{"id":{"repo_id":"tu-berlin","oai_identifier":"oai:depositonce.tu-berlin.de:11303/24202"},"canonical_url":"https://search.dev.ndltd.org/etd/tu-berlin/oai:depositonce.tu-berlin.de:11303/24202","repository":{"repo_id":"tu-berlin","name":"Technische Universität Berlin","base_url":"https://api-depositonce.tu-berlin.de/server/oai/request"},"display":{"title":"Exploring the lifecycle of artificial intelligence – deep learning for 1D to 4D biomedical data","abstract":"This thesis, titled \"Exploring the Lifecycle of Artificial Intelligence - Deep Learning for 1D to 4D Biomedical Data\" presents a comprehensive exploration of the Artificial Intelligence (AI) lifecycle as applied to various domains within the life sciences and medicine, specifically focusing on the application of Machine Learning (ML) and Deep Learning (DL) techniques to data ranging from one to four dimensional data. The study is structured around the AI lifecycle, which is elaborated upon in Chapter 2. This lifecycle encompasses the critical stages of problem definition, data acquisition, model training, evaluation, interpretation, and deployment, offering a holistic approach to the development and implementation of AI systems in medical contexts. A significant portion of the thesis delves into the interpretation of Electrocardiography (ECG) data, as detailed in Chapter 3. This chapter introduces the PTB-XL database, a large-scale dataset for ECG analysis, and discusses the application of eXplainable Artificial Intelligence techniques to enhance model transparency and trustworthiness. The study demonstrates how XAI can be employed to make AI models more interpretable and clinically relevant, which is crucial for their adoption in medical diagnostics. The thesis then shifts focus to the segmentation of 2D histological images, which is critical for accurate medical diagnostics, particularly in dermatology. As described in Chapter 4, this work explores advanced semantic segmentation techniques to identify and classify different biological structures within a Whole Slide Image (WSI). The chapter highlights the challenges and potential solutions in implementing DL models for this purpose, emphasizing the importance of precise segmentation in improving diagnostic accuracy. Further expanding on the dimensional complexity, Chapter 5 addresses the analysis of 3D neutrophil cell nuclei shapes using confocal fluorescence microscopy. This chapter provides insights into the cellular morphology, employing feature extraction, dimensionality reduction, and clustering methods to understand the variations in neutrophil shapes. The analysis underscores the potential of AI in uncovering morphological features that are not easily discernible through traditional methods. The final application discussed in this thesis involves the analysis of 4D lymphocyte dynamics in human lymphoid tissue, as outlined in Chapter 6. This chapter examines the temporal dynamics of lymphocytes using time-lapse confocal microscopy, demonstrating how AI can be utilized to gain novel insights into cellular behavior over time. The work presented here highlights the significant contributions of AI to understanding complex biological processes, potentially leading to new discoveries in immunology. In conclusion, this thesis not only advances the state-of-the-art in the application of AI to the life sciences but also provides a robust framework for the systematic development and deployment of AI models in medical research. By adopting the AI lifecycle approach, the research aims to address each stage, from problem definition to deployment, in a comprehensive manner, with the goal of developing AI solutions that are not only technically sound but also have potential for clinical applicability. The findings of this thesis suggest potential implications for the future of AI in medicine, particularly in enhancing the interpretability, transparency, and trustworthiness of AI models, which are essential for their integration into clinical practice.","abstract_html":"This thesis, titled "Exploring the Lifecycle of Artificial Intelligence - Deep Learning for 1D to 4D Biomedical Data" presents a comprehensive exploration of the Artificial Intelligence (AI) lifecycle as applied to various domains within the life sciences and medicine, specifically focusing on the application of Machine Learning (ML) and Deep Learning (DL) techniques to data ranging from one to four dimensional data. The study is structured around the AI lifecycle, which is elaborated upon in Chapter 2. This lifecycle encompasses the critical stages of problem definition, data acquisition, model training, evaluation, interpretation, and deployment, offering a holistic approach to the development and implementation of AI systems in medical contexts. A significant portion of the thesis delves into