Yazar "Moghaddamnia, Sanam" seçeneğine göre listele

Listeleniyor 1 - 8 / 8
  • [ X ]
    Öğe
    A Machine Learning Approach to Predict the Sepsis Status: Analyzing the Connection Between Relevant Laboratory Values and Other Physiological Measurements Obtained in Intensive Care Unit
    (2021) Moghaddamnia, Sanam; Bozkurt, Mehmet Recep
    Sepsis is a life-threatening disease caused by the body's response to an infection, which leads to tissue damage, organ failure and death. According to literature, about 1.7 million Americans each year suffer from sepsis with approximately 270,000 fatal cases, and more than one third of mortalities in hospitals are associated with sepsis. It is estimated that worldwide about 30 million people suffer from sepsis each year, and 6 million patients with sepsis die. Above all, the cost of sepsis care is extremely high and constitutes a health risk mostly in developing countries. Having regard to the above, early diagnosis and therapy of sepsis play a key role in increasing the chance of cure for many patients. In this regard, many clinicians and researchers have proposed new definitions for sepsis (Singer et al. 2016), however the subject of early diagnosis methods and their practical usability and limits for sepsis remains to a large extent unexplored. Therefore, a cost-effective self-diagnostic technic has been proposed in this work, which is based on the XGBoost algorithm and deep convolution neural network model for sepsis prediction. Septic data were used for the model training, and the highest accuracy and the area under the ROC curve (AUC) obtained to evaluate the performance of the model were 96.49% and 70.78%, respectively. The novelty of this study lies in the investigations that unlike other studies are performed using vital signs determined independently from laboratory measurements. It has shown that the proposed model is faster and can provide more accurate results, and thus has great potential to be used as an efficient technique in predicting sepsis.
  • Küçük Resim
    Öğe
    Efficiency of deep neural networks for joint angle modeling in digital gait assessment
    (Springer Open, 2021) Alcaraz, Javier Conte; Moghaddamnia, Sanam; Peissig, Jürgen
    Reliability and user compliance of the applied sensor system are two key issues of digital healthcare and biomedical informatics. For gait assessment applications, accurate joint angle measurements are important. Inertial measurement units (IMUs) have been used in a variety of applications and can also provide significant information on gait kinematics. However, the nonlinear mechanism of human locomotion results in moderate estimation accuracy of the gait kinematics and thus joint angles. To develop “digital twins” as a digital counterpart of body lower limb joint angles, three-dimensional gait kinematic data were collected. This work investigates the estimation accuracy of different neural networks in modeling lower body joint angles in the sagittal plane using the kinematic records of a single IMU attached to the foot. The evaluation results based on the root mean square error (RMSE) show that long short-term memory (LSTM) networks deliver superior performance in nonlinear modeling of the lower limb joint angles compared to other machine learning (ML) approaches. Accordingly, deep learning based on the LSTM architecture is a promising approach in modeling of gait kinematics using a single IMU, and thus can reduce the required physical IMUs attached on the subject and improve the practical application of the sensor system.
  • [ X ]
    Öğe
    Efficiency of the Memory Polynomial Model in Realizing Digital Twins for Gait Assessment
    (Ieee, 2019) Conte Alcaraz, Javier; Moghaddamnia, Sanam; Fuhrwerk, Martin; Peissig, Juergen
    One of the key issues of multi-sensory digital healthcare and therapy is the reliability and user compliance of the applied sensor system. In the context of digital gait analysis and rehabilitation, different technologies have been proposed allowing objective gait assessment and precise quantification of the rehabilitation progress using Inertial Measurement Unit (IMU) platforms. However, this depends largely on the estimation accuracy of the kinematics (body joint angles). This paper presents the concept of a digital equivalent based on the Memory Polynomial Model (MPM) to reduce the number of IMUs needed for the measurements and to simulate the physical mechanism of lower body joint angles based on acceleration data. The MPM parameter estimation is based on the Least Square (LS) approach and is performed using accelerometer records of non-pathological gait patterns. The Normalized Mean Square Error (NMSE) is used to evaluate the performance of the model. According to the results an NMSE of -20 dB is achieved, which indicates the great potential of applying the MPM to develop a digital twin. That kind of twin can serve as a prototype of the physical counterpart to improve the wearability of the sensor system and to reduce physically induced measurement errors as well.
