An Open-source Image Analysis Toolbox for Quantitative Retinal Optical Coherence Tomography Angiography.

Background: Qualitative and quantitative assessment of retinal perfusion using optical coherence tomography angiography (OCTA) has shown to be effective in the treatment and management of various retinal and optic nerve diseases. However, manual analyses of OCTA images to calculate metrics related to Foveal Avascular Zone (FAZ) morphology, and retinal vascular density and morphology are costly, time-consuming, subject to human error, and are exposed to both inter and intra operator variability. Objective: This study aimed to develop an open-source software framework for quantitative OCTA (QOCTA). Particularly, for analyzing OCTA images and measuring several indices describing microvascular morphology, vessel morphology, and FAZ morphology. Material and Methods: In this analytical study, we developed a toolbox or QOCTA using image processing algorithms provided in MATLAB. The software automatically determines FAZ and measures several parameters related to both size and shape of FAZ including area, perimeter, Feret's diameter circularity, axial ratio, roundness, and solidity. The microvascular structure is derived from the processed image to estimate the vessel density (VD). To assess the reliability of the software, three independent operators measured the mentioned parameters for the eyes of 21 subjects. The consistency of the values was assessed using the intraclass correlation coefficient (ICC) index. Results: Excellent consistency was observed between the measurements completed for the superficial layer, ICC >0.9. For the deep layer, good reliability in the measurements was achieved, ICC >0.7. Conclusion: The developed software is reliable; hence, it can facilitate quantitative OCTA, further statistical comparison in cohort OCTA studies, and can assist with obtaining deeper insights into retinal variations in various populations.

A Hardware-Software System for Accurate Segmentation of Phonocardiogram Signal.

Background: Phonocardiogram (PCG) signal provides valuable information for diagnosing heart diseases. However, its applications in quantitative analyses of heart function are limited because the interpretation of this signal is difficult. A key step in quantitative PCG is the identification of the first and second sounds (S1 and S2) in this signal. Objective: This study aims to develop a hardware-software system for synchronized acquisition of two signals electrocardiogram (ECG) and PCG and to segment the recorded PCG signal via the information provided in the acquired ECG signal. Material and Methods: In this analytical study, we developed a hardware-software system for real-time identification of the first and second heart sounds in the PCG signal. A portable device to capture synchronized ECG and PCG signals was developed. Wavelet de-noising technique was used to remove noise from the signal. Finally, by fusing the information provided by the ECG signal (R-peaks and T-end) into a hidden Markov model (HMM), the first and second heart sounds were identified in the PCG signal. Results: ECG and PCG signals from 15 healthy adults were acquired and analyzed using the developed system. The average accuracy of the system in correctly detecting the heart sounds was 95.6% for S1 and 93.4% for S2. Conclusion: The presented system is cost-effective, user-friendly, and accurate in identifying S1 and S2 in PCG signals. Therefore, it might be effective in quantitative PCG and diagnosing heart diseases.

Recommended maximum holding time of common static sitting postures of office workers.

Objectives. A posture maintained for a long period can be harmful to the health of office workers. This study aimed to estimate the recommended ergonomic duration for maintaining different sitting postures. Methods. Forty healthy male and female students participated in this experiment designed to measure perceived discomfort caused by maintaining common static sitting postures of office workers in a simple ergonomic set-up for 4 min. The Borg CR10 scale was given to the participants to assess the discomfort in different body parts, before and after each experiment. Based on the mean group discomfort level of 2, the recommended holding time of each posture was estimated. Results. The recommended holding time and its discomfort score for each studied posture were tabulated. The shortest holding time of a posture was obtained for the moderate neck flexion (1.61 min), and the longest holding time was obtained for a leg posture with 90° knee flexion (6.45 min). Conclusions. The recommended holding time in this study may help to assess the risk of musculoskeletal disorders (MSDs) in office workers and train the individuals involved in office tasks in proper sitting behavior.

Machine Learning Models for Predicting Breast Cancer Risk in Women Exposed to Blue Light from Digital Screens.

