
-
A Personalized Health-Monitoring System for Elderly by Combining Rules and Case-based Reasoning
- 2015
-
Ingår i: Studies in Health Technology and Informatics, Volume 21. ; , s. 249-254
-
Konferensbidrag (refereegranskat)abstract
- Health-monitoring system for elderly in home environment is a promising solution to provide efficient medical services that increasingly interest by the researchers within this area. It is often more challenging when the system is self-served and functioning as personalized provision. This paper proposed a personalized self-served health-monitoring system for elderly in home environment by combining general rules with a case-based reasoning approach. Here, the system generates feedback, recommendation and alarm in a personalized manner based on elderly’s medical information and health parameters such as blood pressure, blood glucose, weight, activity, pulse, etc. A set of general rules has used to classify individual health parameters. The case-based reasoning approach is used to combine all different health parameters, which generates an overall classification of health condition. According to the evaluation result considering 323 cases and k=2 i.e., top 2 most similar retrieved cases, the sensitivity, specificity and overall accuracy are achieved as 90%, 97% and 96% respectively. The preliminary result of the system is acceptable since the feedback; recommendation and alarm messages are personalized and differ from the general messages. Thus, this approach could be possibly adapted for other situations in personalized elderly monitoring.
|
|
| 9. |
- Ahmed, Mobyen Uddin, et al.
(författare)
-
An Overview on the Internet of Things for Health Monitoring Systems
- 2016
-
Ingår i: 2nd EAI International Conference on IoT Technologies for HealthCare HealthyIoT2015. - Cham : Springer International Publishing. ; , s. 429-436
-
Konferensbidrag (refereegranskat)abstract
- The aging population and the increasing healthcare cost in hospitals are spurring the advent of remote health monitoring systems. Advances in physiological sensing devices and the emergence of reliable low-power wireless network technologies have enabled the design of remote health monitoring systems. The next generation Internet, commonly referred to as Internet of Things (IoT), depicts a world populated by devices that are able to sense, process and react via the Internet. Thus, we envision health monitoring systems that support Internet connection and use this connectivity to enable better and more reliable services. This paper presents an overview on existing health monitoring systems, considering the IoT vision. We focus on recent trends and the development of health monitoring systems in terms of: (1) health parameters, (2) frameworks, (3) wireless communication, and (4) security issues. We also identify the main limitations, requirements and advantages within these systems.
|
|
| 10. |
- Ahmed, Mobyen Uddin, 1976-, et al.
(författare)
-
Big Data Analytics in Health Monitoring at Home
- 2017
-
Ingår i: Medicinteknikdagarna 2017 MTD 2017.
-
Konferensbidrag (refereegranskat)abstract
- This paper proposed a big data analytics approach applied in the projects ESS-H and E-care@home in the context of biomedical and health informatics with the advancement of information fusion, data abstraction, data mining, knowledge discovery, learning, and reasoning [1][2]. Data are collected through the projects, considering both the health parameters, e.g. temperature, bio-impedance, skin conductance, heart sound, blood pressure, pulse, respiration, weight, BMI, BFP, movement, activity, oxygen saturation, blood glucose, heart rate, medication compliance, ECG, EMG, and EEG, and the environmental parameters e.g. force/pressure, infrared (IR), light/luminosity, photoelectric, room-temperature, room-humidity, electrical usage, water usage, RFID localization and accelerometers. They are collected as semi-structured/unstructured, continuous/periodic, digital/paper record, single/multiple patients, once/several-times, etc. and stored in a central could server [5]. Thus, with the help of embedded system, digital technologies, wireless communication, Internet of Things (IoT) and smart sensors, massive quantities of data (so called ‘Big Data’) with value, volume, velocity, variety, veracity and variability are achieved [2]. The data analysis work in the following three steps. In Step1, pre-processing, future extraction and selection are performed based on a combination of statistical, machine learning and signal processing techniques. A novel strategy to fuse the data at feature level and as well as at data level considers a defined fusion mechanism [3]. In Step2, a combination of potential sequences in the learning and search procedure is investigated. Data mining and knowledge discovery, using the refined data from the above for rule extraction and knowledge mining, with support for anomaly detection, pattern recognition and regression are also explored here [4]. In Step3, adaptation of knowledge representation approaches is achieved by combining different artificial intelligence methods [3] [4]. To provide decision support a hybrid approach is applied utilizing different machine learning algorithms, e.g. case-based reasoning, and clustering [4]. The approach offers several data analytics tasks, e.g. information fusion, anomaly detection, rules and knowledge extraction, clustering, pattern identification, correlation analysis, linear regression, logic regression, decision trees, etc. Thus, the approach assist in decision support, early detection of symptoms, context awareness and patient’s health status in a personal environment.
|
|
