Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
SHAMSUL: Systematic Holistic Analysis to investigate Medical Significance Utilizing Local interpretability methods in deep learning for chest radiography pathology prediction
Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.ORCID iD: 0000-0002-1101-3793
Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.ORCID iD: 0000-0002-1912-712x
Skatturinn (Iceland Revenue and Customs), Iceland.
Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.ORCID iD: 0000-0001-5924-5457
2023 (English)In: Nordic Machine Intelligence, E-ISSN 2703-9196, Vol. 3, no 1, p. 27-47Article in journal (Refereed) Published
Abstract [en]

The interpretability of deep neural networks has become a subject of great interest within the medical and healthcare domain. This attention stems from concerns regarding transparency, legal and ethical considerations, and the medical significance of predictions generated by these deep neural networks in clinical decision support systems. To address this matter, our study delves into the application of four well-established interpretability methods: Local Interpretable Model-agnostic Explanations (LIME), Shapley Additive exPlanations (SHAP), Gradient-weighted Class Activation Mapping (Grad-CAM), and Layer-wise Relevance Propagation (LRP). Leveraging the approach of transfer learning with a multi-label-multi-class chest radiography dataset, we aim to interpret predictions pertaining to specific pathology classes. Our analysis encompasses both single-label and multi-label predictions, providing a comprehensive and unbiased assessment through quantitative and qualitative investigations, which are compared against human expert annotation. Notably, Grad-CAM demonstrates the most favorable performance in quantitative evaluation, while the LIME heatmap score segmentation visualization exhibits the highest level of medical significance. Our research underscores both the outcomes and the challenges faced in the holistic approach adopted for assessing these interpretability methods and suggests that a multimodal-based approach, incorporating diverse sources of information beyond chest radiography images, could offer additional insights for enhancing interpretability in the medical domain.

Place, publisher, year, edition, pages
2023. Vol. 3, no 1, p. 27-47
Keywords [en]
Deep Learning, Interpretability Methods, LIME, SHAP, Grad-CAM, LRP, Chest X-ray, Heatmap Score Visualization, Clinical Decision Support System
National Category
Computer Sciences
Research subject
Computer and Systems Sciences
Identifiers
URN: urn:nbn:se:su:diva-223825DOI: 10.5617/nmi.10471OAI: oai:DiVA.org:su-223825DiVA, id: diva2:1813030
Available from: 2023-11-17 Created: 2023-11-17 Last updated: 2025-02-27Bibliographically approved
In thesis
1. Advancing Clinical Decision Support Using Machine Learning & the Internet of Medical Things: Enhancing COVID-19 & Early Sepsis Detection
Open this publication in new window or tab >>Advancing Clinical Decision Support Using Machine Learning & the Internet of Medical Things: Enhancing COVID-19 & Early Sepsis Detection
2024 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

This thesis presents a critical examination of the positive impact of Machine Learning (ML) and the Internet of Medical Things (IoMT) for advancing the Clinical Decision Support System (CDSS) in the context of COVID-19 and early sepsis detection.

It emphasizes the transition towards patient-centric healthcare systems, which necessitate personalized and participatory care—a transition that could be facilitated by these emerging fields. The thesis accentuates how IoMT could serve as a robust platform for data aggregation, analysis, and transmission, which could empower healthcare providers to deliver more effective care. The COVID-19 pandemic has particularly stressed the importance of such patient-centric systems for remote patient monitoring and disease management.

The integration of ML-driven CDSSs with IoMT is viewed as an extremely important step in healthcare systems that could offer real-time decision-making support and enhance patient health outcomes. The thesis investigates ML's capability to analyze complex medical datasets, identify patterns and correlations, and adapt to changing conditions, thereby enhancing its predictive capabilities. It specifically focuses on the development of IoMT-based CDSSs for COVID-19 and early sepsis detection, using advanced ML methods and medical data.

Key issues addressed cover data annotation scarcity, data sparsity, and data heterogeneity, along with the aspects of security, privacy, and accessibility. The thesis also intends to enhance the interpretability of ML prediction model-based CDSSs. Ethical considerations are prioritized to ensure adherence to the highest standards.

The thesis demonstrates the potential and efficacy of combining ML with IoMT to enhance CDSSs by emphasizing the importance of model interpretability, system compatibility, and the integration of multimodal medical data for an effective CDSS.

Overall, this thesis makes a significant contribution to the fields of ML and IoMT in healthcare, featuring their combined potential to enhance CDSSs, particularly in the areas of COVID-19 and early sepsis detection.

The thesis hopes to enhance understanding among medical stakeholders and acknowledges the need for continuous development in this sector.

