Special Issue "Machine Learning in Healthcare and Biomedical Application"

A special issue of Algorithms (ISSN 1999-4893). This special issue belongs to the section "Algorithms for Multidisciplinary Applications".

Deadline for manuscript submissions: closed (31 March 2021).

Special Issue Editor

Dr. Alessia Sarica
E-Mail Website
Guest Editor
Neuroscience Research Center, Department of Medical and Surgical Sciences, “Magna Graecia” University, Italy
Interests: machine learning; decision trees; ensemble learning; computer-aided diagnosis; neurosciences; neuroimaging

Special Issue Information

Dear Colleagues,

I invite you to submit original scientific contributions on the topic “Machine Learning in Healthcare and Biomedical Application”. The huge advances in Machine Learning (ML) have increased the opportunity to improve and speed up clinical decisions in numerous biomedical fields. The present Special Issue focuses on the revolutionary changes to medicine brought about by ML. In particular, we aim to present the most recent discoveries on the application of new or state-of-the-art ML algorithms (e.g., supervised and unsupervised learning; feature selection, extraction and reduction; ensemble learning; deep learning; interpretability and explainability of ML models) in areas related, but not limited to

  • Disease identification, differential diagnosis, and prognosis;
  • Bioimage processing and analysis;
  • Emotion recognition in healthcare;
  • Cognitive and psychological profiling;
  • Epidemic outbreak prediction;
  • Personalized medicine.

Contributions such as systematic reviews or meta-analyses on the above-mentioned topics are also welcome.

Dr. Alessia Sarica
Guest Editor

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 papers will be 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. Algorithms 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 1400 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

  • machine learning for healthcare
  • computer-aided diagnosis
  • automatic clinical decision
  • epidemic outbreak prediction algorithms
  • personalized medicine

Published Papers (5 papers)

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

Research

Open AccessArticle
A Deep Learning Model for Data-Driven Discovery of Functional Connectivity
Algorithms 2021, 14(3), 75; https://0-doi-org.brum.beds.ac.uk/10.3390/a14030075 - 26 Feb 2021
Viewed by 696
Abstract
Functional connectivity (FC) studies have demonstrated the overarching value of studying the brain and its disorders through the undirected weighted graph of functional magnetic resonance imaging (fMRI) correlation matrix. However, most of the work with the FC depends on the way the connectivity [...] Read more.
Functional connectivity (FC) studies have demonstrated the overarching value of studying the brain and its disorders through the undirected weighted graph of functional magnetic resonance imaging (fMRI) correlation matrix. However, most of the work with the FC depends on the way the connectivity is computed, and it further depends on the manual post-hoc analysis of the FC matrices. In this work, we propose a deep learning architecture BrainGNN that learns the connectivity structure as part of learning to classify subjects. It simultaneously applies a graphical neural network to this learned graph and learns to select a sparse subset of brain regions important to the prediction task. We demonstrate that the model’s state-of-the-art classification performance on a schizophrenia fMRI dataset and demonstrate how introspection leads to disorder relevant findings. The graphs that are learned by the model exhibit strong class discrimination and the sparse subset of relevant regions are consistent with the schizophrenia literature. Full article
(This article belongs to the Special Issue Machine Learning in Healthcare and Biomedical Application)
Show Figures

Figure 1

Open AccessArticle
A Novel Approach for Cognitive Clustering of Parkinsonisms through Affinity Propagation
Algorithms 2021, 14(2), 49; https://0-doi-org.brum.beds.ac.uk/10.3390/a14020049 - 04 Feb 2021
Viewed by 753
Abstract
Cluster analysis is widely applied in the neuropsychological field for exploring patterns in cognitive profiles, but traditional hierarchical and non-hierarchical approaches could be often poorly effective or even inapplicable on certain type of data. Moreover, these traditional approaches need the initial specification of [...] Read more.
Cluster analysis is widely applied in the neuropsychological field for exploring patterns in cognitive profiles, but traditional hierarchical and non-hierarchical approaches could be often poorly effective or even inapplicable on certain type of data. Moreover, these traditional approaches need the initial specification of the number of clusters, based on a priori knowledge not always owned. For this reason, we proposed a novel method for cognitive clustering through the affinity propagation (AP) algorithm. In particular, we applied the AP clustering on the regression residuals of the Mini Mental State Examination scores—a commonly used screening tool for cognitive impairment—of a cohort of 49 Parkinson’s disease, 48 Progressive Supranuclear Palsy and 44 healthy control participants. We found four clusters, where two clusters (68 and 30 participants) showed almost intact cognitive performance, one cluster had a moderate cognitive impairment (34 participants), and the last cluster had a more extensive cognitive deficit (8 participants). The findings showed, for the first time, an intra- and inter-diagnostic heterogeneity in the cognitive profile of Parkinsonisms patients. Our novel method of unsupervised learning could represent a reliable tool for supporting the neuropsychologists in understanding the natural structure of the cognitive performance in the neurodegenerative diseases. Full article
(This article belongs to the Special Issue Machine Learning in Healthcare and Biomedical Application)
Show Figures

Figure 1

Open AccessArticle
Mobile-Aware Deep Learning Algorithms for Malaria Parasites and White Blood Cells Localization in Thick Blood Smears
Algorithms 2021, 14(1), 17; https://0-doi-org.brum.beds.ac.uk/10.3390/a14010017 - 11 Jan 2021
Viewed by 603
Abstract
Effective determination of malaria parasitemia is paramount in aiding clinicians to accurately estimate the severity of malaria and guide the response for quality treatment. Microscopy by thick smear blood films is the conventional method for malaria parasitemia determination. Despite its edge over other [...] Read more.
Effective determination of malaria parasitemia is paramount in aiding clinicians to accurately estimate the severity of malaria and guide the response for quality treatment. Microscopy by thick smear blood films is the conventional method for malaria parasitemia determination. Despite its edge over other existing methods of malaria parasitemia determination, it has been critiqued for being laborious, time consuming and equally requires expert knowledge for an efficient manual quantification of the parasitemia. This pauses a big challenge to most low developing countries as they are not only highly endemic but equally low resourced in terms of technical personnel in medical laboratories This study presents an end-to-end deep learning approach to automate the localization and count of P.falciparum parasites and White Blood Cells (WBCs) for effective parasitemia determination. The method involved building computer vision models on a dataset of annotated thick blood smear images. These computer vision models were built based on pre-trained deep learning models including Faster Regional Convolutional Neural Network (Faster R-CNN) and Single Shot Multibox Detector (SSD) models that help process the obtained digital images. To improve model performance due to a limited dataset, data augmentation was applied. Results from the evaluation of our approach showed that it reliably detected and returned a count of parasites and WBCs with good precision and recall. A strong correlation was observed between our model-generated counts and the manual counts done by microscopy experts (posting a spear man correlation of ρ = 0.998 for parasites and ρ = 0.987 for WBCs). Additionally, our proposed SSD model was quantized and deployed on a mobile smartphone-based inference app to detect malaria parasites and WBCs in situ. Our proposed method can be applied to support malaria diagnostics in settings with few trained Microscopy Experts yet constrained with large volume of patients to diagnose. Full article
(This article belongs to the Special Issue Machine Learning in Healthcare and Biomedical Application)
Show Figures

