Skip to main content
Fachgebiete
Automobil + Motoren Bauwesen + Immobilien Business IT + Informatik Elektrotechnik + Elektronik Energie + Nachhaltigkeit Finance + Banking Management + Führung Marketing + Vertrieb Maschinenbau + Werkstoffe Versicherung + Risiko
DE
EN
Bücher
Zeitschriften
Themenseiten
Organisationspsychologie Projektmanagement Marketing Smart Manufacturing
Weitere Formate
Podcasts Webinare Technik Webinare Wirtschaft Kongresse Veranstaltungskalender Awards
Jetzt Einzelzugang starten
Zugang für Unternehmen
Referenzkunden
SLX-Digitalkonferenz: Zukunftswerkstatt 2024

Mehr Umsatz mit Top-Kontakten

Am 7. Mai erfahren Sie, wie Vertriebsteams Leadgenerierung, Leadnurturing und Leadstrategien effizient steuern können.
Erweiterte Suche
Anmelden
nach oben

2019 | Buch

Introduction Kapitel lesen Erstes Kapitel lesen

Applying Machine Learning for Automated Classification of Biomedical Data in Subject-Independent Settings

verfasst von: Dr. Thuy T. Pham

Verlag: Springer International Publishing

Buchreihe : Springer Theses

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten
insite
SUCHEN

Über dieses Buch

This book describes efforts to improve subject-independent automated classification techniques using a better feature extraction method and a more efficient model of classification. It evaluates three popular saliency criteria for feature selection, showing that they share common limitations, including time-consuming and subjective manual de-facto standard practice, and that existing automated efforts have been predominantly used for subject dependent setting. It then proposes a novel approach for anomaly detection, demonstrating its effectiveness and accuracy for automated classification of biomedical data, and arguing its applicability to a wider range of unsupervised machine learning applications in subject-independent settings.

Inhaltsverzeichnis

Frontmatter
Chapter 1. Introduction
Abstract
This work is aimed to demonstrate the contribution of feature learning in classification applications, especially for biomedical data. Unsupervised learning and subject-independent settings are desirable deployment manners for the area. One reason is that these abilities can help deploy classification tasks in out-of-the-lab wearable devices. Another reason is reducing labour costs and subjectivity associated with human involvement. In this thesis, three examples are studied: human body movement assessment where acceleration data is used (Case 1), respiratory artefact removal where lung function tests are carried out (Case 2), and spike sorting for electrophysiological data (Case 3). Manual classification is often considered the de facto standard practice but it is time-consuming and subjective. Existing automated efforts have been predominantly designed for subject-dependent settings. Unsupervised sorters using simple statistics have only yielded modestly accurate results.
Thuy T. Pham
Chapter 2. Background
Abstract
In this chapter, relevant machine learning background is provided. The first section is about classification algorithms, then performance metrics for studied applications are introduced, followed by related techniques that are used to select salient features in the applications. Finally, automatic feature selection models are reviewed. In this thesis, the term classification is mentioned to a task that predicts a class label of an instance regardless of prediction model algorithms. The terms sorting and clustering are used in the third case of application (Chap. 6) are used in the context of unknown class number situation.
Thuy T. Pham
Chapter 3. Algorithms
Abstract
In Chap. 2, general technical background of automated classification and feature engineering concepts were discussed. In this chapter, more detailed algorithms of the proposed work are presented. Specifically, three scenarios of automated classification are introduced in the context of a desirable subject-independent settings. There are two separate sections for two functional blocks for a classification task: feature engineering and sorting algorithms. The first block is a common scheme for several scenarios, on the other hand the second block is specific to each application scenario. The first section describes a voting-based feature (data representation) selection process and its criteria. Then the next parts formulate the problem for three classification scenarios: point anomaly detection (Sect. 3.2.1), collective anomaly detection (Sect. 3.2.2), and unsupervised multi-class sorting (Sect. 3.2.3).
Thuy T. Pham
Chapter 4. Point Anomaly Detection: Application to Freezing of Gait Monitoring
Abstract
This chapter describes details of settings and experiment reports for a point anomaly detection application. First, a dataset of freezing of gait (FoG) in patients with advanced Parkinson’s disease is explained. Then, specific classification settings for FoG detection are provided. Final sections are results of feature ranking and performance comparisons with existing methods.
Thuy T. Pham
Chapter 5. Collective Anomaly Detection: Application to Respiratory Artefact Removals
Abstract
In this chapter, data sets and feature learning result observations in respiratory artefact removal for lung function tests, specifically the forced oscillation technique (FOT) are presented. The first section introduces the FOT method and respiratory artefacts. The next two sections describe our FOT data sets and performance metrics used to evaluate the proposed scheme in Sect. 5. Section 6 discusses feature selection for FOT data, and two different models for artefact detectors are presented in Sects. 7 and 8. The last four sections reports results/discussion of feature ranking and performance comparison between our proposed detectors and existing methods.
Thuy T. Pham
Chapter 6. Spike Sorting: Application to Motor Unit Action Potential Discrimination
Abstract
While in Chaps. 4 and 5 two-class discrimination applications are demonstrated that the proposed feature engineering play an important part in improving the accuracy performance results, this chapter illustrates the contribution of the feature learning scheme in a classification problem where some class information (e.g., number of classes) is not predefined. For example, in application of motor unit action potential (MUAP) sorting for intramuscular electromyography (nEMG) data, called nEMG spike sorting.
Thuy T. Pham
Chapter 7. Conclusion
Abstract
This chapter summarises the proposed feature engineering method for classification applications (Chap. 3) and the main observations in several experiments presented in Chap. 4, 5, 6. Discussions also include limitations and future works for each scenario. In general, the contribution of a systematic application of feature engineering to accuracy performance is shown in all three cases of real-life biomedical data classification.
Thuy T. Pham
Backmatter
Metadaten
Titel
Applying Machine Learning for Automated Classification of Biomedical Data in Subject-Independent Settings
verfasst von
Dr. Thuy T. Pham
Copyright-Jahr
2019
Electronic ISBN
978-3-319-98675-3
Print ISBN
978-3-319-98674-6
DOI
https://doi.org/10.1007/978-3-319-98675-3

Neuer Inhalt

    Bildnachweise