the interpretation of Electrocardiography (ECG) data, as detailed in Chapter 3. This chapter introduces the PTB-XL database, a large-scale dataset for ECG analysis, and discusses the application of eXplainable Artificial Intelligence techniques to enhance model transparency and trustworthiness. The study demonstrates how XAI can be employed to make AI models more interpretable and clinically relevant, which is crucial for their adoption in medical diagnostics. The thesis then shifts focus to the segmentation of 2D histological images, which is critical for accurate medical diagnostics, particularly in dermatology. As described in Chapter 4, this work explores advanced semantic segmentation techniques to identify and classify different biological structures within a Whole Slide Image (WSI). The chapter highlights the challenges and potential solutions in implementing DL models for this purpose, emphasizing the importance of precise segmentation in improving diagnostic accuracy. Further expanding on the dimensional complexity, Chapter 5 addresses the analysis of 3D neutrophil cell nuclei shapes using confocal fluorescence microscopy. This chapter provides insights into the cellular morphology, employing feature extraction, dimensionality reduction, and clustering methods to understand the variations in neutrophil shapes. The analysis underscores the potential of AI in uncovering morphological features that are not easily discernible through traditional methods. The final application discussed in this thesis involves the analysis of 4D lymphocyte dynamics in human lymphoid tissue, as outlined in Chapter 6. This chapter examines the temporal dynamics of lymphocytes using time-lapse confocal microscopy, demonstrating how AI can be utilized to gain novel insights into cellular behavior over time. The work presented here highlights the significant contributions of AI to understanding complex biological processes, potentially leading to new discoveries in immunology. In conclusion, this thesis not only advances the state-of-the-art in the application of AI to the life sciences but also provides a robust framework for the systematic development and deployment of AI models in medical research. By adopting the AI lifecycle approach, the research aims to address each stage, from problem definition to deployment, in a comprehensive manner, with the goal of developing AI solutions that are not only technically sound but also have potential for clinical applicability. The findings of this thesis suggest potential implications for the future of AI in medicine, particularly in enhancing the interpretability, transparency, and trustworthiness of AI models, which are essential for their integration into clinical practice.","abstract_has_math":false,"creators":["Wagner, Patrick"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Samek, Wojciech"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025","date_published":"2025","updated_at":"2026-07-27T21:28:49Z","subjects":[],"languages":["en"],"rights":[],"rights_urls":["https://creativecommons.org/licenses/by-sa/4.0/"],"identifier_entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://doi.org/10.14279/depositonce-23016"],"render_values":[{"text":"https://doi.org/10.14279/depositonce-23016","href":"https://doi.org/10.14279/depositonce-23016","code":true}]}]},"links":{"outbound_url":"https://depositonce.tu-berlin.de/handle/11303/24202","outbound_label":"Repository record","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Samek, Wojciech"]},{"key":"dc:creator","label":"Author","values":["Wagner, Patrick"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-04-10T10:18:56Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2025-04-10T10:18:56Z"]},{"key":"dc:date.issued","label":"Date","values":["2025"]},{"key":"dc:type","label":"Dc Type","values":["Doctoral Thesis"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://creativecommons.org/licenses/by-sa/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://depositonce.tu-berlin.de/handle/11303/24202","https://doi.org/10.14279/depositonce-23016"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This thesis, titled \"Exploring the Lifecycle of Artificial Intelligence - Deep Learning for 1D to 4D Biomedical Data\" presents a comprehensive exploration of the Artificial Intelligence (AI) lifecycle as applied to various domains within the life sciences and medicine, specifically focusing on the application of Machine Learning (ML) and Deep Learning (DL) techniques to data ranging from one to four dimensional data. The study is structured around the AI lifecycle, which is elaborated upon in Chapter 2. This lifecycle encompasses the critical stages of problem definition, data acquisition, model training, evaluation, interpretation, and deployment, offering a holistic approach