Background: Nowadays, there is a growing global concern over rapidly increasing screen time (smartphones, tablets, and computers). An accumulating body of evidence indicates that prolonged exposure to short-wavelength visible light (blue component) emitted from digital screens may cause cancer. The application of machine learning (ML) methods has significantly improved the accuracy of predictions in fields such as cancer susceptibility, recurrence, and survival. Objective: To develop an ML model for predicting the risk of breast cancer in women via several parameters related to exposure to ionizing and non-ionizing radiation. Material and Methods: In this analytical study, three ML models Random Forest (RF), Support Vector Machine (SVM), and Multi-Layer Perceptron Neural Network (MLPNN) were used to analyze data collected from 603 cases, including 309 breast cancer cases and 294 gender and age-matched controls. Standard face-to-face interviews were performed using a standard questionnaire for data collection. Results: The examined models RF, SVM, and MLPNN performed well for correctly classifying cases with breast cancer and the healthy ones (mean sensitivity> 97.2%, mean specificity >96.4%, and average accuracy >97.1%). Conclusion: Machine learning models can be used to effectively predict the risk of breast cancer via the history of exposure to ionizing and non-ionizing radiation (including blue light and screen time issues) parameters. The performance of the developed methods is encouraging; nevertheless, further investigation is required to confirm that machine learning techniques can diagnose breast cancer with relatively high accuracies automatically.

A Mixed-Methods Investigation of Occupational Health Specialists' Knowledge and Application of Pen-and-Paper Observational Methods for Ergonomics Assessment.

OCCUPATIONAL APPLICATIONSErgonomic risk assessment is a key step in managing work-related musculoskeletal disorders. Diverse assessment methods exist, and errors may occur if inappropriate methods are selected. Understanding the level of knowledge, how to use methods, and exploring factors affecting erroneous usage of these methods, can provide useful information for health and safety regulatory authorities and decision-makers to identify problems and determine an action plan to eliminate them. We found that Iranian occupational health specialists have little knowledge about the types of pen-and-paper observational methods (OMs), and most of them use a limited number of these methods. Content analysis of interviews identified three main categories of influential factors and 12 subcategories. The main categories were educational, individual, and organizational factors. These results suggest the need for more effort to ensure that practitioners possess better knowledge and skills in the selection and application of pen-and-paper OMs.

Background Ergonomic risk assessment is a key step in managing ergonomics in the workplace. Errors in selecting risk assessment methods may occur during risk assessment and could invalidate the reliability of the results obtained.Purpose The aims of the present study were to investigate Iranian occupational health specialists' knowledge and application of pen-and-paper observational methods (OMs) for ergonomics assessment and to determine their influential factors.Methods We used a mixed-method design in two phases, combining quantitative and qualitative approaches. Quantitative data were collected from 237 occupational health specialists who were included in the study through purposive sampling of experts. In the qualitative phase, in-depth semi-structured interviews were conducted with participants from the first phase, and the factors associated with the selection and application of the pen-and-paper OMs were examined over a 6-month period.Results Participants were familiar with 13 out of 34 methods studied, and they had performed risk assessments with only nine of them. The most frequently used methods were RULA (95.4%), REBA (84.8%), QEC (65.4%), and ROSA (54.9%). Content analysis of interviews identified three main categories of influential factors and 12 subcategories. The main categories included educational, individual, and organizational factors.Conclusion Iranian occupational health specialists were found to have relatively limited knowledge about pen-and-paper OMs, and most of them use a limited number of these methods. These results suggest the need for more effort to ensure that practitioners possess better knowledge and skills in the selection and application of these methods. Creating stricter regulations regarding the qualifications of practitioners, modifying academic curriculum content, continuously monitoring the performance of practitioners, and holding retraining workshops in a purposeful manner could help minimize errors in selecting and applying pen-and-paper OMs.

Developing a Decision Aid Tool for selecting pen-paper observational ergonomics techniques: a quasi-experimental study.