Abstract [sv]

Denna avhandling presenterar en kritisk granskning av den positiva effekten av maskininlärning (ML) och Internet of Medical Things (IoMT) för att främja det kliniska beslutsstödsystemet (CDSS) kopplat till covid-19 och tidig upptäckt av sepsis.

Avhandlingen betonar övergången mot patientcentrerade vårdsystem som kräver personlig och deltagande vård – en övergång som skulle kunna underlättas av dessa framväxande områden. Studien visar hur IoMT kan fungera som en robust plattform för dataaggregering, analys och överföring, vilket kan ge vårdgivare möjlighet att erbjuda mer effektiv vård. Covid-19-pandemin har särskilt betonat vikten av sådana patientcentrerade system för fjärrövervakning av patienter och sjukdomshantering.

Integreringen av ML-drivna CDSS med IoMT ses som ett extremt viktigt steg i vårdsystemen som kan erbjuda stöd för beslutsfattande i realtid och förbättra patienternas hälsoutfall. Avhandlingen undersöker maskininlärningens förmåga att analysera komplexa medicinska dataset, identifiera mönster och korrelationer, samt göra anpassningar till föränderliga förhållanden, vilket därmed förbättrar dess prediktiva förmågor. Den fokuserar specifikt på utvecklingen av IoMT-baserade CDSS för covid-19 och tidig upptäckt av sepsis, med användning av avancerade ML-metoder och medicinska data.

Nyckelfrågor som adresseras täcker bristen på dataannotering, dataspridning och dataheterogenitet, tillsammans med aspekter av säkerhet, integritet och tillgänglighet. Avhandlingen avser också att förbättra tolkbarheten av ML-prediktionsmodellbaserade CDSS. Etiska överväganden prioriteras för att säkerställa efterlevnad av de högsta standarderna.

Avhandlingen demonstrerar potentialen och effektiviteten i att kombinera ML med IoMT för att förbättra CDSS genom att betona vikten av modelltolkbarhet, systemkompatibilitet och integrering av multimodala medicinska data för ett effektivt CDSS.

Sammantaget bidrar denna avhandling till områdena ML och IoMT inom hälsovården, med deras kombinerade potential att förbättra CDSS, särskilt inom områdena covid-19 och tidig upptäckt av sepsis.

Förhoppningen är att avhandlingen ska förbättra förståelsen bland medicinska intressenter och understryka behovet av kontinuerlig utveckling inom denna sektor.

Place, publisher, year, edition, pages
Stockholm: Department of Computer and Systems Sciences, Stockholm University, 2024. p. 85
Series
Report Series / Department of Computer & Systems Sciences, ISSN 1101-8526 ; 24-004
Keywords
Internet of Medical Things, Patient-Centric Healthcare, Clinical Decision Support System, Predictive Modeling in Healthcare, Health Informatics, Healthcare analytics, COVID-19, Sepsis, COVID-19 Detection, Early Sepsis Detection, Lung Segmentation Detection, Medical Data Annotation Scarcity, Medical Data Sparsity, Medical Data Heterogeneity, Medical Data Security & Privacy, Practical Usability Enhancement, Low-End Device Adaptability, Medical Significance, Interpretability, Visualization, LIME, SHAP, Grad-CAM, LRP, Electronic Health Records, Thermal Image, Tabular Medical Data, Chest X-ray, Machine Learning, Deep Learning, Federated Learning, Semi-Supervised Machine Learning, Multi-Task Learning, Transfer Learning, Multi-Modality, Natural Language Processing, ClinicalBERT, GAN
National Category
Computer Sciences
Research subject
Computer and Systems Sciences
Identifiers
urn:nbn:se:su:diva-226027 (URN)978-91-8014-655-5 (ISBN)978-91-8014-656-2 (ISBN)
Public defence
2024-03-28, Lilla Hörsalen, NOD-huset, Borgarfjordsgatan 12, Kista, 13:00 (English)
Opponent
Supervisors
Available from: 2024-03-05 Created: 2024-01-31 Last updated: 2024-02-26Bibliographically approved

Open Access in DiVA

SHAMSUL(20813 kB)597 downloads
File information
File name FULLTEXT01.pdfFile size 20813 kBChecksum SHA-512
5ba37916e24b670cd882606219f76e2e7496da9e7ab3da3f92e0a9dbc64ae05bb76b938b1c50bfee4c5c7581ca7eaaeae742afe6ab9bbc5b5253bb00f1bfb40b
Type fulltextMimetype application/pdf

Other links

Publisher's full text

Authority records

Alam, Mahbub UlHollmén, JaakkoRahmani, Rahim

Search in DiVA

By author/editor
Alam, Mahbub UlHollmén, JaakkoRahmani, Rahim
By organisation
Department of Computer and Systems Sciences
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 597 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 2117 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf