Marateb, H.R.,
Roohafza, H.,
Noohi, F.,
Hosseini S.G.,
Alemzadeh-Ansari M.,
Bagherieh, S.,
Mansourian, M.,
Mousavi, A.F.,
Seyedhosseini S.M.,
Farshidi H.,
Ahmadi N.,
Yazdani A.,
Sadeghi, M. Publication Date: 2023
Current Problems in Cardiology (1462806)(7)
This study aims to provide a comprehensive risk-assessment model including lifestyle, psychological parameters, and traditional risk factors to determine the risk of major adverse cardiovascular events (MACE) in patients with the first acute ST-segment elevation myocardial infarction episode. Patients were recruited from new hospital admissions of acute ST-segment elevation myocardial infarction and will be followed up to 3 years. Clinical and paraclinical characteristics, lifestyle, psychological, and MACE information are collected and will be used in the risk-assessment model. Totally, 1707 patients were recruited (male: 81.4%, mean age: 56.60 ± 10.34). Primary percutaneous coronary intervention was the most prevalent type of coronary revascularization (81.9%). In case of baseline psychological characteristics, mean depression score was 5.40 ± 4.88, and mean distress score was 7.64 ± 5.08. A comprehensive approach, focusing on medical, lifestyle, and psychological factors, will lead to better identification of cardiovascular disease patients at risk of developing MACE through comprehensive risk-assessment models. © 2022 Elsevier Inc.
Publication Date: 2019
Journal Of Medical Signals And Sensors (22287477)9(4)pp. 227-233
Background: Decellularization techniques have been widely used in tissue engineering recently. However, applying these methods which are based on removing cells and maintaining the extracellular matrix (ECM) encountered some difficulties for dense tissues such as articular cartilage. Together with chemical agents, using physical methods is suggested to help decellularization of tissues. Methods: In this study, to improve decellularization of articular cartilage, the effects of direct and indirect ultrasonic waves as a physical method in addition to sodium dodecyl sulfate (SDS) as chemical agents with 0.1% and 1% (w/v) concentrations were examined. Decellularization process was evaluated by nucleus staining with hematoxylin and eosin (H and E) and by staining glycosaminoglycans (GAG) and collagen. Results: The H and E staining indicated that 1% (w/v) SDS in addition to ultrasonic bath for 5 h significantly decreased the cell nucleus residue to lacuna ratio by 66%. Scanning electron microscopy showed that using direct sonication caused formation of micropores on the surface of the sample which results in better penetration of decellularization material and better cell attachment after decellularization. Alcian Blue and Picrosirius Red staining represented GAG and collagen, respectively, which maintained in ECM structure after decellularization by ultrasonic bath and direct sonicator. Conclusion: Ultrasonic bath can help better penetration of the decellularization material into the cartilage. This improves the speed of the decellularization process while it has no significant defect on the structure of the tissue. © 2019 Journal of Medical Signals & Sensors.
Publication Date: 2021
Diagnostics (20754418)(3)
The World Health Organization (WHO) suggests that mental disorders, neurological disorders, and suicide are growing causes of morbidity. Depressive disorders, schizophrenia, bipolar disorder, Alzheimer’s disease, and other dementias account for 1.84%, 0.60%, 0.33%, and 1.00% of total Disability Adjusted Life Years (DALYs). Furthermore, suicide, the 15th leading cause of death worldwide, could be linked to mental disorders. More than 68 computer-aided diagnosis (CAD) methods published in peer-reviewed journals from 2016 to 2021 were analyzed, among which 75% were published in the year 2018 or later. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol was adopted to select the relevant studies. In addition to the gold standard, the sample size, neuroimaging techniques or biomarkers, validation frameworks, the classifiers, and the performance indices were analyzed. We further discussed how various performance indices are essential based on the biostatistical and data mining perspective. Moreover, critical information related to the Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) guidelines was analyzed. We discussed how balancing the dataset and not using external validation could hinder the generalization of the CAD methods. We provided the list of the critical issues to consider in such studies. © 2021 by the authors. Licensee MDPI, Basel, Switzerland.
Publication Date: 2022
2025 29th International Computer Conference, Computer Society of Iran, CSICC 2025pp. 23-74
Brain-computer interface (BCI) aims to translate human intention into a control output signal. In motor-imaginary (MI) BCI, the imagination of movement modifies the cortex brain activity. Such activities are then used in pattern recognition to identify certain movement classes. MI-BCI could be used to enhance the life quality of physically impaired subjects. Several challenges exist in MI-BCI, including selecting appropriate channels, usually linked with a suitable classifier choice. The entire procedure must be real time in practical applications. A variety of channel selection and classification methods were used for MI-BCI in the literature. Also, hybrid machine learning (ML) and deep learning (DL) methods were used in the literature. In this chapter, different channel selection, ML and DL methods, validation frameworks, and performance indices of EEG-based methods were investigated. Three hundred and twenty-two papers published between January 2000 and March 2021 were analyzed in this systematic review. Specific challenges and future directions were then provided. © 2022 Elsevier Inc. All rights reserved.
Karimian, A.,
Thompson, C.J.,
Sarkar s., ,
Raisali g., ,
Pani r., ,
Davilu h., ,
Sardari d., Publication Date: 2004
IEEE Nuclear Science Symposium Conference Record (10957863)4pp. 2339-2341
Early detection of breast cancer is very important for efficient and effective treatment. Conventional whole body PET (WB-PET) scanners provide metabolic images of breast cancer with several shortcomings related to the general-purpose nature of these systems. We propose a Cylindrical Breast PET (CYBPET) system for breast imaging, for the patients in the prone position. An individual pendulous breast is covered by a thin plastic to make a mild pressure and surrounded by the crystals inside the CYBPET ring and will positioned in the center of field of view. Each breast is imaged separately. The rest of the body has been shielded properly to minimize the contribution of scattered photons from the other breast and the rest of the body. This proposed system configuration has several advantages over the WB-PET system configuration for breast imaging, namely: I) greatly reduced attenuation and scattering effects in breast imaging; II) reduced interference from activity in parts the torso in the FOV; III) elimination a separate attenuation scan due to the homogeneity of breast; IV) reduced number of detectors and its' related electronics to reduce the price of the system. Before building the CYBPET instrument, we have simulated it using the PETSIM Monte Carlo program. Simulations of CYBPET and a (WB-PET) for a 8mm tumor inside the breast with a lesion to background activity concentration of 10 to 1 were made. The NECR (max) for CYBPET (WB-PET) were 203(184) (kcps) in the activity concentration of 3.2(10.4) (μCi/cc). Also the coincidence efficiency show the amounts of 470 and 132 (kcps/μCi/cc) for CYBPET and WB-PET respectively. These results and the advantages listed above suggest improved lesion visualization in the early disease stages. The only shortcoming, like the PEM, SPEM and manunography systems is, this system would not be able to detect the lesions which are very close to chest wall. © 2004 IEEE.
In this article, a new framework is proposed to address multi-class Motor Imagery Brain-Computer Interface (MIBCI) problems containing a small portion of labeled datasets. In this framework, the combination of Independent Component Analysis (ICA), multi-class Common Spatial Pattern (CSP), and a functional Application Programming Interface (API) model assumes a pivotal role. In the feature extraction stage of the work, a concatenated altered signal affected by spatial weights is proposed for each trial in three frequency ranges. This distribution of features can both provide suitable feature maps for augmentation, preparing data for the deep learning analysis, and underscore distinguishable features of MI classes. In the classification stage, spatial and temporal features are dominated by using the effective combination of a one-dimensional Convolutional Neural Network (CNN) and a two-staged Bidirectional Long Short-Term Memory (BLSTM) in three branches containing different distributions of frequency. Given that, the model simultaneously learns past-tofuture and future-to-past patterns in two stages. The experimental result on datasets 2a BCI-Competition IV illustrates that the proposed method can be liable, practical and more competitive than the other popular methods pointed out in this paper. All in all, the proposed framework can alleviate the issue of small portions of labeled datasets in MI problems.
Publication Date: 2022
Applied Biochemistry and Biotechnology (02732289)194(5)pp. 2077-2092
The prevalence of diabetes has increased over the past years. Therefore, developing minimally invasive, user-friendly, and cost-effective glucose biosensors is necessary especially in low-income and developing countries. Cellulose paper–based analytical devices have attracted the attention of many researchers due to affordability, not requiring trained personnel, and complex equipment. This paper describes a microfluidic paper-based analytical device (μPAD) for detecting glucose concentration in tear range with the naked eye. The paper-based biosensor fabricated by laser CO2; and glucose oxidase/horseradish peroxidase (GOx/HRP) enzyme solution coupled with tetramethylbenzidine (TMB) were utilized as reagents. A sample volume of 10 μl was needed for the biosensor operation and the results were observable within 5 min. The color intensity–based and distance-based results were analyzed by ImageJ and Tracker to evaluate the device performance. Distance-based results showed a linear behavior in 0.1–1.2 mM with an R2 = 0.9962 and limit of detection (LOD) of 0.1 mM. The results could be perceived by the naked eye without needing additional equipment or trained personnel in a relatively short time (3–5 min). © 2022, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.
