ISSN: 2709-4111
Submit Article
Submission Instructions
Open Access Information
Ethics and Malpractice Statement
Most Recent Issue
2021
Issue 2Issue 1

Visit the new journal website to submit and consult our contents: https://publications.eai.eu/index.php/bebi/index

Editor(s)-in-Chief: Nikolaos Bourbakis, Jeffrey P. Tsai and Nenad Filipovic

EAI Endorsed Transactions on Bioengineering and Bioinformatics (BEBI), formerly EAI Endorsed Transactions on Ambient Systems, aims at building intelligent synergies and systems among Bioinformatics, Computational Biology, Bioengineering and Biomedicine these complementary disciplines that hold great promise for the advancement of research and development in complex medical and biological systems, agriculture, environment, public health, drug design. Research and development in these areas are impacting the science and technology in fields such as medicine, food production, forensics, etc. by advancing fundamental concepts in molecular biology, by helping us understand living organisms at multiple levels, by developing innovative implants and bio-prosthetics, and by improving tools and techniques for the detection, prevention and treatment of diseases. The BEBI will provide a common platform for the cross fertilization of ideas, and for shaping knowledge and scientific achievements by bridging these very important and complementary disciplines into an interactive and attractive forum.

  • Editors-in-Chief
  • Nikolaos Bourbakis, Wright State University in Ohio, United States
  • Jeffrey P. Tsai, Asia University, Taiwan
  • Nenad Filipovic, University of Kragujevac, Serbia
  • Editorial board
  • Benny Lo (Imperial College London, UK)
  • Henrique C Fernandes (Federal University of Uberlandia, Brazil)
  • Koji Kashihara (Ritsumeikan University, Japan)
  • Kostas Michalopoulos (Convergent Engineering Inc., USA)
  • Luca Citi (University of Essex, UK)
  • George Lambrou (Kapodistriakon University of Athens, Greece)
  • Constantinos Pattichis (University of Cyprus, Cyprus)
  • Nikolaos Tsekos (University of Houston, USA)
  • Athanasios Tsanas (University of Edinburgh, UK)
  • Alexandros Karargyris (IBM, USA)
  • Iosif Papadakis-Ktistakis (ASML, USA)
  • Ioannis Pavlidis (University of Houston, USA)
  • Michael Raymer (Wright State University, USA)
  • Katia Del Rio-Tsonis (Miami University, USA)
  • Pothitos Pytichoutis (University of Dayton, USA)
  • Sameer Antani (National Institutes of Health, USA)
  • Michalis Zervakis (Technical University of Crete, Greece)
  • John R. Rose (University of South Carolina, USA)
  • George M. Spyrou (The Cyprus School of Molecular Medicine, Cyprus)
  • Roman Snytsar (Research and AI Microsoft, USA)
  • Fani Deligianni (Glasgow University, Scotland)
  • Kristen Beck (IBM Research, USA)
  • Ehsan Dehghan (IBM Research, Canada)
  • Zoran Bosnic (University of Ljubljana, Slovenia)
  • Marko Robnik-Sikonja (University of Ljubljana, Slovenia)
  • Ka-Lok Ng (Asia University, Taiwan)
  • Tijana Sustersic (Faculty of Engineering, University of Kragujevac)
  • Zoran Bosnić (University of Ljubljana, Slovenia)
  • Marko Robnik-Šikonja (University of Ljubljana, Slovenia)

