Machine Learning for Biomedical Applications

A special issue of Bioengineering (ISSN 2306-5354). This special issue belongs to the section "Biosignal Processing".

Deadline for manuscript submissions: closed (30 September 2022) | Viewed by 51347

Special Issue Editors

Department of Chemical, Materials and Production Engineering, University of Naples Federico II, Piazzale Tecchio 80, 80125 Naples, Italy
Interests: biomedical engineering; bioengineering; biomedical data analysis; biomedical signal processing; drug delivery systems; biomaterials; polymer microparticles; lean six sigma in healthcare
Special Issues, Collections and Topics in MDPI journals
Department of Electrical Engineering and Information Technology, University of Naples Federico II, Via Claudio 21, 80125 Naples, Italy
Interests: machine learning; statistics; gait analysis; health technology assessment; lean six sigma; biomedical engineering
Special Issues, Collections and Topics in MDPI journals
Department of Advanced Medical and Surgical Sciences, University of Campania “Luigi Vanvitelli”, Piazza Luigi Miraglia, 2, 80138 Naples, Italy
Interests: biomedical engineering; biosignal and bioimage processing; ergonomics; rehabilitation engineering, gait analysis, wearable sensors; telemedicine; machine learning; biostatistics
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

In recent years, the collection and the analysis of medical data coupled with the advances in the artificial intelligence have allowed us to achieve potentially promising advances in the biomedical field. In this context, machine learning has demonstrated itself to be a very promising tool to both gain new insights from medical-related data and to pursue different objectives and find new alternative solutions which are potentially applicable to solve biomedical complex issues in a more automated, favorable, and rapid way. This Special Issue seeks high-quality contributions (articles, reviews, communications, etc.) which propose novel research and address recent progress applying machine learning strategies to the biomedical field, including, but not limited to, image and signal processing aimed at diagnosis and rehabilitation.

The topics of interest include, but are not limited to, the following application fields for machine learning:

  • Biomedical and health informatics;
  • Biomedical data processing from wearable sensors;
  • Biomedical signal processing;
  • Healthcare innovation;
  • Image processing applications (e.g., radiomics, …)
  • Internet of Things;
  • Machine learning for bio-robotics and biomechanics;
  • Neural and rehabilitation engineering;
  • Support decision-making for medical diagnoses.

Dr. Giuseppe Cesarelli
Dr. Alfonso Ponsiglione
Dr. Carlo Ricciardi
Dr. Leandro Donisi
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Bioengineering is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2700 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • biomedical informatics
  • biomedical signal processing
  • computer-aided diagnosis
  • human activity recognition
  • image processing
  • inertial measurements units and sensors for IoT
  • machine learning
  • medical diagnoses
  • modeling and simulation
  • neural networks
  • rehabilitation medicine

Related Special Issue

Published Papers (21 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Jump to: Review, Other

15 pages, 9666 KiB  
Article
Understanding the Role of Connectivity Dynamics of Resting-State Functional MRI in the Diagnosis of Autism Spectrum Disorder: A Comprehensive Study
by Yaser ElNakieb, Mohamed T. Ali, Ahmed Elnakib, Ahmed Shalaby, Ali Mahmoud, Ahmed Soliman, Gregory Neal Barnes and Ayman El-Baz
Bioengineering 2023, 10(1), 56; https://doi.org/10.3390/bioengineering10010056 - 02 Jan 2023
Cited by 9 | Viewed by 2062
Abstract
In addition to the standard observational assessment for autism spectrum disorder (ASD), recent advancements in neuroimaging and machine learning (ML) suggest a rapid and objective alternative using brain imaging. This work presents a pipelined framework, using functional magnetic resonance imaging (fMRI) that allows [...] Read more.
In addition to the standard observational assessment for autism spectrum disorder (ASD), recent advancements in neuroimaging and machine learning (ML) suggest a rapid and objective alternative using brain imaging. This work presents a pipelined framework, using functional magnetic resonance imaging (fMRI) that allows not only an accurate ASD diagnosis but also the identification of the brain regions contributing to the diagnosis decision. The proposed framework includes several processing stages: preprocessing, brain parcellation, feature representation, feature selection, and ML classification. For feature representation, the proposed framework uses both a conventional feature representation and a novel dynamic connectivity representation to assist in the accurate classification of an autistic individual. Based on a large publicly available dataset, this extensive research highlights different decisions along the proposed pipeline and their impact on diagnostic accuracy. A large publicly available dataset of 884 subjects from the Autism Brain Imaging Data Exchange I (ABIDE-I) initiative is used to validate our proposed framework, achieving a global balanced accuracy of 98.8% with five-fold cross-validation and proving the potential of the proposed feature representation. As a result of this comprehensive study, we achieve state-of-the-art accuracy, confirming the benefits of the proposed feature representation and feature engineering in extracting useful information as well as the potential benefits of utilizing ML and neuroimaging in the diagnosis and understanding of autism. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Figure 1