Figure 1

Open AccessArticle
On a Controlled Se(Is)(Ih)(Iicu)AR Epidemic Model with Output Controllability Issues to Satisfy Hospital Constraints on Hospitalized Patients
Algorithms 2020, 13(12), 322; https://0-doi-org.brum.beds.ac.uk/10.3390/a13120322 - 03 Dec 2020
Cited by 1 | Viewed by 557
Abstract
An epidemic model, the so-called SE(Is)(Ih)(Iicu)AR epidemic model, is proposed which splits the infectious subpopulation of the classical SEIR (Susceptible-Exposed-Infectious-Recovered) model into four subpopulations, namely asymptomatic infectious and three categories of symptomatic infectious, namely slight infectious, non-intensive care infectious, and intensive care hospitalized [...] Read more.
An epidemic model, the so-called SE(Is)(Ih)(Iicu)AR epidemic model, is proposed which splits the infectious subpopulation of the classical SEIR (Susceptible-Exposed-Infectious-Recovered) model into four subpopulations, namely asymptomatic infectious and three categories of symptomatic infectious, namely slight infectious, non-intensive care infectious, and intensive care hospitalized infectious. The exposed subpopulation has four different transitions to each one of the four kinds of infectious subpopulations governed under eventually different proportionality parameters. The performed research relies on the problem of satisfying prescribed hospitalization constraints related to the number of patients via control interventions. There are four potential available controls which can be manipulated, namely the vaccination of the susceptible individuals, the treatment of the non-intensive care unit hospitalized patients, the treatment of the hospitalized patients at the intensive care unit, and the transmission rate which can be eventually updated via public interventions such as isolation of the infectious, rules of groups meetings, use of face masks, decrees of partial or total quarantines, and others. The patients staying at the non-intensive care unit and those staying at the intensive care unit are eventually, but not necessarily, managed as two different hospitalized subpopulations. The controls are designed based on output controllability issues in the sense that the levels of hospital admissions are constrained via prescribed maximum levels and the measurable outputs are defined by the hospitalized patients either under a joint consideration of the sum of both subpopulations or separately. In this second case, it is possible to target any of the two hospitalized subpopulations only or both of them considered as two different components of the output. Different algorithms are given to design the controls which guarantee, if possible, that the prescribed hospitalization constraints hold. If this were not possible, because the levels of serious infection are too high according to the hospital availability means, then the constraints are revised and modified accordingly so that the amended ones could be satisfied by a set of controls. The algorithms are tested through numerically worked examples under disease parameterizations of COVID-19. Full article
(This article belongs to the Special Issue Machine Learning in Healthcare and Biomedical Application)
Show Figures

Figure 1

Open AccessArticle
Lung Lobe Segmentation Based on Lung Fissure Surface Classification Using a Point Cloud Region Growing Approach
Algorithms 2020, 13(10), 263; https://0-doi-org.brum.beds.ac.uk/10.3390/a13100263 - 15 Oct 2020
Viewed by 713
Abstract
In anatomy, the lung can be divided by lung fissures into several pulmonary lobe units with specific functions. Identifying the lung lobes and the distribution of various diseases among different lung lobes from CT images is important for disease diagnosis and tracking after [...] Read more.
In anatomy, the lung can be divided by lung fissures into several pulmonary lobe units with specific functions. Identifying the lung lobes and the distribution of various diseases among different lung lobes from CT images is important for disease diagnosis and tracking after recovery. In order to solve the problems of low tubular structure segmentation accuracy and long algorithm time in segmenting lung lobes based on lung anatomical structure information, we propose a segmentation algorithm based on lung fissure surface classification using a point cloud region growing approach. We cluster the pulmonary fissures, transformed into point cloud data, according to the differences in the pulmonary fissure surface normal vector and curvature estimated by principal component analysis. Then, a multistage spline surface fitting method is used to fill and expand the lung fissure surface to realize the lung lobe segmentation. The proposed approach was qualitatively and quantitatively evaluated on a public dataset from Lobe and Lung Analysis 2011 (LOLA11), and obtained an overall score of 0.84. Although our approach achieved a slightly lower overall score compared to the deep learning based methods (LobeNet_V2 and V-net), the inter-lobe boundaries from our approach were more accurate for the CT images with visible lung fissures. Full article
(This article belongs to the Special Issue Machine Learning in Healthcare and Biomedical Application)
Show Figures

Figure 1

Back to TopTop