Nikolaos Bourbakis1

Dr. Nikolaos Bourbakis (IEEE Life Fellow) currently is a Distinguished Professor in Information Technology, CSE Dept., Professor (j.a.) of Geriatrics School of Medicine and the Director of the Center of Assistive Research Technologies (CART) at Wright State University (WSU), Dayton, OH. He has conducted research in Applied AI (Projects: SE-Expert System, Hermes-Vision Architecture, Kydon- Image Recognition Architecture, BouMer-Learing, DIAS Architecture, Handwritten OCR-Words, Glossa-NL Understanding, Technical Document Understanding, SKD Body Activity Recognition), Cyber Security (Projects: SCAN, AMYNA) and Bioengineering (Projects: Tyflos-Wearable Decvice for Assisting Blind for Navigation, Reading and Recognition; SPN-Brain Analysis for Recognition; Wireless Capsule Endoscopy; Ultrasound Imaging-Liver-Heart; Nosokoma & Prognosis for Assistive Living) funded ($15M) by government agencies and industry. He has published 450 articles in refereed IEEE/ International Journals, book-chapters and Conference Proceedings, 10 books and 60 technical reports. He has advised and graduated 30 PhDs and 40 Master students. He is/was the founder and the EIC of the International Journal on AI Tools (WS Pub. 1992 -) and the International Journal on Monitoring & Surveillance Tech Research (IGI-Global Pub. 2013 -2019); the Editor in Charge of a Research Series of Books in AI (WS Publisher, 1993-2001); the Founder and General Chair in International IEEE Computer Society Conferences (ICTAI, BIBE, IISA), Symposia and Workshops; the Associate Editor (12) in numerous IEEE and International Journals; and a Guest Editor in 25 special issues in IEEE and International Journals. He is/was a Distinguished IEEE Computer Society Speaker; an NSF University Research Programs Evaluator; an IEEE Computer Society Golden Core Member; the Founder-President of the Biological & Artificial Intelligence Society-Foundation (BAIF); an External Evaluator in University Promotion Committees, an Official Nominator of the National Academy of Achievements for Computer Science Programs; and a Keynote Speaker in numerous International/IEEE Conferences. His research work has been internationally recognized and has won several prestigious awards. Some of them are: IBM Author recognition Award 1991; IEEE Computer Society Outstanding Contribution Award 1992; Member of the New York Academy of Sciences (1993); IEEE Outstanding Paper Award AUTESTCON 1994; IEEE Computer Society Technical Research Achievement Award 1998; IEEE I&S Outstanding Leadership Award 1998; IEEE ICTAI 10 years Research Contribution Award 1999; Pattern Recognition Society Journal best papers selection for 1999; IEEE Symposium on BIBE Outstanding Leadership Award, 2003; ASC Outstanding Scientists & Engineers Research Award-2005; SETN Research Recognition- 2006; Degree of Honor and Recognition, University of Patras, Greece 2007; IEEE Appreciation Award for organizing ICTAI- 08, Dayton OH, 2008; IEEE Computer Society Appreciation Award as the BIBE Steering Chair 2009; Dr. F. Russ IEEE Biomedical Engineering award, Dayton OH, 2010; Pattern Recognition Society Journal “Most Cited Article for 2006-2010”, and “Honorary mentioned for several years”; IJAIT best papers selection for the years 2006-2010; IEEE BIBE- Keynote Speaker Recognition Award Mobilehealth, Taichung, Taiwan, 2011; WSU Outstanding Service Award, 2001-2012; IEEE Int. Conference ICTAI-2012 CV Ramammoorthy Best paper Award, Piraeus, Greece; “In honor and recognition for his outstanding scholarly achievements and contributions in the field of Computer Science”, Dept. of Informatics, University of Piraeus, Greece, July 2013; IEEE EMBS-GR Award of Achievements for his contribution on Assistive Devices for the Blind, IEEE IC-BIBE, Nov 2013, Chania , Greece; IEEE Conference BIBE-2014, Biomedical Engineering Best paper Award FL, USA, Nov. 2014; UNESCOGR Recognition Leadership Award, IEEE Conference IISA, Corfu, Greece, July 2015; IEEE Conference on Tools with AI, CV Ramamoorthy Best paper Award, Vietri sul Mare, Italy Nov. 2015; IEEE Computer Society 28 years ICTAI Outstanding Service & Leadership Recognition, IEEE ICTAI, S. Jose, CA, 2016; IEEE Computer Society Keynote Speaker Recognition Award 2017, Boston MA; IEEE ICTAI-2018, 30 years Leadership Recognition, Volos, Nov 2018, Greece; IEEE ICTAI-2019 the Best Paper Award (Nov. 2019, Portland OR, USA) was given on his name for his 31 years outstanding contribution to the ICTAI series; Nominated for Honorary Doctoral Degree from the University of Piraeus, Greece 2020.

Jeffrey P. Tsai2

Asia University, Taiwan

Nenad Filipovic3

University of Kragujevac, Serbia


1: Wright State University in Ohio, United States
2: Asia University, Taiwan
3: University of Kragujevac, Serbia

  • Estimating animal pose using deep learning: a trained deep learning model outperforms morphological analysis

    Appears in:
    bebi 22(4): e5

    Authors:
    J. Banzon, S. Lee, D. H. Kim, K. Le

    Published:
    22nd Apr 2022

    Abstract:
    INTRODUCTION: Analyzing animal behavior helps researchers understand their decision-making process and helper tools are rapidly becoming an indispensable part of many interdisciplinary studies. However, researchers are often challenged to estimate animal pose because of the limitation of the tools and its vulnerability to a specific environment. Over the years, deep learning has been introduced as an alternative solution to overcome these challenges. OBJECTIVES: This study investigates how deep learning models can be applied for the accurate prediction of animal behavior, comparing with traditional morphological analysis based on image pixels. METHODS: Transparent Omnidirectional Locomotion Compensator (TOLC), a tracking device, is used to record videos with a wide range of animal behavior. Recorded videos contain two insects: a walking red imported fire ant (Solenopsis invicta) and a walking fruit fly (Drosophila melanogaster). Body parts such as the head, legs, and thorax, are estimated by using an open-source deep-learning toolbox. A deep learning model, ResNet-50, is trained to predict the body parts of the fire ant and the fruit fly respectively. 500 image frames for each insect were annotated by humans and then compared with the predictions of the deep learning model as well as the points generated from the morphological analysis. RESULTS: The experimental results show that the average distance between the deep learning-predicted centroids and the human-annotated centroids is 2.54, while the average distance between the morphological analysis-generated centroids and the human-annotated centroids is 6.41 over the 500 frames of the fire ant. For the fruit fly, the average distance of the centroids between the deep learning- predicted and the human-annotated is 2.43, while the average distance of the centroids between the morphological analysis-generated and the human-annotated is 5.06 over the 477 image frames. CONCLUSION: In this paper, we demonstrate that the deep learning model outperforms traditional morphological analysis in terms of estimating animal pose in a series of video frames.