27 pages, 4366 KiB  
Article
Automated Lung-Related Pneumonia and COVID-19 Detection Based on Novel Feature Extraction Framework and Vision Transformer Approaches Using Chest X-ray Images
by Chiagoziem C. Ukwuoma, Zhiguang Qin, Md Belal Bin Heyat, Faijan Akhtar, Abla Smahi, Jehoiada K. Jackson, Syed Furqan Qadri, Abdullah Y. Muaad, Happy N. Monday and Grace U. Nneji
Bioengineering 2022, 9(11), 709; https://doi.org/10.3390/bioengineering9110709 - 18 Nov 2022
Cited by 19 | Viewed by 6060
Abstract
According to research, classifiers and detectors are less accurate when images are blurry, have low contrast, or have other flaws which raise questions about the machine learning model’s ability to recognize items effectively. The chest X-ray image has proven to be the preferred [...] Read more.
According to research, classifiers and detectors are less accurate when images are blurry, have low contrast, or have other flaws which raise questions about the machine learning model’s ability to recognize items effectively. The chest X-ray image has proven to be the preferred image modality for medical imaging as it contains more information about a patient. Its interpretation is quite difficult, nevertheless. The goal of this research is to construct a reliable deep-learning model capable of producing high classification accuracy on chest x-ray images for lung diseases. To enable a thorough study of the chest X-ray image, the suggested framework first derived richer features using an ensemble technique, then a global second-order pooling is applied to further derive higher global features of the images. Furthermore, the images are then separated into patches and position embedding before analyzing the patches individually via a vision transformer approach. The proposed model yielded 96.01% sensitivity, 96.20% precision, and 98.00% accuracy for the COVID-19 Radiography Dataset while achieving 97.84% accuracy, 96.76% sensitivity and 96.80% precision, for the Covid-ChestX-ray-15k dataset. The experimental findings reveal that the presented models outperform traditional deep learning models and other state-of-the-art approaches provided in the literature. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Figure 1

16 pages, 9125 KiB  
Article
A Surrogate Model Based on a Finite Element Model of Abdomen for Real-Time Visualisation of Tissue Stress during Physical Examination Training
by Florence Leong, Chow Yin Lai, Siamak Farajzadeh Khosroshahi, Liang He, Simon de Lusignan, Thrishantha Nanayakkara and Mazdak Ghajari
Bioengineering 2022, 9(11), 687; https://doi.org/10.3390/bioengineering9110687 - 14 Nov 2022
Cited by 2 | Viewed by 3770
Abstract
Robotic patients show great potential for helping to improve medical palpation training, as they can provide feedback that cannot be obtained in a real patient. They provide information about internal organ deformation that can significantly enhance palpation training by giving medical trainees visual [...] Read more.
Robotic patients show great potential for helping to improve medical palpation training, as they can provide feedback that cannot be obtained in a real patient. They provide information about internal organ deformation that can significantly enhance palpation training by giving medical trainees visual insight based on the pressure they apply for palpation. This can be achieved by using computational models of abdomen mechanics. However, such models are computationally expensive, and thus unable to provide real-time predictions. In this work, we proposed an innovative surrogate model of abdomen mechanics by using machine learning (ML) and finite element (FE) modelling to virtually render internal tissue deformation in real time. We first developed a new high-fidelity FE model of the abdomen mechanics from computerized tomography (CT) images. We performed palpation simulations to produce a large database of stress distribution on the liver edge, an area of interest in most examinations. We then used artificial neural networks (ANNs) to develop the surrogate model and demonstrated its application in an experimental palpation platform. Our FE simulations took 1.5 h to predict stress distribution for each palpation while this only took a fraction of a second for the surrogate model. Our results show that our artificial neural network (ANN) surrogate has an accuracy of 92.6%. We also showed that the surrogate model is able to use the experimental input of palpation location and force to provide real-time projections onto the robotics platform. This enhanced robotics platform has the potential to be used as a training simulator for trainees to hone their palpation skills. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Figure 1

19 pages, 7429 KiB  
Article
Level-Set-Based Kidney Segmentation from DCE-MRI Using Fuzzy Clustering with Population-Based and Subject-Specific Shape Statistics
by Moumen El-Melegy, Rasha Kamel, Mohamed Abou El-Ghar, Norah S. Alghamdi and Ayman El-Baz
Bioengineering 2022, 9(11), 654; https://doi.org/10.3390/bioengineering9110654 - 05 Nov 2022
Cited by 3 | Viewed by 1272
Abstract
The segmentation of dynamic contrast-enhanced magnetic resonance images (DCE-MRI) of the kidney is a fundamental step in the early and noninvasive detection of acute renal allograft rejection. In this paper, a new and accurate DCE-MRI kidney segmentation method is proposed. In this method, [...] Read more.
The segmentation of dynamic contrast-enhanced magnetic resonance images (DCE-MRI) of the kidney is a fundamental step in the early and noninvasive detection of acute renal allograft rejection. In this paper, a new and accurate DCE-MRI kidney segmentation method is proposed. In this method, fuzzy c-means (FCM) clustering is embedded into a level set method, with the fuzzy memberships being iteratively updated during the level set contour evolution. Moreover, population-based shape (PB-shape) and subject-specific shape (SS-shape) statistics are both exploited. The PB-shape model is trained offline from ground-truth kidney segmentations of various subjects, whereas the SS-shape model is trained on the fly using the segmentation results that are obtained for a specific subject. The proposed method was evaluated on the real medical datasets of 45 subjects and reports a Dice similarity coefficient (DSC) of 0.953 ± 0.018, an intersection-over-union (IoU) of 0.91 ± 0.033, and 1.10 ± 1.4 in the 95-percentile of Hausdorff distance (HD95). Extensive experiments confirm the superiority of the proposed method over several state-of-the-art level set methods, with an average improvement of 0.7 in terms of HD95. It also offers an HD95 improvement of 9.5 and 3.8 over two deep neural networks based on the U-Net architecture. The accuracy improvements have been experimentally found to be more prominent on low-contrast and noisy images. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Graphical abstract