BACKGROUND: A significant error that may occur during ergonomic risk assessment and invalidate assessment reliability corresponds to technique selection. This study aimed to develop a new tool called the Decision Aid Tool (DAT) to reduce pen-paper observational technique selection errors. METHODS: This quasi-experiment before-after study was performed in three phases. In the first phase, the participants' skills in technique selection were examined by showing them 20 videos of different single-task jobs. In the second phase, the DAT was designed using pen-paper observational techniques. Finally, in the third phase, 115 occupational health specialists included in the study through purposive sampling of experts evaluated the tool's efficacy. RESULTS: The results of the first phase showed that 62% of participants made an error in selecting the proper technique. The mean and standard deviation scores from the first and third phases were 11.4 ± 6.59 and 39.01 ±1.89, respectively. The mean scores increased significantly after using DAT, and 97.5% of participants could correctly select task techniques. CONCLUSIONS: The efficacy of DAT was confirmed in a quasi-experimental before-and-after study. Using DAT increases the participants' ability to choose the correct technique. The DAT can be functional for practitioners to select the pen-paper observational techniques correctly under the purpose of assessment, the body areas, and the characteristics of the task to be assessed.

A Comparison between the Perturbed-Chain Statistical Associating Fluid Theory Equation of State and Machine Learning Modeling Approaches in Asphaltene Onset Pressure and Bubble Point Pressure Prediction during Gas Injection.

Predicting asphaltene onset pressure (AOP) and bubble point pressure (Pb) is essential for optimization of gas injection for enhanced oil recovery. Pressure-Volume-Temperature or PVT studies along with equations of state (EoSs) are widely used to predict AOP and Pb. However, PVT experiments are costly and time-consuming. The perturbed-chain statistical associating fluid theory or PC-SAFT is a sophisticated EoS used for prediction of the AOP and Pb. However, this method is computationally complex and has high data requirements. Hence, developing precise and reliable smart models for prediction of the AOP and Pb is inevitable. In this paper, we used machine learning (ML) methods to develop predictive tools for the estimation of the AOP and Pb using experimental data (AOP data set: 170 samples; Pb data set: 146 samples). Extra trees (ET), support vector machine (SVM), decision tree, and k-nearest neighbors ML methods were used. Reservoir temperature, reservoir pressure, SARA fraction, API gravity, gas-oil ratio, fluid molecular weight, monophasic composition, and composition of gas injection are considered as input data. The ET (R 2: 0.793, RMSE: 7.5) and the SVM models (R 2: 0.988, RMSE: 0.76) attained more reliable results for estimation of the AOP and Pb, respectively. Generally, the accuracy of the PC-SAFT model is higher than that of the AI/ML models. However, our results confirm that the AI/ML approach is an acceptable alternative for the PC-SAFT model when we face lack of data and/or complex mathematical equations. The developed smart models are accurate and fast and produce reliable results with lower data requirements.

Optimizing artificial meniscus by mechanical stimulation of the chondrocyte-laden acellular meniscus using ad hoc bioreactor.

BACKGROUND: Tissue engineering focuses on reconstructing the damaged meniscus by mimicking the native meniscus. The application of mechanical loading on chondrocyte-laden decellularized whole meniscus is providing the natural microenvironment. The goal of this study was to evaluate the effects of dynamic compression and shear load on chondrocyte-laden decellularized meniscus. MATERIAL AND METHODS: The fresh samples of rabbit menisci were decellularized, and the DNA removal was confirmed by histological assessments and DNA quantification. The biocompatibility, degradation and hydration rate of decellularized menisci were evaluated. The decellularized meniscus was injected at a density of 1 × 105 chondrocyte per scaffold and was subjected to 3 cycles of dynamic compression and shear stimuli (1 h of 5% strain, ± 25°shear at 1 Hz followed by 1 h rest) every other day for 2 weeks using an ad hoc bioreactor. Cytotoxicity, GAG content, ultrastructure, gene expression and mechanical properties were examined in dynamic and static condition and compared to decellularized and intact menisci. RESULTS: Mechanical stimulation supported cell viability and increased glycosaminoglycan (GAG) accumulation. The expression of collagen-I (COL-I, 10.7-folds), COL-II (6.4-folds), aggrecan (AGG, 3.2-folds), and matrix metalloproteinase (MMP3, 2.3-folds) was upregulated compared to the static conditions. Furthermore, more aligned fibers and enhanced tensile strength were observed in the meniscus treated in dynamic condition with no sign of mineralization. CONCLUSION: Compress and shear stimulation mimics the loads on the joint during walking and be able to improve cell function and ultrastructure of engineered tissue to recreate a functional artificial meniscus.