  • Determination of VEGF and CXCR4 in Tumor and Peritumoral Tissue of Patients with Breast Cancer as a Predictive Factor

    Appears in:
    bebi 22(4): e4

    Authors:
    Milena Milutinović, Slobodanka Mitrović, Nataša Zdravković, Danijela Nikodijević, Aleksandar Cvetković, Vladislava Stojić, Danijela Cvetković, Jovana Jovankić

    Published:
    25th Mar 2022

    Abstract:
    INTRODUCTION: Despite the obvious progress in the field of diagnosis and therapy, further measures are needed to increase the effectiveness of treatment and reduce morbidity and mortality from breast cancer. OBJECTIVES: To study the influence of peritumoral tissue on the growth and development of the tumor itself. METHODS: An immunofluorescence method was used to determine the protein expression of VEGF and CXCR-4 in tumor and peritumoral tissue. RESULTS: Peritumoral tissue is not only a passive factor, but actively participates in the process of tumor growth and development, as well as in the processes of recurrence and metastasis. CONCLUSION: Markers of neoangiogenesis in tumor and peritumoral tissue such as protein expression of VEGF and CXCR-4 receptors may serve as reliable predictors of disease outcome in breast cancer patients, which may provide useful suggestions in treatment choices.

  • A smart cropping pipeline to improve prostate’s peripheral zone segmentation on MRI using Deep Learning

    Appears in:
    bebi 22(4): e3

    Authors:
    Nikolaos Tachos, Dimitios I. Fotiadis, Dimitris Zaridis, Eugenia Mylona, Manolis Tsiknakis, Kostas Marias

    Published:
    24th Feb 2022

    Abstract:
    INTRODUCTION: Although accurate segmentation of the prostatic subregions is a crucial step for prostate cancer diagnosis, it remains a challenge. OBJECTIVES: To propose a deep learning (DL)-based cropping pipeline to improve the performance of DL networks for segmenting the prostate’s peripheral zone. METHODS: A U-net network was trained to crop the area around the peripheral zone on MRI in order to reduce the class imbalance between foreground and background pixels. The DL-cropping was compared with the standard center-cropping using three segmentation networks. RESULTS: The DL-cropping improved significantly the segmentation performance in terms of Dice score, Sensitivity, Hausdorff Distance, and Average Surface Distance, for all three networks. The improvement in Dice Score was 34%, 13% and 16% for the U-net, Dense U-net and Bridged U-net, respectively. CONCLUSION: For all the evaluated networks, the proposed DL-cropping technique outperformed the standard center-cropping.

  • Influences of seasonal and demographic factors on the COVID-19 pandemic dynamics

    Appears in:
    bebi 21(4): e2

    Authors:
    O. Rodionov, I. Nesteruk, S. Walczak, A. V. Nikitin

    Published:
    8th Dec 2021

    Abstract:
    INTRODUCTION: The number of COVID-19 cases per capita (accumulated (CC) and daily (DCC)) are important characteristics of the pandemic dynamics indicating the effectiveness of quarantine, testing, and vaccination. They also indicate the appearance of new waves (e.g., caused by new coronavirus strains) and may be the result of various demographic and seasonal factors. OBJECTIVES: We investigate the influence of the volume and the density of population and the urbanisation level on the CC values accumulated in European countries and regions of Ukraine at the end of June 2021 and the impact of seasonal factors on the DCC values by comparing their dynamics in the spring and summer of 2020 and 2021 for northern and southern regions. METHODS: The influence of demographic factors on CC values was investigated with the use of linear regression. Since DCC values are very random and demonstrate some weekly period, we have used the 7-days smoothing proposed before. The second year of the pandemic allows us to compare its dynamics in the spring and the summer of 2020 with the same period in 2021 and investigate the influence of seasonal factors. We have chosen some northern countries and regions: Ukraine, EU, the UK, USA and some countries located in the tropical zone and Southern Hemisphere: India, Brazil, South Africa and Argentina. The dynamics in these regions was compared with the global one. RESULTS: The accumulated number of cases per capita CC does not depend on the demographic factors used for analysis, although it may differ by about 4 times for different regions of Ukraine and more than 9 times for different European countries. The number of COVID-19 per capita registered in Ukraine is comparable with the same characteristic in other European countries but much higher than in China, South Korea and Japan. Some seasonal similarities are visible for global dynamics, EU and South Africa. Before July 2020, the southern countries demonstrated exponential growth, but northern regions showed some stabilization trends. CONCLUSION: The CC values in Europe do not show any visible dependence on the volume of population, its density and the urbanization level. More or less similar seasonal behaviour of DCC values are visible for global dynamics in July and August. Unfortunately, we cannot conclude where the quarantine restrictions were the most effective since the dynamics of the pandemic are influenced by many other factors not considered in this study, in particular, the emergence of new strains and the large number of unreported cases.