15 pages, 611 KiB  
Article
A Bioinformatics View on Acute Myeloid Leukemia Surface Molecules by Combined Bayesian and ABC Analysis
by Michael C. Thrun, Elisabeth K. M. Mack, Andreas Neubauer, Torsten Haferlach, Miriam Frech, Alfred Ultsch and Cornelia Brendel
Bioengineering 2022, 9(11), 642; https://doi.org/10.3390/bioengineering9110642 - 03 Nov 2022
Cited by 1 | Viewed by 1986
Abstract
“Big omics data” provoke the challenge of extracting meaningful information with clinical benefit. Here, we propose a two-step approach, an initial unsupervised inspection of the structure of the high dimensional data followed by supervised analysis of gene expression levels, to reconstruct the surface [...] Read more.
“Big omics data” provoke the challenge of extracting meaningful information with clinical benefit. Here, we propose a two-step approach, an initial unsupervised inspection of the structure of the high dimensional data followed by supervised analysis of gene expression levels, to reconstruct the surface patterns on different subtypes of acute myeloid leukemia (AML). First, Bayesian methodology was used, focusing on surface molecules encoded by cluster of differentiation (CD) genes to assess whether AML is a homogeneous group or segregates into clusters. Gene expressions of 390 patient samples measured using microarray technology and 150 samples measured via RNA-Seq were compared. Beyond acute promyelocytic leukemia (APL), a well-known AML subentity, the remaining AML samples were separated into two distinct subgroups. Next, we investigated which CD molecules would best distinguish each AML subgroup against APL, and validated discriminative molecules of both datasets by searching the scientific literature. Surprisingly, a comparison of both omics analyses revealed that CD339 was the only overlapping gene differentially regulated in APL and other AML subtypes. In summary, our two-step approach for gene expression analysis revealed two previously unknown subgroup distinctions in AML based on surface molecule expression, which may guide the differentiation of subentities in a given clinical–diagnostic context. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Graphical abstract

14 pages, 2915 KiB  
Article
A Study of Projection-Based Attentive Spatial–Temporal Map for Remote Photoplethysmography Measurement
by Dae-Yeol Kim, Soo-Young Cho, Kwangkee Lee and Chae-Bong Sohn
Bioengineering 2022, 9(11), 638; https://doi.org/10.3390/bioengineering9110638 - 02 Nov 2022
Cited by 3 | Viewed by 1753
Abstract
The photoplethysmography (PPG) signal contains various information that is related to CVD (cardiovascular disease). The remote PPG (rPPG) is a method that can measure a PPG signal using a face image taken with a camera, without a PPG device. Deep learning-based rPPG methods [...] Read more.
The photoplethysmography (PPG) signal contains various information that is related to CVD (cardiovascular disease). The remote PPG (rPPG) is a method that can measure a PPG signal using a face image taken with a camera, without a PPG device. Deep learning-based rPPG methods can be classified into three main categories. First, there is a 3D CNN approach that uses a facial image video as input, which focuses on the spatio-temporal changes in the facial video. The second approach is a method that uses a spatio-temporal map (STMap), and the video image is pre-processed using the point where it is easier to analyze changes in blood flow in time order. The last approach uses a preprocessing model with a dichromatic reflection model. This study proposed the concept of an axis projection network (APNET) that complements the drawbacks, in which the 3D CNN method requires significant memory; the STMap method requires a preprocessing method; and the dyschromatic reflection model (DRM) method does not learn long-term temporal characteristics. We also showed that the proposed APNET effectively reduced the network memory size, and that the low-frequency signal was observed in the inferred PPG signal, suggesting that it can provide meaningful results to the study when developing the rPPG algorithm. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Figure 1

18 pages, 671 KiB  
Article
Feature–Classifier Pairing Compatibility for sEMG Signals in Hand Gesture Recognition under Joint Effects of Processing Procedures
by Mohammed Asfour, Carlo Menon and Xianta Jiang
Bioengineering 2022, 9(11), 634; https://doi.org/10.3390/bioengineering9110634 - 02 Nov 2022
Cited by 3 | Viewed by 1274
Abstract
Gesture recognition using surface electromyography (sEMG) serves many applications, from human–machine interfaces to prosthesis control. Many features have been adopted to enhance recognition accuracy. However, studies mostly compare features under a prechosen feature window size or a classifier, biased to a specific application. [...] Read more.
Gesture recognition using surface electromyography (sEMG) serves many applications, from human–machine interfaces to prosthesis control. Many features have been adopted to enhance recognition accuracy. However, studies mostly compare features under a prechosen feature window size or a classifier, biased to a specific application. The bias is evident in the reported accuracy drop, around 10%, from offline gesture recognition in experiment settings to real-time clinical environment studies. This paper explores the feature–classifier pairing compatibility for sEMG. We demonstrate that it is the primary determinant of gesture recognition accuracy under various window sizes and normalization ranges, thus removing application bias. The proposed pairing ranking provides a guideline for choosing the proper feature or classifier in future research. For instance, random forest (RF) performed best, with a mean accuracy of around 74.0%; however, it was optimal with the mean absolute value feature (MAV), giving 86.8% accuracy. Additionally, our ranking showed that the proper pairing enables low-computational models to surpass complex ones. The Histogram feature with linear discriminant analysis classifier (HIST-LDA) was the top pair with 88.6% accuracy. We also concluded that a 1250 ms window and a (−1, 1) signal normalization were the optimal procedures for gesture recognition on the used dataset. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Graphical abstract