Engineered artificial articular cartilage made of decellularized extracellular matrix by mechanical and IGF-1 stimulation.

Cartilage engineering has the potential to overcome clinical deficiency in joint disorders. Decellularized extracellular matrix (dECM) has great biocompatibility and bioactivity and can be considered an appropriate natural scaffold for tissue engineering applications. Both insulin-like growth factor-1 (IGF-1) and mechanical compression stimulate the production of cartilage ECM, modulate mechanical properties, and gene expression. The current investigation aimed to fabricate a high-quality moldable artificial cartilage by exposing the chondrocytes in biomimicry conditions using cartilage dECM, IGF-1, and mechanical stimulations. In this study, an ad hoc bioreactor was designed to apply dynamic mechanical stimuli (10 % strain, 1 Hz) on chondrocyte-laden cartilage dECM-constructs with/without IGF-1 supplementation for 2 weeks, 3 h/day. Our data revealed that mechanical stimulation had no adverse effect on cell viability and proliferation. However, it elevated the expression of chondrogenic markers such as collagen type II (COL2A1), aggrecan (ACAN), and proteoglycan-4 (PRG-4), and reduced the expression of matrix metalloproteinase-3 (MMP-3). Mechanical stimulation also promoted higher newly formed glycosaminoglycan (GAG) and produced more aligned fibers that can be responsible for higher Young's modulus of the engineered construct. Even though IGF-1 demonstrated some extent of improvement in developing neocartilage, it was not as effective as mechanical stimulation. Neither IGF-1 nor compression elevated the collagen type I expression. Compression and IGF-1 showed a synergistic impact on boosting the level of COL2A1 but not the other factors. In conclusion, mechanical stimulation on moldable cartilage dECM can be considered a good technique to fabricate artificial cartilage with higher functionality.

Assessment of macular findings by OCT angiography in patients without clinical signs of diabetic retinopathy: radiomics features for early screening of diabetic retinopathy.

This cross-sectional study aimed to quantitatively analyze the optical coherence tomography angiography (OCTA) images using MATLAB-based software and evaluate the initial changes in macular vascular density and the distortion of the foveal avascular zone (FAZ), before the clinical appearance of diabetic retinopathy. For this purpose, 21 diabetic patients without any clinical features indicating DR, and 21 healthy individuals matched with patients based on their demographic characteristics were included. Macular thickness, macular vascular density, and morphological changes of FAZ were assessed using OCTA. The diagnostic ability of morphological parameters was evaluated by receiver operating curve analysis. The intraclass correlation coefficient (ICCC) index was used to check the consistency of the extracted values. There was no significant difference in age, gender, LogMAR visual acuity, spherical equivalent, and intra-ocular pressure amongst patients and controls. No correlation was found between age and the FAZ area as well as vascular density. The vascular structure of the superficial layer showed FAZ enlargement, reduced vascular density in the macular area, and significant deviations of FAZ shape parameters (convexity and Frequency Domain Irregularity) in patients compared with healthy individuals. Measurements were highly correlated between separate imaging sessions with ICCC of over 0.85 for all parameters. The represented data suggests that radiomics parameters can be applied as both an early screening tool and guidance for better follow-up of diabetic patients who have not had any sign of DR in fundoscopic exams.

A machine learning-based system for detecting leishmaniasis in microscopic images.

BACKGROUND: Leishmaniasis, a disease caused by a protozoan, causes numerous deaths in humans each year. After malaria, leishmaniasis is known to be the deadliest parasitic disease globally. Direct visual detection of leishmania parasite through microscopy is the frequent method for diagnosis of this disease. However, this method is time-consuming and subject to errors. This study was aimed to develop an artificial intelligence-based algorithm for automatic diagnosis of leishmaniasis. METHODS: We used the Viola-Jones algorithm to develop a leishmania parasite detection system. The algorithm includes three procedures: feature extraction, integral image creation, and classification. Haar-like features are used as features. An integral image was used to represent an abstract of the image that significantly speeds up the algorithm. The adaBoost technique was used to select the discriminate features and to train the classifier. RESULTS: A 65% recall and 50% precision was concluded in the detection of macrophages infected with the leishmania parasite. Also, these numbers were 52% and 71%, respectively, related to amastigotes outside of macrophages. CONCLUSION: The developed system is accurate, fast, easy to use, and cost-effective. Therefore, artificial intelligence might be used as an alternative for the current leishmanial diagnosis methods.