Publication Date: 2013
Nuclear Technology and Radiation Protection (14513994)28(3)pp. 278-283
Ionizing particles have been used for the treatment of atherosclerosis. Internal irradiation is commonly carried out by means of several methods (catheter-based systems, radioactive stents or balloons) to reduce the probability of restenosis. 90Y, due to some of its characteristics, is an appropriate radioisotope for intravascular brachytherapy. However, since there are some critical tissues in the vicinity of the heart like the breast and lymph nodes, it is necessary to perform a dosimetry calculation around the artery under radiotherapy to justify the treatment method. In this study, a 3-D dose distribution was obtained for the coronary vessel and its surrounding tissues for a standard 90Y stent in a MCNPX program. The results were compared with other investigations on restenosis prevention using 90Y-coated stents. The calculations represented a 28-day cumulative dose between 1230 cGy and 2400 cGy at 0.1 mm from the stent surface, while this quantity was about 23.8 cGy at 8.5 mm from the stent surface. An assessment of the dose equivalent and effective dose was also performed at r = 8.5 mm for the mentioned surrounding tissues which may be located in the area, based on the latest changes in ICRP recommendations. Additionally, the dose equivalent calculated within the treatment period for these organs was compared with published dotimetry data for 90Sr/90Y seed sources in order to evaluate radiation protection concerns about these two radiotherapy methods. It has been found that, depending on stent parameters, 90Y stent implantation might increase the unfavorable side effects for the patient, but to a much lesser degree than the other methods.
Publication Date: 2025
Magnetic Resonance Materials in Physics, Biology and Medicine (13528661)38(2)pp. 299-315
Objective: This study presents a novel deep learning-based framework for precise brain MR region segmentation, aiming to identify the location and the shape details of different anatomical structures within the brain. Materials and methods: The approach uses a two-stage 3D segmentation technique on a dataset of adult subjects, including cognitively normal participants and individuals with cognitive decline. Stage 1 employs a 3D U-Net to segment 13 brain regions, achieving a mean DSC of 0.904 ± 0.060 and a mean HD95 of 1.52 ± 1.53 mm (a mean DSC of 0.885 ± 0.065 and a mean HD95 of 1.57 ± 1.35 mm for smaller parts). For challenging regions like hippocampus, thalamus, cerebrospinal fluid, amygdala, basal ganglia, and corpus callosum, Stage 2 with SegResNet refines segmentation, improving mean DSC to 0.921 ± 0.048 and HD95 to 1.17 ± 0.69 mm. Results: Statistical analysis reveals significant improvements (p-value < 0.001) for these regions, with DSC increases ranging from 1.3 to 3.2% and HD95 reductions of 0.06–0.33 mm. Comparisons with recent studies highlight the superior performance of the performed method. Discussion: The inclusion of a second stage for refining the segmentation of smaller regions demonstrates substantial improvements, establishing the framework’s potential for precise and reliable brain region segmentation across diverse cognitive groups. © The Author(s), under exclusive licence to European Society for Magnetic Resonance in Medicine and Biology (ESMRMB) 2025.
Publication Date: 2011
Annals of DAAAM and Proceedings of the International DAAAM Symposium (17269679)pp. 167-168
In this paper, the characteristics of the human perception system as well as the image features are exploited in a dual-resolution vision system for segmentation/object detection in outdoor scenes. The texture details are deliberately removed and similar color shades are combined together in a low-resolution version of the image, to reduce the excess image information. Using a color clustering algorithm, the color regions of the low-resolution image are found. Then, a weighted graph is constructed, whose nodes contain the detailed features of the regions, derived from the highresolution image. The weight of the edge between two nodes, Nodes' Merging Potential (NMP), denotes the advantage of merging them together to construct the fundamental image regions. This graph is then pruned regarding the NMP values, so that the main segments are developed and then identified. The proposed algorithm has shown high speed and accuracy for segmentation/object detection in outdoor scenes.