  • An Approach to Improve the Water Quality on Industrial Effluent by Phytoremediation with Water Hyacinth (Eichhornia Crassipes)

    Appears in:
    bebi 21(4): e1

    Authors:
    Bhishma Karki, S. Sankar, A Shaji George, Digvijay Pandey, Pankaj Dadeech, M R Sundarakumar, Binay Kumar Pandey

    Published:
    12nd Oct 2021

    Abstract:
    INTRODUCTION: Industries and factories are the backbone of any nation to increase the economy through its productivity and develop the county globally. But the wastages and sewages produced by each industry are creating pollution and make cleanliness environment on the society. OBJECTIVES: The recycling process and decontamination process of that industrial effluent is a challenging task for all the industries in and around the world. Ancient techniques are followed in industrial effluent cleaning processes but the output of those is not sufficient in terms of recycling. METHODS: Phytoremediation is the new technique used to control industrial effluent with the help of water hyacinth available in the water places. This paper discussed all possible solutions of industrial effluent removal along with water quality improvement techniques. Decontamination of the industrial effluent using water hyacinth is mainly dealt in leaves, stems, and root parts. There are only water hyacinth is used to remove industrial effluent, no other products like powder, liquid plants, and other chemicals are used. RESULTS: Total duration of 21 days has to be considered for the growth of water hyacinth used in water places to monitor the recycling of industrial effluent at different parts. A water quality test has been conducted to identify the level of contaminants in the Industrial effluent. Furthermore, BOD (Biological oxygen demand), COD (Chemical oxygen demand), TDS (Total dissolved solids), TSS (Total suspended solids), Turbidity, and chloride levels are taken as a parameter from the industrial effluents. CONCLUSION: Every 24hours, for continuous 3-5days the color of effluent has been notified to predict the changes. Turbidity and other impurities are noted by the effects of the water hyacinth used as results.

  • Semi-supervised Learning for COVID-19 Image Classification via ResNet

    Appears in:
    bebi 21(3): e5

    Authors:
    Lijun Qian, Seungchan Kim, Xishuang Dong, Xiangfang Li, Lucy Nwosu

    Published:
    25th Aug 2021

    Abstract:
    INTRODUCTION: Coronavirus disease 2019 (COVID-19) is an ongoing global pandemic in over 200 countries and territories. Analysis of X-ray imaging data can play a critical role in timely and accurate screening and fighting against COVID-19. OBJECTIVES: Supervised deep learning dominates COVID-19 pathology data analytics. However, it requires a substantial amount of annotated X-ray images to train models, which is often not applicable to data analysis for emerging events. METHODS: The proposed model with two paths is built based on Residual Neural Network for COVID-19 image classification to reduce labeling efforts, where the two paths refer to a supervised path and an unsupervised path, respectively. RESULTS: Experimental results demonstrate that the proposed model can achieve promising performance even when trained on very few labeled training image. CONCLUSION: The proposed model can reduces the efforts of building deep learning models significantly for COVID-19 image classification.

  • Non-linear dynamics of morphogenetic renal branching and Bifurcation analysis of chaotic patterns in autoregulation mechanisms

    Appears in:
    bebi 21(3): e4

    Authors:
    Amareswara Prasad Chunduru

    Published:
    14th Jul 2021

    Abstract:
    INTRODUCTION: Modelling of kidney physiology can add to comprehension of its work by formalizing existing information into numerical conditions and computational methods. The study evaluates the mathematical models that have been created to comprehend kidney physiology and pathophysiology. OBJECTIVE: Kidneys play a critical role in maintaining the body's water balance, electrolyte equilibrium, and acid-base balance, Through current knowledge with numerical models and computational methods, kidney physiology modeling will improve understanding of kidney function. METHOD: A L-System fractal system is designed to develop symmetrical branching tree systems that can fuse the physiological concepts of arterial tree branching to find the efficiency of blood flow in the renal arterial tree. Hopf Bifurcation analysis is also performed on mathematical models of autoregulation mechanisms that evaluate kidney physiology and glomerular filtration. RESULT: Because of the fractal structure of arterial branching, the flow rate is reduced in line with Strahler's order, so that work required (energy loss) is minimized to the cube root of flow rate. According to bifurcation analysis, mean arterial pressures between 70 and 100 mmHg can cause glomerulosclerosis, and a high gain in TGF signal can cause Limit cycle oscillations. CONCLUSION: The study concludes that nature has developed an optimal way of transferring blood from the Aorta to the Capillary bed by an evolutionary process such that the energy loss along the pathway is progressively reduced. Bifurcation analysis concludes that long and sustained oscillations due to underlying conditions such as diabetes, hypertension can lead to kidney damage.