  • [ X ]
    Öğe
    Estimation of Blood Calcium and Potassium Values from ECG Records
    (Sciendo, 2024) Babur, Sebahattin; Moghaddamnia, Sanam; Bozkurt, Mehmet Recep
    The identification of diseases caused by changes in ion concentration is quite difficult and yet plays a decisive role in the success of clinical care, diagnosis and treatment. The clinically proven approach to diagnosing electrolyte concentration imbalance is blood tests. There is a need to provide a non-invasive diagnostic method that is not of a temporary nature. Bio-signals such as the electrocardiogram (ECG) can be used to meet this demand and become diagnostic tools that facilitate home monitoring of electrolyte concentration on a permanent basis. This study investigates the feasibility and efficiency of methods based on machine learning (ML) and ECG recordings in monitoring critical levels of existing potassium and calcium concentration. Morphological, frequency and frequency-time domain features were extracted to automatically estimate calcium and potassium levels. Furthermore, the potential of estimates based on modeling approaches will be demonstrated to gain insights into relevant clinical findings and improve the performance of monitoring approaches. Using the hold-out validation method, the best results in terms of mean square error (MSE) and R for estimating the calcium value are 0.7157 and 0.57347, using fuzzy inference systems (FIS). Here, R represents the proportion of the variance in the calcium value that is explained by the model.
  • Küçük Resim
    Öğe
    Functional quality assessment of whole-body vibration training devices based on instantaneous amplitude and frequency of photogrammetric vibration measurements
    (Medical Engineering & Physics, 2023) Moghaddamnia, Sanam; Rofallski, Robin; Luhmann, Thomas; Kaeding, Tobias S.
    The practical use of whole-body vibration training (WBVT) and such research may be negatively influenced by generated vibrations with amplitudes, frequencies, and/or patterns that deviate from preset adjustments on WBVT devices. This study examined whether prolonged regular use can generate respective deviations. Four WBVT devices, used for 19 months in a research project on the effects of WBVT, were analyzed using photo grammetry before start of the research project and after 19 months. Divergences between preset and measured amplitudes and frequencies were calculated for all measurements. To quantify how well the output of devices correlates with the target setting, the vibration characteristics were calculated. In particular, exact long-term measurements related to the vibration amplitude is conducted and analyzed for the first time, which has been found as an important measure of the device functional quality. One device had a significantly (p < 0.01) larger machine run time than the other three. This one showed the most pronounced signs of functional impairments concerning instantaneous amplitudes, frequencies and the mode of vibration after prolonged use. These results based on photometric measurements underline again that prolonged use can result in divergences between preset and actual applied amplitudes, frequencies, mode of vibration and other accuracy measurement metrics.
  • Küçük Resim
    Öğe
    Monitoring the rehabilitation progress using a DCNN and kinematic data for digital healthcare
    (Institute of Electrical and Electronics Engineers, 2021) Alcaraz, Javier Conte; Moghaddamnia, Sanam; Penner, Maxim; Peissig, Jürgen
    Monitoring the progress of patients during the rehabilitation process after an operation is beneficial for adjusting care and medical treatment in order to improve the patient's quality of life. The supervised methods used for this in the literature need data labeling, which is a time and cost-intensive procedure. In this paper, we propose Deep Convolutional Neural Network (DCNN) for monitoring the progress of the rehabilitation, utilizing the kinematic data from a Wearable Sensor System (WSS). The WSS provides three-dimensional linear acceleration and angular velocity from multiple body parts such as the lower back and lower limbs during walking at any speed on level ground. Twelve patients with hip unilateral arthroplasty completed two weeks of gait training after the operation. The classification results of different Inertial Measurement Unit (IMU) placements revealed that the IMU placed at thigh achieved the highest accuracy. The proposed DCNN achieved up to 98% classification accuracy for the rehabilitation progress monitoring. This approach provides an objective and evidence-based way of understanding clinically important changes in human movement patterns in response to exercise therapy.
  • Küçük Resim
    Öğe
    On the efficiency of LSTM in classifying musical impressions from EEG recordings
    (2023) Kaya, Burak; Habiboğlu, Mehmet Gökhan; Moghaddamnia, Sanam
    The objective of this study is the classification of musical impressions with long short-term memory (LSTM) approach using EEG recordings of 20 subjects, while listening to different music genres. For this purpose, a deep learning model was developed, where relevant features extracted from intrinsic mode functions (IMF) of the clean EEG data are used as the input signals. The classification accuracy of the proposed model is evaluated with various feature sets. The highest classification accuracy is 73.33%, which is achieved by combining higher-order statistics and the first difference of IMF features.
  • Küçük Resim
    Öğe
    Parameter and feature selection in decision trees for the classification of musical impressions from EEG records
    (2023) Ozaltun, Emir Atakan; Moghaddamnia, Sanam; Habiboğlu, Mehmet Gökhan
    Reliable classification of different emotions is an important issue for emotional interaction between humans and computers. Therefore, this study aims at assessing the performance of decision trees in classifying musical impressions from EEG records of 20 subjects, who listened to songs in different music styles. First, features extracted from the clean EEG data used to train the classifier, where different feature combinations and parameter settings are considered. Next, the impact of various hyperparameter values on the classification accuracy is examined and the relevant feature combination is specified. According to the results, an accuracy rate of 76,12% is achieved, when all time domain features are included in the classification.