Nazem, Fatemeh,
Ghasemi, Fahimeh,
Fassihi, Afshin,
Rasti, Reza,
Nazem, F.,
Ghasemi, F.,
Fassihi, A.,
Rasti, R.,
Dehnavi, A.M. Publication Date: 2023
JOURNAL OF MEDICAL SIGNALS & SENSORS (22287477)13(1)pp. 1-10
Background: The first step in developing new drugs is to find binding sites for a protein structure that can be used as a starting point to design new antagonists and inhibitors. The methods relying on convolutional neural network for the prediction of binding sites have attracted much attention. This study focuses on the use of optimized neural network for three-dimensional (3D) non-Euclidean data. Methods: A graph, which is made from 3D protein structure, is fed to the proposed GU-Net model based on graph convolutional operation. The features of each atom are considered as attributes of each node. The results of the proposed GU-Net are compared with a classifier based on random forest (RF). A new data exhibition is used as the input of RF classifier. Results: The performance of our model is also examined through extensive experiments on various datasets from other sources. GU-Net could predict the more number of pockets with accurate shape than RF. Conclusions: This study will enable future works on a better modeling of protein structures that will enhance knowledge of proteomics and offer deeper insight into drug design process.
Mehdikhani, M.,
Yilgör, P.,
Poursamar, S.A.,
Etemadi, N.,
Gokyer, S.,
Navid, S.,
Farzan, M.,
Farzan, M.,
Babaei, M.,
Rafienia m., M. Publication Date: 2024
International Journal of Biological Macromolecules (01418130)282
Skin injuries resulting from physical trauma pose significant health risks, necessitating advanced wound care solutions. This investigation introduces an innovative bilayer wound dressing composed of 3D-printed propolis-coated polycaprolactone (PCL/PP) and an electrospun composite of polyvinyl alcohol, chitosan, polycaprolactone, and diltiazem (PVA/CTS/PCL/DTZ). SEM analysis revealed a bilayer structure with 89.23 ± 51.47 % porosity and uniformly distributed nanofibers. The scaffold tensile strength, with pore sizes of 100, 300, and 500 μm, was comparable to native skin. However, smaller pore sizes reduced water vapor transmission from 4211.59 ± 168.53 to 2358.49 ± 203.63 g/m2. The incorporation of DTZ lowered the contact angle to 35.23 ± 3.65°, while the addition of PCL reduced the degradation rate and modulated the release of DTZ by approximately 50 %. Moreover, lower pH increased the degradation rate and decreased swelling. The inclusion of propolis enhanced antibacterial activity, and 10 % DTZ promoted the viability, proliferation, and migration of fibroblasts and adipose-derived stem cells. However, increasing DTZ concentration to 12 % reduced cell viability. In vivo tests on rats demonstrated effective wound healing and anti-inflammatory properties of the bilayer samples. Regarding the aforementioned results, the PCL/PP-PVA/CTS/PCL/DTZ (10 % w/w) bilayer wound dressing is a promising candidate for wound healing applications. © 2024
Marateb, H.R.,
Mohebbian, M.R.,
Mansourian, M.,
Mañanas, M.A.,
Mokarian, F. Publication Date: 2017
Computational and Structural Biotechnology Journal (20010370)pp. 75-85
Cancer is a collection of diseases that involves growing abnormal cells with the potential to invade or spread to the body. Breast cancer is the second leading cause of cancer death among women. A method for 5-year breast cancer recurrence prediction is presented in this manuscript. Clinicopathologic characteristics of 579 breast cancer patients (recurrence prevalence of 19.3%) were analyzed and discriminative features were selected using statistical feature selection methods. They were further refined by Particle Swarm Optimization (PSO) as the inputs of the classification system with ensemble learning (Bagged Decision Tree: BDT). The proper combination of selected categorical features and also the weight (importance) of the selected interval-measurement-scale features were identified by the PSO algorithm. The performance of HPBCR (hybrid predictor of breast cancer recurrence) was assessed using the holdout and 4-fold cross-validation. Three other classifiers namely as supported vector machines, DT, and multilayer perceptron neural network were used for comparison. The selected features were diagnosis age, tumor size, lymph node involvement ratio, number of involved axillary lymph nodes, progesterone receptor expression, having hormone therapy and type of surgery. The minimum sensitivity, specificity, precision and accuracy of HPBCR were 77%, 93%, 95% and 85%, respectively in the entire cross-validation folds and the hold-out test fold. HPBCR outperformed the other tested classifiers. It showed excellent agreement with the gold standard (i.e. the oncologist opinion after blood tumor marker and imaging tests, and tissue biopsy). This algorithm is thus a promising online tool for the prediction of breast cancer recurrence. © 2016 The Authors
Marateb, H.R.,
Mansourian, M.,
Faghihimani E.,
Amini, M.,
Farina D. Publication Date: 2014
Computers in Biology and Medicine (18790534)(1)pp. 34-42
Microalbuminuria (MA) is an independent predictor of cardiovascular and renal disease, development of overt nephropathy, and cardiovascular mortality in patients with type 2 diabetes. Detecting MA is an important screening tool to identify people with high risk of cardiovascular and kidney disease. The gold standard to detect MA is measuring 24-h urine albumin excretion. A new method for MA diagnosis is presented in this manuscript which uses clinical parameters usually monitored in type 2 diabetic patients without the need of an additional measurement of urinary albumin. We designed an expert-based fuzzy MA classifier in which rule induction was performed by particle swarm optimization. A variety of classifiers was tested. Additionally, multiple logistic regression was used for statistical feature extraction. The significant features were age, diabetic duration, body mass index and HbA1C (the average level of blood sugar over the previous 3 months, which is routinely checked every 3 months for diabetic patients). The resulting classifier was tested on a sample size of 200 patients with type 2 diabetes in a cross-sectional study. The performance of the proposed classifier was assessed using (repeated) holdout and 10-fold cross-validation. The minimum sensitivity, specificity, precision and accuracy of the proposed fuzzy classifier system with feature extraction were 95%, 85%, 84% and 92%, respectively. The proposed hybrid intelligent system outperformed other tested classifiers and showed "almost perfect agreement" with the gold standard. This algorithm is a promising new tool for screening MA in type-2 diabetic patients. © 2013 Elsevier Ltd.
We proposed an automatic hybrid image segmentation model that integrates the modified statistical expectation maximization (EM) method and the spatial information combined with Support Vector Machines (SVM). To improve the overall segmentation performance different types of information are integrated in this study, which are, voxel location, textural features, MR intensity and relationship with neighboring voxels. The modified EM method is used for intensity based classification as an initial segmentation stage. Secondly simple and beneficial features are extracted from target area of segmented image using gray-level co-occurrence matrix (GLCM) technique. Subsequently, we applied Support Vector Machine (SVM) to rank computed features from the extracted features, which is an enhancement step. To evaluate the performance of the proposed method, experiments carried out on real MRI. The results of proposed method are evaluated against manual segmentation results on real scans. The K-index is calculated to evaluate the performance of the proposed model relative to the expert segmentations. The results demonstrate that the proposed technique has satisfactory results. © 2015 IEEE.
Alamatsaz, N.,
Tabatabaei, L.,
Yazdchi, M.,
Payan, H.,
Alamatsaz, N.,
Nasimi, F. Publication Date: 2024
Biomedical Signal Processing and Control (17468108)90
Objective: Electrocardiogram (ECG) is the most frequent and routine diagnostic tool used for monitoring heart electrical signals and evaluating its functionality. The human heart can suffer from a variety of diseases, including cardiac arrhythmias. Arrhythmia is an irregular heart rhythm that in severe cases can lead to stroke and can be diagnosed via ECG recordings. Since early detection of cardiac arrhythmias is of great importance, computerized and automated classification and identification of these abnormal heart signals have received much attention for the past decades. Methods: This paper introduces a light Deep Learning (DL) approach for high accuracy detection of 8 different cardiac arrhythmias and normal rhythms. To employ DL techniques, the ECG signals were preprocessed using resampling and baseline wander removal techniques. The classification was performed using an 11-layer network employing a combination of Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM). Results: In order to evaluate the proposed technique, ECG signals are chosen from the two physionet databases, the MIT-BIH arrhythmia database and the long-term AF database. The proposed DL framework based on the combination of CNN and LSTM showed promising results than most of the state-of-the-art methods. The proposed method reaches the mean diagnostic accuracy of 98.24%. Conclusion: A trained model for arrhythmia classification using diverse ECG signals were successfully developed and tested. Significance: This study presents a lightweight classification technique with high diagnostic accuracy compared to other notable methods, making it a potential candidate for implementation in Holter monitor devices for arrhythmia detection. Finally, we used SHapley Additive exPlanations (SHAP), the most popular Explainable Artificial Intelligence (XAI) method to understand how our model make predictions. The results indicate that those features (ECG samples) that have contributed the most to a prediction are consonant with clinicians’ decisions. Therefore, the use of interpretable models increases the trust of clinicians in AI and thus leads to decreasing the number of misdiagnoses of cardiovascular diseases. © 2023 Elsevier Ltd
Publication Date: 2005
2025 29th International Computer Conference, Computer Society of Iran, CSICC 2025pp. 2336-2348
One of the most important aspects of agent design in Al is the way agent acts or responds to the environment that the agent is acting upon. An effective action selection and behavioral method gives a powerful advantage in overall agent performance. We define a new method of action selection based on probability/priority models, we thereby introduce an efficient way to determine probabilities and a new priority based system which maps the human knowledge to action selec tion method. Furthermore, a behavior model is introduced to make the model more flexible. Copyright © IICAI 2005.
Karimi, M.T.,
Jamshidi, N.,
Bahreinizad, H.,
Bani, M.S.,
Omar, A.H.H. Publication Date: 2014
Work (18759270)49(4)pp. 663-668
BACKGROUND: Stability during standing is achieved by a complex process which involves the performance of various systems. Using a force plate for analysing the stability for a period of one minute has been reported exclusively by many investigators. Most of people stand for a long period of time when chatting with somebody, doing a job and when waiting in a queue. However nobody has analysed the stability during quiet standing for a prolonged standing (5 minutes). OBJECTIVE: The main aim of this research study was to analyse the performance of the subjects regarding stability for a period of 5 minutes. METHOD: A group of 40 normal subjects from the staff and students of Rehabilitation Faculty of Isfahan University of Medical Sciences were recruited in this research project. They were asked to stand on the force plate (Kistler) for a period of 5 minutes. They were instructed to look straight ahead and with their head erect and their arms at their sides in a comfortable position. The excursions of the COP sway in both planes were measured for all 20 seconds periods of data collection. RESULTS: The results of this research study showed that stability analysing based on the sway of the COP, while the test was collected for one minute, is not recommended. There is a significant difference between the excursions of the COP during the first to fifth minutes. The stability of the subject was optimum in the third and fourth minutes of standing. CONCLUSION: Using the COP sway, based on the first minute of standing, is neither a good representative of the more stable position nor the unstable position. It is recommended to discuss the stability of subjects based on their ability to return from an unstable position to a more stable position. © 2014 -IOS Press and the authors.
Publication Date: 2007
Scientia Iranica (23453605)14(6)pp. 571-578
One of the most important challenges in automatic speech recognition is the case of mismatch between training and test data. Conventional methods for improving recognition robustness seek to eliminate or reduce the mismatch, e.g. enhancement of the speech by adapting the statistical models. Training the model in different situations is another example of these methods. The success with these techniques has been moderate compared to human performance. In this paper, an inspiration from human listeners created the motivation to develop and implement a new bidirectional neural network. This network is capable of modeling the phoneme sequence, using bidirectional connections in an isolated word recognition task. This network can correct the phoneme sequence obtained from the acoustic model to what is learned in the training phase. Acoustic feature vectors are enhanced, based on the inversion techniques in neural networks, by cascading the lexical and the acoustic model. Speech enhancement by this method has a remarkable effect in eliminating mismatch between the training and test data. The efficiency of the lexical model and speech enhancement was observed by a 17.3 percent increase in the phoneme recognition correction ratio. © Sharif University of Technology, December 2007.