Common errors in selecting and implementing pen-paper observational methods by Iranian practitioners for assessing work-related musculoskeletal disorders risk: a systematic review.

Objectives. This study aimed to determine the types and frequency of pen-paper observational methods (OMs) used by Iranian practitioners and to identify their errors in selecting and implementing these methods. Methods. This was a systematic review and analytical study of papers in which the OMs had been used. Scientific databases were analyzed from September 1970 to September 2018. Errors were determined based on a list of wrong practices both in the selection and implementation of methods. Three ergonomists carried out the process of identifying errors independently. Results. The most frequently used methods were rapid upper limb assessment (RULA), quick exposure check (QEC) and rapid entire body assessment (REBA), respectively. Errors in selecting and implementing pen-paper OMs were 53.3 and 36.4%, respectively. Conclusions. Despite the abundant number of pen-paper OMs, Iranian practitioners use few of them. The high rate of errors can indicate a lack of knowledge and skills among practitioners for selecting and implementing OMs. The development of decision-making tools may help practitioners to select appropriate pen-paper OMs for assessing different types of tasks.

Investigation of office workers' sitting behaviors in an ergonomically adjusted workstation.

Objectives. Common ergonomic office workstations are designed for a few optimum postures. Nonetheless, sitting is a dynamic activity and the ideal sitting posture is rarely maintained in practice. Therefore, the present study aimed to investigate the sitting behavior of office workers in an actual working environment using ergonomically adjusted workstations to examine whether they promote maintaining appropriate sitting postures. Methods. Sitting behaviors (frequency of postures and position changes in different body parts) were explored among 26 office workers during a 60-min sitting duration, using the posture recording and classification method developed by Graf et al. The rapid upper limb assessment (RULA) method was also used to assess postural load. Then, the results of the RULA method were compared with the results from investigating the sitting behavior of office workers. Results. Common ergonomic workstations were effective in eliminating some awkward postures. However, some important risk factors such as holding postures with an inappropriate lumbar spine curve (86% of the observations) and maintaining a posture for a long time (for 7-12 min) were observed in the participants' sitting behaviors, while they were neglected in the RULA method. Conclusions. The common ergonomic workstations could not guarantee the users' appropriate sitting behaviors.

A machine learning model for predicting favorable outcome in severe traumatic brain injury patients after 6 months.

BACKGROUND: Traumatic brain injury (TBI), which occurs commonly worldwide, is among the more costly of health and socioeconomic problems. Accurate prediction of favorable outcomes in severe TBI patients could assist with optimizing treatment procedures, predicting clinical outcomes, and result in substantial economic savings. METHODS: In this study, we examined the capability of a machine learning-based model in predicting "favorable" or "unfavorable" outcomes after 6 months in severe TBI patients using only parameters measured on admission. Three models were developed using logistic regression, random forest, and support vector machines trained on parameters recorded from 2,381 severe TBI patients admitted to the neuro-intensive care unit of Rajaee (Emtiaz) Hospital (Shiraz, Iran) between 2015 and 2017. Model performance was evaluated using three indices: sensitivity, specificity, and accuracy. A ten-fold cross-validation method was used to estimate these indices. RESULTS: Overall, the developed models showed excellent performance with the area under the curve around 0.81, sensitivity and specificity of around 0.78. The top-three factors important in predicting 6-month post-trauma survival status in TBI patients are "Glasgow coma scale motor response," "pupillary reactivity," and "age." CONCLUSIONS: Machine learning techniques might be used to predict the 6-month outcome in TBI patients using only the parameters measured on admission when the machine learning is trained using a large data set.