to the development and implementation of AI systems in medical contexts. A significant portion of the thesis delves into the interpretation of Electrocardiography (ECG) data, as detailed in Chapter 3. This chapter introduces the PTB-XL database, a large-scale dataset for ECG analysis, and discusses the application of eXplainable Artificial Intelligence techniques to enhance model transparency and trustworthiness. The study demonstrates how XAI can be employed to make AI models more interpretable and clinically relevant, which is crucial for their adoption in medical diagnostics. The thesis then shifts focus to the segmentation of 2D histological images, which is critical for accurate medical diagnostics, particularly in dermatology. As described in Chapter 4, this work explores advanced semantic segmentation techniques to identify and classify different biological structures within a Whole Slide Image (WSI). The chapter highlights the challenges and potential solutions in implementing DL models for this purpose, emphasizing the importance of precise segmentation in improving diagnostic accuracy. Further expanding on the dimensional complexity, Chapter 5 addresses the analysis of 3D neutrophil cell nuclei shapes using confocal fluorescence microscopy. This chapter provides insights into the cellular morphology, employing feature extraction, dimensionality reduction, and clustering methods to understand the variations in neutrophil shapes. The analysis underscores the potential of AI in uncovering morphological features that are not easily discernible through traditional methods. The final application discussed in this thesis involves the analysis of 4D lymphocyte dynamics in human lymphoid tissue, as outlined in Chapter 6. This chapter examines the temporal dynamics of lymphocytes using time-lapse confocal microscopy, demonstrating how AI can be utilized to gain novel insights into cellular behavior over time. The work presented here highlights the significant contributions of AI to understanding complex biological processes, potentially leading to new discoveries in immunology. In conclusion, this thesis not only advances the state-of-the-art in the application of AI to the life sciences but also provides a robust framework for the systematic development and deployment of AI models in medical research. By adopting the AI lifecycle approach, the research aims to address each stage, from problem definition to deployment, in a comprehensive manner, with the goal of developing AI solutions that are not only technically sound but also have potential for clinical applicability. The findings of this thesis suggest potential implications for the future of AI in medicine, particularly in enhancing the interpretability, transparency, and trustworthiness of AI models, which are essential for their integration into clinical practice.","Diese Dissertation mit dem Titel \"Exploring the Lifecycle of Artificial Intelligence - Deep Learning for 1D to 4D Biomedical Data\" (zu deutsch: \"Erforschung des Lebenszyklus der künstlichen Intelligenz - Deep Learning für 1D bis 4D biomedizinischen Daten'') bietet eine umfassende Untersuchung des Lebenszyklus von Künstlicher Intelligenz (KI) in verschiedenen Bereichen der Lebenswissenschaften und der Medizin, wobei der Schwerpunkt auf der Anwendung von Maschinellem Lernen (ML)- und Deep Learning (DL)-Techniken auf Daten von 1D bis 4D liegt. Die Studie beginnt in Kapitel 2 mit dem KI-Lebenszyklus, welcher die Phasen der Problemdefinition, Datenerfassung, Modelltraining, Bewertung, Interpretation und Implementierung umfasst. Ein wesentlicher Teil der Dissertation befasst sich mit der Interpretation von Elektrokardiogramm (EKG)-Daten, wie in Kapitel 3 beschrieben. Dieses Kapitel stellt die PTB-XL-Datenbank vor, einen groß angelegten Datensatz für die EKG-Analyse, und erörtert die Anwendung von erklärbarer KI (XAI)-Techniken zur Verbesserung der Modelltransparenz und Vertrauenswürdigkeit. Die Studie zeigt, wie XAI eingesetzt werden kann, um KI-Modelle interpretierbarer und klinisch relevanter zu machen, was entscheidend für ihre Akzeptanz in der medizinischen Diagnostik ist. Anschließend wird in Kapitel 4 die Segmentierung verschiedener biologischer Strukturen in 2D-histologischen Ganzschnittaufnahmen behandelt, die für eine präzise medizinische Diagnostik, insbesondere in der Dermatologie, von entscheidender Bedeutung ist. Das Kapitel beleuchtet die Herausforderungen und potenziellen Lösungen bei der Implementierung von DL-Modellen zu diesem Zweck und betont die Bedeutung einer präzisen Segmentierung zur Verbesserung der diagnostischen Genauigkeit. Mit zunehmender dimensionaler Komplexität befasst sich Kapitel 5 mit der Analyse von 3D-Formen der Neutrophilen-Zellkerne mittels konfokaler Fluoreszenzmikroskopie. Dieses Kapitel liefert Einblicke