  • Semantic segmentation of chest X-ray images based on the severity of COVID-19 infected patients

    Appears in:
    bebi 21(3): e3

    Authors:
    N. Filipović, J. Musulin, I. Lorencin, N. Anđelić, Z. Car, D. Štifanić, A. Blagojević, T. Šušteršič, S. Baressi Šegota, Z. Jurilj, S. Vlahinić

    Published:
    7th Jul 2021

    Abstract:
    INTRODUCTION: As a result of this global health crisis caused by the COVID-19 pandemic, the medical industry is searching for innovations that have the potential to automate the diagnostic process of COVID-19 and serve as an assistive tool for clinicians. OBJECTIVES: X-ray images have shown to be useful in the diagnosis of COVID-19. The goal of this research is to demonstrate an approach for automatic segmentation of lungs in chest X-ray images. METHODS: In this research DeepLabv3+ with Xception_65, MobileNetV2, and ResNet101 as backbones are used in order to perform lung segmentation. RESULTS: The proposed approach was experimented on X-ray images and has achieved an average mIOU of 0.910, F1 of 0.925, accuracy of 0.968, precision of 0.916, sensitivity of 0.935, and specificity of 0.977. CONCLUSION: Based on the obtained results, the proposed approach proved to be successful in terms of lung segmentation in chest X-ray images and has a great potential for clinical use.

  • Automated Pipeline for Continual Data Gathering and Retraining of the Machine Learning-Based COVID-19 Spread Models

    Appears in:
    bebi 21(3): e2

    Authors:
    S. Vlahinić, D. Štifanić, N. Anđelić, T. Šušteršič, A. Blagojević, I. Lorencin, S. Baressi Šegota, J. Musulin, Z. Car

    Published:
    4th May 2021

    Abstract:
    INTRODUCTION: The development of epidemiological curve models is one of the key factors in the combat of epidemiological diseases such as COVID-19. OBJECTIVES: The goal of this paper is to develop a system for automatic training and testing of AI-based regressive models of epidemiological curves using public data, which involves automating the data acquisition and speeding up the training of the models. METHODS: The research applies Multilayer Perceptron (MLP) for the creation of models, implemented within a system for automatic data fetching and training, and e valuated using the coefficient of determination (R2). Training time is lowered through the application of data filtering and simplifying the model selection. RESULTS: The developed system can train high precision models rapidly, allowing for quick model delivery All trained models achieve scores which are higher than 0.95. CONCLUSION: The results show that the development of a quick COVID-19 spread modeling system is possible.

  • Human Motion Enhancement via Joint Optimization of Kinematic and Anthropometric Constraints

    Appears in:
    bebi 21(3): e1

    Authors:
    Le Zhou, Guoliang Fan, Jerome Hausselle, Nate Lannan

    Published:
    9th Apr 2021

    Abstract:
    INTRODUCTION: Recently, low cost RGB-D depth sensors have emerged as a promising alternative for motion capture for clinical gait assessment. However, depth sensor based Mocap (D-Mocap) suffers from low accuracy and poor stability for 3D joint estimation due to noise, self-occlusion, interference, and other technical limitations, which prevents it from being widely used in health-related applications. OBJECTIVES: The primary objective of this study is to integrate non-linear Kalman filters (KFs) and evolutionary algorithms to enhance the quality of D-Mocap data by jointly considering kinematic or anthropometric constraints at the joint-level and skeleton-level, respectively. METHODS: We propose a hybrid approach to synergistically integrate the Tobit Kalman filter (TKF) and the Differential Evolutionary (DE) algorithm for human motion enhancement that is referred to as TKF-DE. Specifically, the joint-level TKF provides the predictive distribution of each joint that is kinematically admissible in time and probabilistically amenable in space for skeleton-level DE optimization in terms of all bone lengths. Two predictive distributions of the TKF, i.e., the Gaussian and uniform, are tested and compared in terms of their effectiveness in generating the initial DE population. RESULTS: Two sets of motion capture data are used to validate the proposed TKF-DE methods, one simulated and one real-world. The first dataset is from the Carnegie Mellon University Database (CMU) which contains a multitude of various motions and is simplified to a 21-joint skeleton and corrupted with additive white Gaussian noise (AWGN). The second dataset was collected at two labs at Oklahoma State University (OSU). The Orbbec depth sensor and the Nuitrack SDK were used for D-Mocap data acquisition along with an optical Mocap system that was time-synchronized and skeleton-matched with the D-Mocap system as a reference for evaluation. The results confirm that the proposed TKF-DE algorithms significantly outperform other nonlinear KFs, including the extended KF (EKF), the unscented KF (UKF), and the TKF alone, with improved accuracy and stability of the estimation of joint positions, bone lengths, and joint angles. It is also shown that the Gaussian-based predictive distribution is better than the uniform one, further validating the eÿcacy and synergy of the two key components in the TKF-DE algorithm. CONCLUSION: Our research synergistically integrates the TKF and DE algorithms in one framework referred to as the TKF-DE, that takes advantage of kinematic and anthropometric constraints for human motion enhancement. The experimental results on two D-Mocap datasets show that the proposed TKF-DE method significantly improves the quality of D-Mocap data in terms of joint positions, bone lengths, and joint angles. This study takes one step closer to bringing the D-Mocap technology to possible health-related applications.