19 pages, 4326 KiB  
Article
PHF3 Technique: A Pyramid Hybrid Feature Fusion Framework for Severity Classification of Ulcerative Colitis Using Endoscopic Images
by Jing Qi, Guangcong Ruan, Jia Liu, Yi Yang, Qian Cao, Yanling Wei and Yongjian Nian
Bioengineering 2022, 9(11), 632; https://doi.org/10.3390/bioengineering9110632 - 01 Nov 2022
Cited by 2 | Viewed by 1800
Abstract
Evaluating the severity of ulcerative colitis (UC) through the Mayo endoscopic subscore (MES) is crucial for understanding patient conditions and providing effective treatment. However, UC lesions present different characteristics in endoscopic images, exacerbating interclass similarities and intraclass differences in MES classification. In addition, [...] Read more.
Evaluating the severity of ulcerative colitis (UC) through the Mayo endoscopic subscore (MES) is crucial for understanding patient conditions and providing effective treatment. However, UC lesions present different characteristics in endoscopic images, exacerbating interclass similarities and intraclass differences in MES classification. In addition, inexperience and review fatigue in endoscopists introduces nontrivial challenges to the reliability and repeatability of MES evaluations. In this paper, we propose a pyramid hybrid feature fusion framework (PHF3) as an auxiliary diagnostic tool for clinical UC severity classification. Specifically, the PHF3 model has a dual-branch hybrid architecture with ResNet50 and a pyramid vision Transformer (PvT), where the local features extracted by ResNet50 represent the relationship between the intestinal wall at the near-shot point and its depth, and the global representations modeled by the PvT capture similar information in the cross-section of the intestinal cavity. Furthermore, a feature fusion module (FFM) is designed to combine local features with global representations, while second-order pooling (SOP) is applied to enhance discriminative information in the classification process. The experimental results show that, compared with existing methods, the proposed PHF3 model has competitive performance. The area under the receiver operating characteristic curve (AUC) of MES 0, MES 1, MES 2, and MES 3 reached 0.996, 0.972, 0.967, and 0.990, respectively, and the overall accuracy reached 88.91%. Thus, our proposed method is valuable for developing an auxiliary assessment system for UC severity. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Graphical abstract

16 pages, 3351 KiB  
Article
Machine Learning for Aiding Blood Flow Velocity Estimation Based on Angiography
by Swati Padhee, Mark Johnson, Hang Yi, Tanvi Banerjee and Zifeng Yang
Bioengineering 2022, 9(11), 622; https://doi.org/10.3390/bioengineering9110622 - 28 Oct 2022
Cited by 2 | Viewed by 1976
Abstract
Computational fluid dynamics (CFD) is widely employed to predict hemodynamic characteristics in arterial models, while not friendly to clinical applications due to the complexity of numerical simulations. Alternatively, this work proposed a framework to estimate hemodynamics in vessels based on angiography images using [...] Read more.
Computational fluid dynamics (CFD) is widely employed to predict hemodynamic characteristics in arterial models, while not friendly to clinical applications due to the complexity of numerical simulations. Alternatively, this work proposed a framework to estimate hemodynamics in vessels based on angiography images using machine learning (ML) algorithms. First, the iodine contrast perfusion in blood was mimicked by a flow of dye diffusing into water in the experimentally validated CFD modeling. The generated projective images from simulations imitated the counterpart of light passing through the flow field as an analogy of X-ray imaging. Thus, the CFD simulation provides both the ground truth velocity field and projective images of dye flow patterns. The rough velocity field was estimated using the optical flow method (OFM) based on 53 projective images. ML training with least absolute shrinkage, selection operator and convolutional neural network was conducted with CFD velocity data as the ground truth and OFM velocity estimation as the input. The performance of each model was evaluated based on mean absolute error and mean squared error, where all models achieved or surpassed the criteria of 3 × 10−3 and 5 × 10−7 m/s, respectively, with a standard deviation less than 1 × 10−6 m/s. Finally, the interpretable regression and ML models were validated with over 613 image sets. The validation results showed that the employed ML model significantly reduced the error rate from 53.5% to 2.5% on average for the v-velocity estimation in comparison with CFD. The ML framework provided an alternative pathway to support clinical diagnosis by predicting hemodynamic information with high efficiency and accuracy. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Graphical abstract

15 pages, 2640 KiB  
Article
Classification of Autism and Control Gait in Children Using Multisegment Foot Kinematic Features
by Ashirbad Pradhan, Victoria Chester and Karansinh Padhiar
Bioengineering 2022, 9(10), 552; https://doi.org/10.3390/bioengineering9100552 - 14 Oct 2022
Cited by 4 | Viewed by 2257
Abstract
Previous research has demonstrated that children with autism walk with atypical ankle kinematics and kinetics. Although these studies have utilized single-segment foot (SSF) data, multisegment foot (MSF) kinematics can provide further information on foot mechanics. Machine learning (ML) tools allow the combination of [...] Read more.
Previous research has demonstrated that children with autism walk with atypical ankle kinematics and kinetics. Although these studies have utilized single-segment foot (SSF) data, multisegment foot (MSF) kinematics can provide further information on foot mechanics. Machine learning (ML) tools allow the combination of MSF kinematic features for classifying autism gait patterns. In this study, multiple ML models are investigated, and the most contributing features are determined. This study involved 19 children with autism and 21 age-matched controls performing walking trials. A 34-marker system and a 12-camera motion capture system were used to compute SSF and MSF angles during walking. Features extracted from these foot angles and their combinations were used to develop support vector machine (SVM) models. Additional techniques-S Hapley Additive exPlanations (SHAP) and the Shapley Additive Global importancE (SAGE) are used for local and global importance of the black-box ML models. The results suggest that models based on combinations of MSF kinematic features classify autism patterns with an accuracy of 96.3%, which is higher than using SSF kinematic features (83.8%). The relative angle between the metatarsal and midfoot segments had the highest contribution to the classification of autism gait patterns. The study demonstrated that kinematic features from MSF angles, supported by ML models, can provide an accurate and interpretable classification of autism and control patterns in children. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Graphical abstract