in die zelluläre Morphologie, indem es Merkmalsextraktion, Dimensionsreduktion und Clusterverfahren einsetzt, um die Variationen in den Formen der Neutrophilen zu verstehen. Die Analyse unterstreicht das Potenzial von KI, morphologische Merkmale aufzudecken, die mit konventionellen Methoden schwer zugänglich oder nur begrenzt erkennbar sind. Die letzte in dieser Dissertation behandelte Anwendung umfasst die Analyse der Dynamik von Lymphozyten in 4D-Zeitraffer-Aufnahmen in menschlichem Lymphgewebe, wie in Kapitel 6 dargestellt. Dieses Kapitel untersucht die Dynamiken von Lymphozyten mithilfe von konfokaler Mikroskopie und zeigt, wie KI genutzt werden kann, um neue Erkenntnisse über das Verhalten von Zellen im Zeitverlauf zu gewinnen. Die hier vorgestellte Arbeit hebt die bedeutenden Beiträge von KI zum Verständnis komplexer biologischer Prozesse hervor, die potenziell zu neuen Entdeckungen in der Immunologie führen können. Abschließend verbessert diese Dissertation nicht nur den aktuellen Stand der Technik bei der Anwendung von KI in den Lebenswissenschaften, sondern bietet auch ein systematischen Einblick in KI-Modellen in der medizinischen Forschung. Die Forschung verdeutlicht, dass entlang des gesamten KI-Lebenszyklus, von der Problemdefinition bis hin zur Implementierung, umfangreiche Entwicklungen stattfinden, um KI-Lösungen zu schaffen, die nicht nur technisch fundiert, sondern auch klinisch anwendbar sind. Die Ergebnisse deuten auf mögliche Implikationen für die Zukunft der KI in der Medizin hin, insbesondere bei der Verbesserung der Interpretierbarkeit und Transparenz von KI-Modellen, die für ihre Integration in die klinische Praxis unerlässlich sind."]},{"key":"dc:title","label":"Title","values":["Exploring the lifecycle of artificial intelligence – deep learning for 1D to 4D biomedical data"]}]}],"canonical_facts":{"dc:contributor.advisor":["Samek, Wojciech"],"dc:creator":["Wagner, Patrick"],"dc:date.accessioned":["2025-04-10T10:18:56Z"],"dc:date.available":["2025-04-10T10:18:56Z"],"dc:date.issued":["2025"],"dc:description.abstract":["This thesis, titled \"Exploring the Lifecycle of Artificial Intelligence - Deep Learning for 1D to 4D Biomedical Data\" presents a comprehensive exploration of the Artificial Intelligence (AI) lifecycle as applied to various domains within the life sciences and medicine, specifically focusing on the application of Machine Learning (ML) and Deep Learning (DL) techniques to data ranging from one to four dimensional data. The study is structured around the AI lifecycle, which is elaborated upon in Chapter 2. This lifecycle encompasses the critical stages of problem definition, data acquisition, model training, evaluation, interpretation, and deployment, offering a holistic approach to the development and implementation of AI systems in medical contexts. A significant portion of the thesis delves into the interpretation of Electrocardiography (ECG) data, as detailed in Chapter 3. This chapter introduces the PTB-XL database, a large-scale dataset for ECG analysis, and discusses the application of eXplainable Artificial Intelligence techniques to enhance model transparency and trustworthiness. The study demonstrates how XAI can be employed to make AI models more interpretable and clinically relevant, which is crucial for their adoption in medical diagnostics. The thesis then shifts focus to the segmentation of 2D histological images, which is critical for accurate medical diagnostics, particularly in dermatology. As described in Chapter 4, this work explores advanced semantic segmentation techniques to identify and classify different biological structures within a Whole Slide Image (WSI). The chapter highlights the challenges and potential solutions in implementing DL models for this purpose, emphasizing the importance of precise segmentation in improving diagnostic accuracy. Further expanding on the dimensional complexity, Chapter 5 addresses the analysis of 3D neutrophil cell nuclei shapes using confocal fluorescence microscopy. This chapter provides insights into the cellular morphology, employing feature extraction, dimensionality reduction, and clustering methods to understand the variations in neutrophil shapes. The analysis underscores the potential of AI in uncovering morphological features that are not easily discernible through traditional methods. The final application discussed in this thesis involves the analysis of 4D lymphocyte dynamics in human lymphoid tissue, as outlined in Chapter 6. This chapter examines the temporal dynamics of lymphocytes using time-lapse confocal microscopy, demonstrating how AI can be utilized to gain novel insights into cellular behavior over time. The work presented here highlights the significant contributions of AI to understanding complex biological processes, potentially leading to