Bahrami, A.,
Karimian, A.,
Fatemizadeh, E.,
Arabi, H.,
Zaidi, H. Publication Date: 2020
Medical Physics (24734209)47(10)pp. 5158-5171
Purpose: Despite the proven utility of multiparametric magnetic resonance imaging (MRI) in radiation therapy, MRI-guided radiation treatment planning is limited by the fact that MRI does not directly provide the electron density map required for absorbed dose calculation. In this work, a new deep convolutional neural network model with efficient learning capability, suitable for applications where the number of training subjects is limited, is proposed to generate accurate synthetic computed tomography (sCT) images from MRI. Methods: This efficient convolutional neural network (eCNN) is built upon a combination of the SegNet architecture (a 13-layer encoder-decoder structure similar to the U-Net network) without softmax layers and the residual network. Moreover, maxpooling indices and high resolution features from the encoding network were incorporated into the corresponding decoding layers. A dataset containing 15 co-registered MRI-CT pairs of male pelvis (1861 two-dimensional images) were used for training and evaluation of MRI to CT synthesis process using a fivefold cross-validation scheme. The performance of the eCNN model was compared to an atlas-based sCT generation technique as well as the original U-Net model considering CT images as reference. The mean error (ME), mean absolute error (MAE), Pearson correlation coefficient (PCC), structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR) metrics were calculated between sCT and ground truth CT images. Results: The eCNN model exhibited effective learning capability using only 12 training subjects. The model achieved a ME and MAE of 2.8 ± 10.3 and 30.0 ± 10.4 HU, respectively, which is substantially lower than values achieved by the atlas-based (−0.8 ± 35.4 and 64.6 ± 21.2) and U-Net (7.4 ± 11.9 and 44.0 ± 8.8) methods, respectively. Conclusion: The proposed eCNN model exhibited efficient convergence rate with a low number of training subjects, while providing accurate synthetic CT images. The eCNN model outperformed the original U-Net model and showed superior performance to the atlas-based technique. © 2020 American Association of Physicists in Medicine
Publication Date: 2025
Scientific Reports (20452322)15(1)
In PET systems, the SNR relies on the coincidence time resolution (CTR) of 511 keV photon pairs. This research investigates the impact of reflectors, surface treatments, materials, and scintillation crystal length on the CTR of a brainPET detector using dual-layer offset scintillators (DLOs). This study is based on a brainPET, under development at the University of Manitoba, to propose a new design to achieve an improved CTR. Four different pairs of LYSO crystals with distinct optical compositions, surface treatments, and reflective materials were simulated (using GATEv9.3). Each model comprises two LYSO crystal with dimensions of 3 × 3 × 10 mm3. Considering the initial experimental data from the brainPET lab, simulation results showed that the crystal with a roughened surface and ESR reflector demonstrated 13.6% energy resolution and an average 17.8% improvement in CTR compared to other models. In addition, a more comprehensive model, including a dual-layer offset detector was designed. The bottom and top layers have 25 × 19 and 24 × 18 crystals with thickness of 12 and 8 mm, respectively in the DLO model. The simulation investigation showed that the DLO configuration could enhance the time resolution by 17.5% and the energy resolution by 5.4% which are considerably comparable to the state-of-the-art brainPET systems. © The Author(s) 2025.
Publication Date: 2016
Journal of Mechanics in Medicine and Biology (17936810)16(3)
The dynamic study of frog's swimming style contributes to the modeling of the nature-inspired robots. To study the torque matrix produced in the joints during continuous modeling, the dynamic model of the Xenopus laevis swimming is reproduced in the coronal plane. The necessary kinematic data for the modeling is extracted from the frog movement graphs and diagrams during swimming. In the dynamic model, legs are considered as a group of rigid links. In order to verify this method, the generated forward force in half a cycle is studied. Unlike the previous studies, the role of geometry, dimensions and mechanical properties of the legs' fundamental links in generating thrust force is modeled in this study, leading to finding the most proper form for this mechanism design. © 2016 World Scientific Publishing Company.
Karimian, A.,
Mohamadrezaee m., ,
Saddadi, F.,
Forozani g.h., Publication Date: 2010
Iranian Journal of Physics Research (16826957)10(3)pp. 177-185
In nuclear medicine, studies of important tissues such as cardiac, the emitted photons from the cardiac before reaching the gamma detectors are attenuated and scattered by other tissues inside the thorax. Therefore, the quality and contrast of the image will be reduced. In this research, to improve the quality of cardiac images by SPECT system, the most convenient algorithms for attenuation correction were studied and assessed in the first step. Then the best method using the line source in Transmission Attenuation Correction (TAC) method was modified and the experimental data wase obtained by using this new and modified method, cardiac phantom, Dual Head SPECT system and a line source of 201Tl with the activity of about 0.5 mCi. The data was collected and obtained in two steps: (1) Scanning the cardiac phantom and line source which was beside the cardiac phantom this step involves using emission and transmission simultaneously. (2) Scanning the cardiac phantom in the absence of line source which means using emission data. Next, the suggested attenuation correction formula was used and the calculated attenuation coefficient for each pixel was calculated and applied to each pixel. Our results showed a nice improvement in contrast and visibility of the images by this simple and in imporoved expensive method. The advantages of this method include simplicity, the available radionuclide, improved accuracy, quality and contrast of the final image, and finally, cost - effectiveness. These advantages may help the nuclear medicine centers to improve their ability to detect the physiological and functional defects of the cardiac, especially in the elder and women patients.