  • Absence seizure detection classifying matching pursuit features of EEG signals

    Appears in:
    bebi 20(1): e4

    Authors:
    Michalis Zervakis, Katerina Giannakaki, Giorgos Giannakakis, Pelagia Vorgia

    Downloads:
    1328

    Abstract:
    INTRODUCTION: Absence seizures are characterized by a typical generalized spike-and-wave electroencephalographic (EEG) pattern around 3Hz. The automatic identification of this pattern and consequently its corresponding seizure is a valuable information towards the reliable patient’s clinical image and treatment planning. In this paper, we propose a method for absence seizures detection based on EEG signals decomposition via the Matching Pursuit (MP) algorithm. METHODS: Based on the ictal EEG semiology, MP features were extracted able to track the ictal pattern. Thisanalysis was performed in a clinical dataset of 8 pediatric patients (4 females, 4 males) suffering from activeabsence epilepsy, containing 123 absence seizures in total. Automatic classification schema based on Machine Learning techniques were employed to categorize the MP patterns into non-ictal and ictal states. RESULTS: The seizure detection system achieved a time window based discrimination accuracy of 97.3% byusing a Support Vector Machine (SVM) classifier and 10-fold cross-validation, in that way accomplishing a good state of the art performance. DISCUSSION: Compared to other popular spectral analysis methods, Matching Pursuit appears to be a robustand efficient method regarding absence seizures detection on EEG signals and our results indicate that the MP features proposed in this work are features that can be used effectively in seizure detection procedure.

  • Texture-based Feature Extraction for COVID-19 Pneumonia Classification using Chest Radiography

    Appears in:
    bebi 21(2): e2

    Authors:
    L. V. Moura, A. M. Marques da Silva, R. C. Barros, C. Mattjie, C. M. Dartora

    Downloads:
    1168

    Abstract:
    INTRODUCTION: The identification of COVID-19 pneumonia using chest radiography is challenging. OBJECTIVES: We investigate classification models to differentiate COVID-19-based and typical pneumonia in chest radiography. METHODS: We use 136 segmented chest X-rays to train and evaluate the performance of support vector machine (SVM), random forest (RF), AdaBoost (AB), and logistic regression (LR) classification methods. We use the PyRadiomics to extract statistical texture-based features in the right, left, and in six lung zones. We use a stratified k-folds (k=5) cross-validation within the training dataset, selecting the most relevant features with validation accuracy and relative feature importance. RESULTS: The AB model seems to be the best discriminant method, using six lung zones (AUC = 0.98). CONCLUSION: Our study shows a predominance of radiomic texture-based features related to COVID-19 pneumonia within the right lung, with a tendency within the upper lung zone.

  • Modelling COVID-19 using Fuzzy Cognitive Maps (FCM)

    Appears in:
    bebi 21(2): e3

    Authors:
    Peter P. Groumpos

    Downloads:
    1141

    Abstract:
    INTRODUCTION: The outbreak of COVID-19 has gained ground in many countries, leading towards a global health emergency. Research efforts have been intensified all around the humankind. OBJECTIVE: The main objective of this paper is to study the new pandemic COVID-19 using for the first-time theories of Fuzzy Cognitive Maps (FCM). METHODS: All known studies for COVID-19 are done based on statistical models. These statistical approaches depend solely on correlation factors. The factor of causality has not been considered due to the luck of sufficient mathematical models based on causality. Correlation does not imply causality while causality always implies correlation. The approach of Fuzzy Cognitive Maps (FCM) that is considering the causality factors is proposed, to investigate the whole spectrum of COVID-19. RESULTS: Early theoretical simulation studies using a COVID-19 FCM model have been conducted. Simulations with real patient data give excellent results. The state space Advanced Fuzzy Cognitive Maps (AFCM) is the natural sequence of the classical FCM theories. CONCLUSIONS: This study gives strong evidence that the “generic FCM theories” are probably the only ones that explore the causality between the variables of medical problems in a sound mathematical and scientific foundation.