new discoveries in immunology. In conclusion, this thesis not only advances the state-of-the-art in the application of AI to the life sciences but also provides a robust framework for the systematic development and deployment of AI models in medical research. By adopting the AI lifecycle approach, the research aims to address each stage, from problem definition to deployment, in a comprehensive manner, with the goal of developing AI solutions that are not only technically sound but also have potential for clinical applicability. The findings of this thesis suggest potential implications for the future of AI in medicine, particularly in enhancing the interpretability, transparency, and trustworthiness of AI models, which are essential for their integration into clinical practice.","Diese Dissertation mit dem Titel \"Exploring the Lifecycle of Artificial Intelligence - Deep Learning for 1D to 4D Biomedical Data\" (zu deutsch: \"Erforschung des Lebenszyklus der künstlichen Intelligenz - Deep Learning für 1D bis 4D biomedizinischen Daten'') bietet eine umfassende Untersuchung des Lebenszyklus von Künstlicher Intelligenz (KI) in verschiedenen Bereichen der Lebenswissenschaften und der Medizin, wobei der Schwerpunkt auf der Anwendung von Maschinellem Lernen (ML)- und Deep Learning (DL)-Techniken auf Daten von 1D bis 4D liegt. Die Studie beginnt in Kapitel 2 mit dem KI-Lebenszyklus, welcher die Phasen der Problemdefinition, Datenerfassung, Modelltraining, Bewertung, Interpretation und Implementierung umfasst. Ein wesentlicher Teil der Dissertation befasst sich mit der Interpretation von Elektrokardiogramm (EKG)-Daten, wie in Kapitel 3 beschrieben. Dieses Kapitel stellt die PTB-XL-Datenbank vor, einen groß angelegten Datensatz für die EKG-Analyse, und erörtert die Anwendung von erklärbarer KI (XAI)-Techniken zur Verbesserung der Modelltransparenz und Vertrauenswürdigkeit. Die Studie zeigt, wie XAI eingesetzt werden kann, um KI-Modelle interpretierbarer und klinisch relevanter zu machen, was entscheidend für ihre Akzeptanz in der medizinischen Diagnostik ist. Anschließend wird in Kapitel 4 die Segmentierung verschiedener biologischer Strukturen in 2D-histologischen Ganzschnittaufnahmen behandelt, die für eine präzise medizinische Diagnostik, insbesondere in der Dermatologie, von entscheidender Bedeutung ist. Das Kapitel beleuchtet die Herausforderungen und potenziellen Lösungen bei der Implementierung von DL-Modellen zu diesem Zweck und betont die Bedeutung einer präzisen Segmentierung zur Verbesserung der diagnostischen Genauigkeit. Mit zunehmender dimensionaler Komplexität befasst sich Kapitel 5 mit der Analyse von 3D-Formen der Neutrophilen-Zellkerne mittels konfokaler Fluoreszenzmikroskopie. Dieses Kapitel liefert Einblicke in die zelluläre Morphologie, indem es Merkmalsextraktion, Dimensionsreduktion und Clusterverfahren einsetzt, um die Variationen in den Formen der Neutrophilen zu verstehen. Die Analyse unterstreicht das Potenzial von KI, morphologische Merkmale aufzudecken, die mit konventionellen Methoden schwer zugänglich oder nur begrenzt erkennbar sind. Die letzte in dieser Dissertation behandelte Anwendung umfasst die Analyse der Dynamik von Lymphozyten in 4D-Zeitraffer-Aufnahmen in menschlichem Lymphgewebe, wie in Kapitel 6 dargestellt. Dieses Kapitel untersucht die Dynamiken von Lymphozyten mithilfe von konfokaler Mikroskopie und zeigt, wie KI genutzt werden kann, um neue Erkenntnisse über das Verhalten von Zellen im Zeitverlauf zu gewinnen. Die hier vorgestellte Arbeit hebt die bedeutenden Beiträge von KI zum Verständnis komplexer biologischer Prozesse hervor, die potenziell zu neuen Entdeckungen in der Immunologie führen können. Abschließend verbessert diese Dissertation nicht nur den aktuellen Stand der Technik bei der Anwendung von KI in den Lebenswissenschaften, sondern bietet auch ein systematischen Einblick in KI-Modellen in der medizinischen Forschung. Die Forschung verdeutlicht, dass entlang des gesamten KI-Lebenszyklus, von der Problemdefinition bis hin zur Implementierung, umfangreiche Entwicklungen stattfinden, um KI-Lösungen zu schaffen, die nicht nur technisch fundiert, sondern auch klinisch anwendbar sind. Die Ergebnisse deuten auf mögliche Implikationen für die Zukunft der KI in der Medizin hin, insbesondere bei der Verbesserung der Interpretierbarkeit und Transparenz von KI-Modellen, die für ihre Integration in die klinische Praxis unerlässlich sind."],"dc:identifier.uri":["https://depositonce.tu-berlin.de/handle/11303/24202","https://doi.org/10.14279/depositonce-23016"],"dc:language.iso":["en"],"dc:rights.uri":["https://creativecommons.org/licenses/by-sa/4.0/"],"dc:title":["Exploring the lifecycle of artificial intelligence – deep learning for 1D to 4D biomedical data"],"dc:type":["Doctoral Thesis"]},"updated_at":"2026-07-27T21:28:49Z"}