AN ARTIFICIAL NEURAL NETWORK-BASED MODEL FOR PREDICTING ANNUAL DOSE IN HEALTHCARE WORKERS OCCUPATIONALLY EXPOSED TO DIFFERENT LEVELS OF IONIZING RADIATION.

We presented an artificial intelligence-based model to predict annual effective dose (AED) value of health workers. Potential factors affecting AED and the results of annual blood tests were collected from 91 radiation workers. Filter-based feature selection strategy revealed that the eight factors plate, red cell distribution width (RDW), educational degree, nonacademic course in radiation protection (hour), working hours per month, department and the number of procedures done per year and work in radiology department or not (0,1) were the most important predictors for AED. The prediction model was developed using a multilayer perceptron neural network and these prediction parameters as inputs. The model provided favorable accuracy in predicting AED value while a regression model did not. There was a strong linear relationship between the predicted AED values and the measured doses (R-value =0.89 for training samples and 0.86 for testing samples). These results are promising and show that artificial neural networks can be used to improve/facilitate dose estimation process.

Predicting the Postoperative Outcome of Percutaneous Nephrolithotomy with Machine Learning System: Software Validation and Comparative Analysis with Guy's Stone Score and the CROES Nomogram.

Purpose: To validate the output of a machine learning-based software as an intelligible interface for predicting multiple outcomes after percutaneous nephrolithotomy (PCNL). We compared the performance of this system with Guy's stone score (GSS) and the Clinical Research Office of Endourological Society (CROES) nomogram. Patients and Methods: Data from 146 adult patients (87 males, 59%) who underwent PCNL at our institute were used. To validate the system, accuracy of the software for predicting each postoperative outcome was compared with the actual outcome. Similarly, preoperative data were analyzed with GSS and CROES nomograms to determine stone-free status as predicted by these nomograms. A receiver operating characteristic (ROC) curve was generated for each scoring system, and the area under the ROC curve (AUC) was calculated and used to assess the predictive performance of all three models. Results: Overall stone-free rate was 72.6% (106/146). Forty of 146 patients (27.4%) were scheduled for 42 ancillary procedures (extracorporeal shockwave lithotripsy [SWL] [n = 31] or repeat PCNL [n = 11]) to manage residual renal stones. Overall, the machine learning system predicted the PCNL outcomes with an accuracy ranging between 80% and 95.1%. For predicting the stone-free status, the AUC for the software (0.915) was significantly larger than the AUC for GSS (0.615) or CROES nomograms (0.621) (p < 0.001). Conclusion: At the internal institutional level, the machine learning-based software was a promising tool for recording, processing, and predicting outcomes after PCNL. Validation of this system against an external dataset is highly recommended before its widespread application.

An artificial intelligence-based clinical decision support system for large kidney stone treatment.

A decision support system (DSS) was developed to predict postoperative outcome of a kidney stone treatment procedure, particularly percutaneous nephrolithotomy (PCNL). The system can serve as a promising tool to provide counseling before an operation. The overall procedure includes data collection and prediction model development. Pre/postoperative variables of 254 patients were collected. For feature vector, we used 26 variables from three categories including patient history variables, kidney stone parameters, and laboratory data. The prediction model was developed using machine learning techniques, which includes dimensionality reduction and supervised classification. A novel method based on the combination of sequential forward selection and Fisher's discriminant analysis was developed to reduce the dimensionality of the feature space and to improve the performance of the system. Multiple classifier scheme was used for prediction. The derived DSS was evaluated by running leave-one-patient-out cross-validation approach on the dataset. The system provided favorable accuracy (94.8%) in predicting the outcome of a treatment procedure. The system also correctly estimated 85.2% of the cases that required stent placement after the removal of a stone. In predicting whether the patient might require a blood transfusion during the surgery or not, the system predicted 95.0% of the cases correctly. The results are promising and show that the developed DSS could be used in assisting urologists to provide counseling, predict a surgical outcome, and ultimately choose an appropriate surgical treatment for removing kidney stones.

A knowledge-based system for brain tumor segmentation using only 3D FLAIR images.