  • Parallel Implementation of String-Based Clustering for HT-SELEX Data

    Appears in:
    bebi 20(1): e3

    Authors:
    Shintaro Kato, Takafumi Aoki, Koichi Ito, Hirotaka Minagawa, Katsunori Horii, Masaki Ito, Takayoshi Ono, Ikuo Shiratori, Iwao Waga

    Downloads:
    1134

    Abstract:
    INTRODUCTION: A clustering method for HT-SELEX is crucial for selecting different types of aptamer candidates. We have developed FSBC method for HT-SELEX data implemented in R. FSBC exhibited the highest accuracy of sequence clustering compared with conventional methods, while the processing time of FSBC is longer than AptaCluster. OBJECTIVES: The objective of this study is to improve the processing time of FSBC. METHODS: We propose pFSBC, which reduces the processing time of ORS estimation in FSBC by introducing parallel implementation. RESULTS: The processing time and clustering accuracy were evaluated with the last round of NCBI SRA data of SRR3279661 from BioProject PRJNA315881 comparing with other conventional clustering methods. We demonstrated that pFSBC exhibited the highest clustering accuracy and the shortest processing time. CONCLUSION: We expect that pFSBC will help to avoid the time-consuming clustering task, and it will provide accurate clustering results for the HT-SELEX data.

  • Well-being Forecasting using a Parametric Transfer-Learning method based on the Fisher Divergence and Hamiltonian Monte Carlo

    Appears in:
    bebi 20(1): e6

    Authors:
    Sergio Gonzalez-Martinez, Luca Ostacoli, Sara Carletto, Tasos Papastylianou, Eirini Christinaki, Manuel Ottaviano, Luca Citi, Riccardo Poli

    Downloads:
    1128

    Abstract:
    INTRODUCTION: Traditional personalised modelling typically requires sufficient personal data for training. This is a challenge in healthcare contexts, e.g. when using smartphones to predict well-being. OBJECTIVE: A method to produce incremental patient-specific models and forecasts even in the early stages of data collection when the data are sporadic and limited. METHODS: We propose a parametric transfer-learning method based on the Fisher divergence, where information from other patients is injected as a prior term into a Hamiltonian Monte Carlo framework. We test our method on the NEVERMIND dataset of self-reported well-being scores. RESULTS: Out of 54 scenarios representing varying training/forecasting lengths and competing methods, our method achieved overall best performance in 50 (92.6%) and demonstrated a significant median difference in45 (83.3%). CONCLUSION: The method performs favourably overall, particularly when long-term forecasts are required given short-term data.

  • Towards exploration and evaluation of sleep staging classification schemes for healthy and patient subjects

    Appears in:
    bebi 20(1): e2

    Authors:
    Christos Timplalexis, Dimitrios Chasanidis, Konstantinos Diamantaras, Ioanna Chouvarda

    Downloads:
    983

    Abstract:
    INTRODUCTION: Sleep stage classification is an important task for the timely diagnosis of sleep-related disorders, which are one the most common indicator of illness. OBJECTIVE: An automated sleep scoring implementation with promising generalization capabilities is presented, aiding towards eliminating the tedious procedure of manual sleep scoring. METHODS:Two Electroencephalogram (EEG) channels and the Electrooculogram (EOG) channel are utilized as inputs for feature extraction both in the time and frequency domain, while temporal feature changes are utilized in order to capture contextual information of the signals. An ensemble tree-based and a neural network approach are presented at the classification process. RESULTS: A total of 66 subjects belonging to three different groups (healthy, placebo, drug intake) were included in the study. The tree-based classification method outperforms the neural network at all cases. CONCLUSION: State of the art results are achieved, while it is highlighted that using jointly the healthy and patient subjects dataset, boosts the model’s accuracy and generalization capability.

  • Hybrid Adaptive Parametric Frequency Analysis

    Appears in:
    bebi 20(1): e5

    Authors:
    Kriton Konstantinidis, Emery N. Brown

    Downloads:
    966

    Abstract:
    INTRODUCTION: High quality spectra are very useful in anesthesia related procedures where Electroencephalogram (EEG) frequency content has been shown to drastically help track different brain states. A recent work (Konstantinidis & Brown, 2019 [1]) introduced the Gaussian Hybrid Autoregressive Model as a parametric method to generate smooth, very high resolution spectrograms of non-stationary EEG data of humans under propofol. OBJECTIVE: In this paper, we extend the model proposed in [1] to incorporate non-Gaussian state noise. METHODS: A Monte Carlo Markov Chain (MCMC) filtering procedure on a self-organizing state-space model is presented. RESULTS: We test the extended model on EEGs from human patients under propofol, ketamine and sevoflurane and illustrate the advantages over its Gaussian counterpart. CONCLUSION: The suitability of the proposed method for online use, in combination with its ability to smoothly track frequency changes in human EEG signals under the most common anesthetics, suggests that it can be used for real-time brain state tracking. Such online use can facilitate the design of more precise closed loop systems for automatic control of brain states under general anesthesia.