16 pages, 1647 KiB  
Article
Implementation of Predictive Algorithms for the Study of the Endarterectomy LOS
by Teresa Angela Trunfio, Anna Borrelli and Giovanni Improta
Bioengineering 2022, 9(10), 546; https://doi.org/10.3390/bioengineering9100546 - 12 Oct 2022
Cited by 11 | Viewed by 934
Abstract
Background: In recent years, the length of hospital stay (LOS) following endarterectomy has decreased significantly from 4 days to 1 day. LOS is influenced by several common complications and factors that can adversely affect the patient’s health and may vary from one healthcare [...] Read more.
Background: In recent years, the length of hospital stay (LOS) following endarterectomy has decreased significantly from 4 days to 1 day. LOS is influenced by several common complications and factors that can adversely affect the patient’s health and may vary from one healthcare facility to another. The aim of this work is to develop a forecasting model of the LOS value to investigate the main factors affecting LOS in order to save healthcare cost and improve management. Methods: We used different regression and machine learning models to predict the LOS value based on the clinical and organizational data of patients undergoing endarterectomy. Data were obtained from the discharge forms of the “San Giovanni di Dio e Ruggi d’Aragona” University Hospital (Salerno, Italy). R2 goodness of fit and the results in terms of accuracy, precision, recall and F1-score were used to compare the performance of various algorithms. Results: Before implementing the models, the preliminary correlation study showed that LOS was more dependent on the type of endarterectomy performed. Among the regression algorithms, the best was the multiple linear regression model with an R2 value of 0.854, while among the classification algorithms for LOS divided into classes, the best was decision tree, with an accuracy of 80%. The best performance was obtained in the third class, which identifies patients with prolonged LOS, with a precision of 95%. Among the independent variables, the most influential on LOS was type of endarterectomy, followed by diabetes and kidney disorders. Conclusion: The resulting forecast model demonstrates its effectiveness in predicting the value of LOS that could be used to improve the endarterectomy surgery planning. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Figure 1

14 pages, 433 KiB  
Article
Predicting the Level of Respiratory Support in COVID-19 Patients Using Machine Learning
by Hisham Abdeltawab, Fahmi Khalifa, Yaser ElNakieb, Ahmed Elnakib, Fatma Taher, Norah Saleh Alghamdi, Harpal Singh Sandhu and Ayman El-Baz
Bioengineering 2022, 9(10), 536; https://doi.org/10.3390/bioengineering9100536 - 09 Oct 2022
Viewed by 1333
Abstract
In this paper, a machine learning-based system for the prediction of the required level of respiratory support in COVID-19 patients is proposed. The level of respiratory support is divided into three classes: class 0 which refers to minimal support, class 1 which refers [...] Read more.
In this paper, a machine learning-based system for the prediction of the required level of respiratory support in COVID-19 patients is proposed. The level of respiratory support is divided into three classes: class 0 which refers to minimal support, class 1 which refers to non-invasive support, and class 2 which refers to invasive support. A two-stage classification system is built. First, the classification between class 0 and others is performed. Then, the classification between class 1 and class 2 is performed. The system is built using a dataset collected retrospectively from 3491 patients admitted to tertiary care hospitals at the University of Louisville Medical Center. The use of the feature selection method based on analysis of variance is demonstrated in the paper. Furthermore, a dimensionality reduction method called principal component analysis is used. XGBoost classifier achieves the best classification accuracy (84%) in the first stage. It also achieved optimal performance in the second stage, with a classification accuracy of 83%. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Figure 1

14 pages, 1502 KiB  
Article
Altered Microcirculation in Alzheimer’s Disease Assessed by Machine Learning Applied to Functional Thermal Imaging Data
by David Perpetuini, Chiara Filippini, Michele Zito, Daniela Cardone and Arcangelo Merla
Bioengineering 2022, 9(10), 492; https://doi.org/10.3390/bioengineering9100492 - 21 Sep 2022
Cited by 4 | Viewed by 1534
Abstract
Alzheimer’s disease (AD) is characterized by progressive memory failures accompanied by microcirculation alterations. Particularly, impaired endothelial microvascular responsiveness and altered flow motion patterns have been observed in AD patients. Of note, the endothelium influences the vascular tone and also the small superficial blood [...] Read more.
Alzheimer’s disease (AD) is characterized by progressive memory failures accompanied by microcirculation alterations. Particularly, impaired endothelial microvascular responsiveness and altered flow motion patterns have been observed in AD patients. Of note, the endothelium influences the vascular tone and also the small superficial blood vessels, which can be evaluated through infrared thermography (IRT). The advantage of IRT with respect to other techniques relies on its contactless features and its capability to preserve spatial information of the peripheral microcirculation. The aim of the study is to investigate peripheral microcirculation impairments in AD patients with respect to age-matched healthy controls (HCs) at resting state, through IRT and machine learning (ML) approaches. Particularly, several classifiers were tested, employing as regressors the power of the nose tip temperature time course in different physiological frequency bands. Among the ML classifiers tested, the Decision Tree Classifier (DTC) delivered the best cross-validated accuracy (accuracy = 82%) when discriminating between AD and HCs. The results further demonstrate the alteration of microvascular patterns in AD in the early stages of the pathology, and the capability of IRT to assess vascular impairments. These findings could be exploited in clinical practice, fostering the employment of IRT as a support for the early diagnosis of AD. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Figure 1