This study aims to develop a semi-automatic system for brain tumor segmentation in 3D MR images. For a given image, noise was corrected using SUSAN algorithm first. A specific region of interest (ROI) that contains tumor was identified and then the intensity non-uniformity in ROI was corrected via the histogram normalization and intensity scaling. Each voxel in ROI was presented using 22 features and then was categorized as tumor or non-tumor by a multiple-classifier system. T1- and T2-weighted images and fluid-attenuated inversion recovery (FLAIR) were examined. The system performance in terms of Dice index (DI), sensitivity (SE) and specificity (SP) was evaluated using 150 simulated and 30 real images from the BraTS 2012 database. The results showed that the presented system with an average DI > 0.85, SE > 0.90, and SP > 0.98 for simulated data and DI > 0.80, SE > 0.84, and SP > 0.98 for real data might be used for accurate extraction of the brain tumors. Moreover, this system is 6 times faster than a similar system that processes the whole image. In comparison with two state-of-the-art tumor segmentation methods, our system improved DI (e.g., by 0.31 for low-grade tumors) and outperformed these algorithms. Considering the costs of imaging procedures, tumor identification accuracy and computation times, the proposed system that augmented general pathological information about tumors and used only 4 features of FLAIR images can be suggested as a brain tumor segmentation system for clinical applications.

Automatic segmentation of vertebrae in 3D CT images using adaptive fast 3D pulse coupled neural networks.

Two systems are presented for segmentation of vertebrae in a 3D computed tomography (CT) image. The first method extracts seven features from each voxel and uses a multi-layer perceptron neural network (MLPNN) to classify the voxel as vertebrae or background. In the second method, the segmentation is completed in two steps: first, a newly developed adaptive pulse coupled neural network (APCNN) directly applied to a given image segments vertebrae, then the result is refined using a median filter. In the developed APCNN, the values for the user-defined parameters of the pulse coupled neural networks (PCNN) are adaptively adjusted for each image individually, instead of using one value for all images as in conventional PCNN. The performance of both systems in terms of Dice index (DI) was evaluated and compared against the state-of-the-art segmentation methods using seventeen clinical and standard CT images. Overall, both systems demonstrated statistically similar and promising performance with average DI > 95%. Compared to existing PCNN-based segmentation algorithms, the accuracy of the proposed APCNN improved by 29.3% on average. The developed APCNN-based system is more accurate than MLPNN-based system and existing PCNN-based algorithms in segmentation of vertebrae with blurred and weak boundaries and in the images contaminated by salt- and- pepper noise. In terms of computation time, the APCNN-based system is 16 times faster than the MLPNN-based system. Consequently, the presented APCNN-based algorithm is both accurate and fast and could be used in clinical environment for segmentation of vertebrae in 3D CT images.

Cross Comparison of Motor Unit Potential Features Used in EMG Signal Decomposition.

Feature extraction is an important step of resolving an electromyographic (EMG) signal into its component motor unit potential trains, commonly known as EMG decomposition. Until now, different features have been used to represent motor unit potentials (MUPs) and improve decomposition processing time and accuracy, but a major limitation is that no systematic comparison of these features exists. In an EMG decomposition system, like any pattern recognition system, the features used for representing MUPs play an important role in the overall performance of the system. A cross comparison of the feature extraction methods used in EMG signal decomposition can assist in choosing the best features for representing MUPs and ultimately may improve EMG decomposition results. This paper presents a survey and cross comparison of these feature extraction methods. Decomposability index, classification accuracy of a -nearest neighbors classifier, and class-feature mutual information were employed for evaluating the discriminative power of various feature extraction techniques commonly used in the literature including time domain, morphological, frequency domain, and discrete wavelets. In terms of data, 45 simulated and 82 real EMG signals were used. Results showed that among time domain features, the first derivative of time samples exhibit the best separability. For morphological features, slope analysis provided the most discriminative power. Discrete Fourier transform coefficients offered the best separability among frequency domain features. However, neither morphological nor frequency domain techniques outperformed time domain features. The detail 4 coefficients in a discrete wavelets decomposition exceeded in evaluation measures when compared with other feature extraction techniques. Using principal component analysis slightly improved the results, but it is time consuming. Overall, considering computation time and discriminative ability, the first derivative of time samples might be efficient in representing MUPs in EMG decomposition and there is no need for sophisticated feature extraction methods.