  • Combined machine learning and finite element simulation approach towards personalized model for prognosis of COVID-19 disease development in patients

    Appears in:
    bebi 21(2): e6

    Authors:
    Anđela Blagojević, Zlatan Car, Nenad Filipović, Dragan Milovanović, Ivan Lorencin, Dejan Baskić, Sandi Baressi Šegota, Danijela Baskić, Tijana Šušteršič

    Downloads:
    900

    Abstract:
    INTRODUCTION: Machine learning algorithms and in silico models for the COVID-19 have been used to classify infectious people and predict their condition in time. OBJECTIVES: This study aims at creating a personalized model that combines machine learning and finite element simulation approach in order to predict development of COVID-19 infection in patients. METHODS: The methodology combines several aspects (1) classification of patients into several classes of clinical condition (2) segmentation of human lungs in X ray images (3) finite element simulation to investigate the spreading of SARS-COV-2 virion in the lungs. RESULTS: The findings show accuracy larger than 90% in all aspects of methodology. FE simulation has revealed that the distribution of airflow in the lung changes in time with the infection. CONCLUSION: The key benefit of our proposed method is that it combines several methods that will be further improved in order to create a truly unique combined methodology for predictive models in patients infected with COVID-19.

  • Robust Heatmap Template Generation for COVID-19 Biomarker Detection

    Appears in:
    bebi 21(2): e4

    Authors:
    Miguel Lerma, Mirtha Lucas, Jacob Furst, Daniela Raicu

    Downloads:
    876

    Abstract:
    INTRODUCTION: Detecting and identifying patterns in chest X-ray images of Covid-19 patients are important tasks for understanding the disease and for making differential diagnosis. OBJECTIVES: The purpose of this work is to develop a technique for detecting biomarkers of four possible conditions in chest X-rays, and study the patterns arising from the location of biomarkers. METHODS: We use transfer learning applied to a pretrained VGG19 neural network to build a model capable of detecting the four conditions in chest X-rays. For biomarkers detection we use Grad-CAM. Patterns in the biomarkers are found by using classical eigenfaces approach. RESULTS: The discovered patterns are consistent across images from a given class of disease, and they are robust with respect to changes in dataset. CONCLUSION: The identified patterns can serve as biomarkers for a given disease in chest X-ray images, and constitute explanations of how the deep learning model makes classification decisions.

  • Safe in COVID-19: A platform to support effective monitoring of incidents during a pandemic

    Appears in:
    bebi 21(2): e5

    Authors:
    G. Kavlentakis, Y. Petrakis, H. Kondylakis, D.G. Katehakis, V. Tzikoulis, S. Kostomanolakis, N. Stathiakis, F. Logothetidis, A. Kouroubali

    Downloads:
    793

    Abstract:
    INTRODUCTION: The ongoing coronavirus pandemic is affecting the lives of millions of people, changing society by enacting new rules for social life, business and travel. The world community (i.e. international organizations, public administrations, businesses, and citizens) has conducted a huge effort in delivering digital solutions to properly address the challenges imposed. A multitude of approaches have been followed, towards achieving a wide spectrum of side goals. OBJECTIVES: This work presents a platform, designed for public health authorities, to effectively track suspect, probable and confirmed incidence cases in a pandemic by means of a mobile app used by citizens to provide immediate feedback.The aim has been to better support human tracers, and the platform must not be confused with proximity tracking apps. METHODS: The outbreak response tool was developed based on official information and guidelines, on top of an already existing personal health record app which has been extended, to properly accommodate specific needs that emerged during the crisis. RESULTS: The developed platform provides the framework to support return to the “new normal” in less time, with reduced stress and more security for individuals, more direct and safer management of patients by physicians, and better possibilities for monitoring the epidemic by public health authorities. Issues relevant to privacy concerns, and interoperability with available patient registries and data analytics tools were also examined to better support public healthcare delivery and contain the spread of the infection. CONCLUSION: In order for the foreseen benefits to be realized, there is a need to respect safety and security regulations, while at the same time conform to international standards and widely accepted medical protocols. Cross-border interoperability and the availability of appropriate links (i.e. publicly available interfaces) to relevant open data and national registries is considered to be of paramount importance.
Publisher
EAI
ISSN
2709-4111
Number of Volumes
1
Last Published
2021-03-24
EAI Logo

About EAI

Community

Publish with EAI