28 pages, 4349 KiB  
Article
Decision Support System for Liver Lesion Segmentation Based on Advanced Convolutional Neural Network Architectures
by Dan Popescu, Andrei Stanciulescu, Mihai Dan Pomohaci and Loretta Ichim
Bioengineering 2022, 9(9), 467; https://doi.org/10.3390/bioengineering9090467 - 13 Sep 2022
Cited by 1 | Viewed by 1715
Abstract
Given its essential role in body functions, liver cancer is the third most common cause of death from cancer, despite being the sixth most common type of cancer worldwide. Following advancements in medicine and image processing, medical image segmentation methods are receiving a [...] Read more.
Given its essential role in body functions, liver cancer is the third most common cause of death from cancer, despite being the sixth most common type of cancer worldwide. Following advancements in medicine and image processing, medical image segmentation methods are receiving a great deal of attention. As a novelty, the paper proposes an intelligent decision system for segmenting liver and hepatic tumors by integrating four efficient neural networks (ResNet152, ResNeXt101, DenseNet201, and InceptionV3). Images from computed tomography for training, validation, and testing were taken from the public LiTS17 database and preprocessed to better highlight liver tissue and tumors. Global segmentation is done by separately training individual classifiers and the global system of merging individual decisions. For the aforementioned application, classification neural networks have been modified for semantic segmentation. After segmentation based on the neural network system, the images were postprocessed to eliminate artifacts. The segmentation results obtained by the system were better, from the point of view of the Dice coefficient, than those obtained by the individual networks, and comparable with those reported in recent works. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Graphical abstract

18 pages, 3090 KiB  
Article
Cuffless Blood Pressure Estimation Using Calibrated Cardiovascular Dynamics in the Photoplethysmogram
by Hamed Samimi and Hilmi R. Dajani
Bioengineering 2022, 9(9), 446; https://doi.org/10.3390/bioengineering9090446 - 06 Sep 2022
Cited by 9 | Viewed by 1997
Abstract
An important means for preventing and managing cardiovascular disease is the non-invasive estimation of blood pressure. There is particular interest in developing approaches that provide accurate cuffless and continuous estimation of this important vital sign. This paper proposes a method that uses dynamic [...] Read more.
An important means for preventing and managing cardiovascular disease is the non-invasive estimation of blood pressure. There is particular interest in developing approaches that provide accurate cuffless and continuous estimation of this important vital sign. This paper proposes a method that uses dynamic changes of the pulse waveform over short time intervals and calibrates the system based on a mathematical model that relates reflective PTT (R-PTT) to blood pressure. An advantage of the method is that it only requires collecting the photoplethysmogram (PPG) using one optical sensor, in addition to initial non-invasive measurements of blood pressure that are used for calibration. This method was applied to data from 30 patients, resulting in a mean error (ME) of 0.59 mmHg, a standard deviation of error (SDE) of 7.07 mmHg, and a mean absolute error (MAE) of 4.92 mmHg for diastolic blood pressure (DBP) and an ME of 2.52 mmHg, an SDE of 12.15 mmHg, and an MAE of 8.89 mmHg for systolic blood pressure (SBP). These results demonstrate the possibility of using the PPG signal for the cuffless continuous estimation of blood pressure based on the analysis of calibrated changes in cardiovascular dynamics, possibly in conjunction with other methods that are currently being researched. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Graphical abstract

18 pages, 27909 KiB  
Article
Development and Evaluation of a Novel Deep-Learning-Based Framework for the Classification of Renal Histopathology Images
by Yasmine Abu Haeyeh, Mohammed Ghazal, Ayman El-Baz and Iman M. Talaat
Bioengineering 2022, 9(9), 423; https://doi.org/10.3390/bioengineering9090423 - 30 Aug 2022
Cited by 8 | Viewed by 3915
Abstract
Kidney cancer has several types, with renal cell carcinoma (RCC) being the most prevalent and severe type, accounting for more than 85% of adult patients. The manual analysis of whole slide images (WSI) of renal tissues is the primary tool for RCC diagnosis [...] Read more.
Kidney cancer has several types, with renal cell carcinoma (RCC) being the most prevalent and severe type, accounting for more than 85% of adult patients. The manual analysis of whole slide images (WSI) of renal tissues is the primary tool for RCC diagnosis and prognosis. However, the manual identification of RCC is time-consuming and prone to inter-subject variability. In this paper, we aim to distinguish between benign tissue and malignant RCC tumors and identify the tumor subtypes to support medical therapy management. We propose a novel multiscale weakly-supervised deep learning approach for RCC subtyping. Our system starts by applying the RGB-histogram specification stain normalization on the whole slide images to eliminate the effect of the color variations on the system performance. Then, we follow the multiple instance learning approach by dividing the input data into multiple overlapping patches to maintain the tissue connectivity. Finally, we train three multiscale convolutional neural networks (CNNs) and apply decision fusion to their predicted results to obtain the final classification decision. Our dataset comprises four classes of renal tissues: non-RCC renal parenchyma, non-RCC fat tissues, clear cell RCC (ccRCC), and clear cell papillary RCC (ccpRCC). The developed system demonstrates a high classification accuracy and sensitivity on the RCC biopsy samples at the slide level. Following a leave-one-subject-out cross-validation approach, the developed RCC subtype classification system achieves an overall classification accuracy of 93.0% ± 4.9%, a sensitivity of 91.3% ± 10.7%, and a high classification specificity of 95.6% ± 5.2%, in distinguishing ccRCC from ccpRCC or non-RCC tissues. Furthermore, our method outperformed the state-of-the-art Resnet-50 model. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Figure 1