Publication Date: 2013
Iranian Journal of Medical Physics (17357241)10(3)pp. 139-146
Introduction: In CT imaging, metallic implants inside the tissues cause metal artifact that reduce the quality of image for diagnosis. In order to reduce the effect of this artifact, a new method with more appropriate results has been presented in this research work. Materials and Methods: The presented method comprised of following steps: a) image enhancement and metal areas extraction, b) sinogram transform of original image, c) metal segments and metal traces inside the sinogram transform of original image segmented by using Fuzzy C means, d) interpolation of metal traces inside the original image sinogram and filtering, and e) adding the image of metal parts to the filtered image to obtain the corrected image. Results: Fifty CT scan images from Alzahra Hospital in Isfahan were used to evaluate the proposed method. The proposed method was applied to images which had implants in regions such as femur, hip, tooth, brain, and stomach. The results showed an intensively reduced in metal artifact and quality improvement of images till 90% for accuracy, compared with the radiologist report. Conclusion: The proposed method reduced the effect of metal artifact by maintaining the specification of other tissues. Furthermore, the consumed time to process the suggested algorithm in this study was less than conventional methods. For instance, the consumed time for CT image, including a metal in the femur region was about 20% of the conventional method.
Publication Date: 2015
Journal Of Medical Signals And Sensors (22287477)5(4)pp. 238-244
Automatic segmentation of multiple sclerosis (MS) lesions in brain magnetic resonance imaging (MRI) has been widely investigated in the recent years with the goal of helping MS diagnosis and patient follow-up. In this research work, Gaussian mixture model (GMM) has been used to segment the MS lesions in MRIs, including T1-weighted (T1-w), T2-w, and T2-fluid attenuation inversion recovery. Usually, GMM is optimized by using expectation-maximization (EM) algorithm. The drawbacks of this optimization method are, it does not converge to optimal maximum or minimum and furthermore, there are some voxels, which do not fit the GMM model and have to be rejected. So, GMM is time-consuming and not too much efficient. To overcome these limitations, in this research study, at the first step, GMM was applied to segment only T1-w images by using 100 various starting points when the maximum number of iterations was considered to be 50. Then segmentation results were used to calculate the parameters of the other two images. Furthermore, FAST-trimmed likelihood estimator algorithm was applied to determine which voxels should be rejected. The output result of the segmentation was classified in three classes; White and Gray matters, cerebrospinal fluid, and some rejected voxels which prone to be MS. In the next phase, MS lesions were detected by using some heuristic rules. This new method was applied on the brain MRIs of 25 patients from two hospitals. The automatic segmentation outputs were scored by two specialists and the results show that our method has the capability to segment the MS lesions with dice similarity coefficient score of 0.82. The results showed a better performance for the proposed approach, in comparison to those of previous works with less time-consuming. © 2015 Journal of Medical Signals & Sensors.
Publication Date: 2015
Journal Of Research In Medical Sciences (17351995)(3)pp. 214-223
Background: Coronary heart diseases/coronary artery diseases (CHDs/CAD), the most common form of cardiovascular disease (CVD), are a major cause for death and disability in developing/developed countries. CAD risk factors could be detected by physicians to prevent the CAD occurrence in the near future. Invasive coronary angiography, a current diagnosis method, is costly and associated with morbidity and mortality in CAD patients. The aim of this study was to design a computer-based noninvasive CAD diagnosis system with clinically interpretable rules. Materials and Methods: In this study, the Cleveland CAD dataset from the University of California UCI (Irvine) was used. The interval-scale variables were discretized, with cut points taken from the literature. A fuzzy rule-based system was then formulated based on a neuro-fuzzy classifier (NFC) whose learning procedure was speeded up by the scaled conjugate gradient algorithm. Two feature selection (FS) methods, multiple logistic regression (MLR) and sequential FS, were used to reduce the required attributes. The performance of the NFC (without/with FS) was then assessed in a hold-out validation framework. Further cross-validation was performed on the best classifier. Results: In this dataset, 16 complete attributes along with the binary CHD diagnosis (gold standard) for 272 subjects (68% male) were analyzed. MLR + NFC showed the best performance. Its overall sensitivity, specificity, accuracy, type I error (α) and statistical power were 79%, 89%, 84%, 0.1 and 79%, respectively. The selected features were “age and ST/heart rate slope categories,” “exercise-induced angina status,” fluoroscopy, and thallium-201 stress scintigraphy results. Conclusion: The proposed method showed “substantial agreement” with the gold standard. This algorithm is thus, a promising tool for screening CAD patients. © 2015, Isfahan University of Medical Sciences(IUMS). All rights reserved.