14 pages, 423 KiB  
Article
Contrastive Self-Supervised Learning for Stress Detection from ECG Data
by Suha Rabbani and Naimul Khan
Bioengineering 2022, 9(8), 374; https://doi.org/10.3390/bioengineering9080374 - 08 Aug 2022
Cited by 8 | Viewed by 2189
Abstract
In recent literature, ECG-based stress assessment has become popular due to its proven correlation to stress and increased accessibility of ECG data through commodity hardware. However, most ECG-based stress assessment models use supervised learning, relying on manually-annotated data. Limited research is done in [...] Read more.
In recent literature, ECG-based stress assessment has become popular due to its proven correlation to stress and increased accessibility of ECG data through commodity hardware. However, most ECG-based stress assessment models use supervised learning, relying on manually-annotated data. Limited research is done in the area of self-supervised learning (SSL) approaches that leverage unlabelled data and none that utilize contrastive SSL. However, with the dominance of contrastive SSL in domains such as computer vision, it is essential to see if the same excellence in performance can be obtained on an ECG-based stress assessment dataset. In this paper, we propose a contrastive SSL model for stress assessment using ECG signals based on the SimCLR framework. We test our model on two ECG-based stress assessment datasets. We show that our proposed solution results in a 9% improvement in accuracy on the WESAD dataset and 3.7% on the RML dataset when compared with SOTA ECG-based SSL models for stress assessment. The development of more accurate stress assessment models, particularly those that employ non-invasive data such as ECG for assessment, leads to developments in wearable technology and the creation of better health monitoring applications in areas such as stress management and relaxation therapy. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Figure 1

20 pages, 2743 KiB  
Article
Analytical Studies of Antimicrobial Peptides as Diagnostic Biomarkers for the Detection of Bacterial and Viral Pneumonia
by Olalekan Olanrewaju Bakare, Arun Gokul and Marshall Keyster
Bioengineering 2022, 9(7), 305; https://doi.org/10.3390/bioengineering9070305 - 11 Jul 2022
Cited by 1 | Viewed by 1708
Abstract
Pneumonia remains one of the leading causes of infectious mortality and significant economic losses among our growing population. The lack of specific biomarkers for correct and timely diagnosis to detect patients’ status is a bane towards initiating a proper treatment plan for the [...] Read more.
Pneumonia remains one of the leading causes of infectious mortality and significant economic losses among our growing population. The lack of specific biomarkers for correct and timely diagnosis to detect patients’ status is a bane towards initiating a proper treatment plan for the disease; thus, current biomarkers cannot distinguish between pneumonia and other associated conditions such as atherosclerotic plaques and human immunodeficiency virus (HIV). Antimicrobial peptides (AMPs) are potential candidates for detecting numerous illnesses due to their compensatory roles as theranostic molecules. This research sought to generate specific data for parental AMPs to identify viral and bacterial pneumonia pathogens using in silico technology. The parental antimicrobial peptides (AMPs) used in this work were AMPs discovered in our previous in silico analyses using the HMMER algorithm, which were used to generate derivative (mutated) AMPs that would bind with greater affinity, in order to detect the bacterial and viral receptors using an in silico site-directed mutagenesis approach. These AMPs’ 3D structures were subsequently predicted and docked against receptor proteins. The result shows putative AMPs with the potential capacity to detect pneumonia caused by these pathogens through their binding precision with high sensitivity, accuracy, and specificity for possible use in point-of-care diagnosis. These peptides’ tendency to detect receptor proteins of viral and bacterial pneumonia with precision justifies their use for differential diagnostics, in an attempt to reduce the problems of indiscriminate overuse, toxicity due to the wrong prescription, bacterial resistance, and the scarcity and high cost of existing pneumonia antibiotics. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Figure 1

12 pages, 474 KiB  
Article
Machine Learning and Regression Analysis to Model the Length of Hospital Stay in Patients with Femur Fracture
by Carlo Ricciardi, Alfonso Maria Ponsiglione, Arianna Scala, Anna Borrelli, Mario Misasi, Gaetano Romano, Giuseppe Russo, Maria Triassi and Giovanni Improta
Bioengineering 2022, 9(4), 172; https://doi.org/10.3390/bioengineering9040172 - 14 Apr 2022
Cited by 23 | Viewed by 2812
Abstract
Fractures of the femur are a frequent problem in elderly people, and it has been demonstrated that treating them with a diagnostic–therapeutic–assistance path within 48 h of admission to the hospital reduces complications and shortens the length of the hospital stay (LOS). In [...] Read more.
Fractures of the femur are a frequent problem in elderly people, and it has been demonstrated that treating them with a diagnostic–therapeutic–assistance path within 48 h of admission to the hospital reduces complications and shortens the length of the hospital stay (LOS). In this paper, the preoperative data of 1082 patients were used to further extend the previous research and to generate several models that are capable of predicting the overall LOS: First, the LOS, measured in days, was predicted through a regression analysis; then, it was grouped by weeks and was predicted with a classification analysis. The KNIME analytics platform was applied to divide the dataset for a hold-out cross-validation, perform a multiple linear regression and implement machine learning algorithms. The best coefficient of determination (R2) was achieved by the support vector machine (R2 = 0.617), while the mean absolute error was similar for all the algorithms, ranging between 2.00 and 2.11 days. With regard to the classification analysis, all the algorithms surpassed 80% accuracy, and the most accurate algorithm was the radial basis function network, at 83.5%. The use of these techniques could be a valuable support tool for doctors to better manage orthopaedic departments and all their resources, which would reduce both waste and costs in the context of healthcare. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Figure 1

Review

Jump to: Research, Other

35 pages, 3264 KiB  
Review
The Role of Medical Image Modalities and AI in the Early Detection, Diagnosis and Grading of Retinal Diseases: A Survey
by Gehad A. Saleh, Nihal M. Batouty, Sayed Haggag, Ahmed Elnakib, Fahmi Khalifa, Fatma Taher, Mohamed Abdelazim Mohamed, Rania Farag, Harpal Sandhu, Ashraf Sewelam and Ayman El-Baz
Bioengineering 2022, 9(8), 366; https://doi.org/10.3390/bioengineering9080366 - 04 Aug 2022
Cited by 6 | Viewed by 2667
Abstract
Traditional dilated ophthalmoscopy can reveal diseases, such as age-related macular degeneration (AMD), diabetic retinopathy (DR), diabetic macular edema (DME), retinal tear, epiretinal membrane, macular hole, retinal detachment, retinitis pigmentosa, retinal vein occlusion (RVO), and retinal artery occlusion (RAO). Among these diseases, AMD and [...] Read more.
Traditional dilated ophthalmoscopy can reveal diseases, such as age-related macular degeneration (AMD), diabetic retinopathy (DR), diabetic macular edema (DME), retinal tear, epiretinal membrane, macular hole, retinal detachment, retinitis pigmentosa, retinal vein occlusion (RVO), and retinal artery occlusion (RAO). Among these diseases, AMD and DR are the major causes of progressive vision loss, while the latter is recognized as a world-wide epidemic. Advances in retinal imaging have improved the diagnosis and management of DR and AMD. In this review article, we focus on the variable imaging modalities for accurate diagnosis, early detection, and staging of both AMD and DR. In addition, the role of artificial intelligence (AI) in providing automated detection, diagnosis, and staging of these diseases will be surveyed. Furthermore, current works are summarized and discussed. Finally, projected future trends are outlined. The work done on this survey indicates the effective role of AI in the early detection, diagnosis, and staging of DR and/or AMD. In the future, more AI solutions will be presented that hold promise for clinical applications. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Figure 1

Other

Jump to: Research, Review

29 pages, 2818 KiB  
Systematic Review
Artificial Intelligence for Detecting and Quantifying Fatty Liver in Ultrasound Images: A Systematic Review
by Fahad Muflih Alshagathrh and Mowafa Said Househ
Bioengineering 2022, 9(12), 748; https://doi.org/10.3390/bioengineering9120748 - 01 Dec 2022
Cited by 12 | Viewed by 2696
Abstract
Background: Non-alcoholic Fatty Liver Disease (NAFLD) is growing more prevalent worldwide. Although non-invasive diagnostic approaches such as conventional ultrasonography and clinical scoring systems have been proposed as alternatives to liver biopsy, their efficacy has been called into doubt. Artificial Intelligence (AI) is now [...] Read more.
Background: Non-alcoholic Fatty Liver Disease (NAFLD) is growing more prevalent worldwide. Although non-invasive diagnostic approaches such as conventional ultrasonography and clinical scoring systems have been proposed as alternatives to liver biopsy, their efficacy has been called into doubt. Artificial Intelligence (AI) is now combined with traditional diagnostic processes to improve the performance of non-invasive approaches. Objective: This study explores how well various AI methods function and perform on ultrasound (US) images to diagnose and quantify non-alcoholic fatty liver disease. Methodology: A systematic review was conducted to achieve this objective. Five science bibliographic databases were searched, including PubMed, Association for Computing Machinery ACM Digital Library, Institute of Electrical and Electronics Engineers IEEE Xplore, Scopus, and Google Scholar. Only peer-reviewed English articles, conferences, theses, and book chapters were included. Data from studies were synthesized using narrative methodologies per Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) criteria. Results: Forty-nine studies were included in the systematic review. According to the qualitative analysis, AI significantly enhanced the diagnosis of NAFLD, Non-Alcoholic Steatohepatitis (NASH), and liver fibrosis. In addition, modalities, image acquisition, feature extraction and selection, data management, and classifiers were assessed and compared in terms of performance measures (i.e., accuracy, sensitivity, and specificity). Conclusion: AI-supported systems show potential performance increases in detecting and quantifying steatosis, NASH, and liver fibrosis in NAFLD patients. Before real-world implementation, prospective studies with direct comparisons of AI-assisted modalities and conventional techniques are necessary. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
Show Figures

Graphical abstract

Back to TopTop