González, Camila ; Ranem, Amin ; Othman, Ahmed ; Mukhopadhyay, Anirban (2022)
Task-Agnostic Continual Hippocampus Segmentation for Smooth Population Shifts.
4th MICCAI workshop on Domain Adaptation and Representation Transfer. Singapore (22.09.2022-22.09.2022)
doi: 10.1007/978-3-031-16852-9_11
Konferenzveröffentlichung, Bibliographie
Kurzbeschreibung (Abstract)
Most continual learning methods are validated in settings where task boundaries are clearly defined and task identity information is available during training and testing. We explore how such methods perform in a task-agnostic setting that more closely resembles dynamic clinical environments with gradual population shifts. We propose ODEx, a holistic solution that combines out-of-distribution detection with continual learning techniques. Validation on two scenarios of hippocampus segmentation shows that our proposed method reliably maintains performance on earlier tasks without losing plasticity.
| Typ des Eintrags: | Konferenzveröffentlichung |
|---|---|
| Erschienen: | 2022 |
| Autor(en): | González, Camila ; Ranem, Amin ; Othman, Ahmed ; Mukhopadhyay, Anirban |
| Art des Eintrags: | Bibliographie |
| Titel: | Task-Agnostic Continual Hippocampus Segmentation for Smooth Population Shifts |
| Sprache: | Englisch |
| Publikationsjahr: | 2022 |
| Verlag: | Springer |
| Buchtitel: | Domain Adaptation and Representation Transfer |
| Reihe: | Lecture Notes in Computer Science |
| Band einer Reihe: | 13542 |
| Veranstaltungstitel: | 4th MICCAI workshop on Domain Adaptation and Representation Transfer |
| Veranstaltungsort: | Singapore |
| Veranstaltungsdatum: | 22.09.2022-22.09.2022 |
| DOI: | 10.1007/978-3-031-16852-9_11 |
| Kurzbeschreibung (Abstract): | Most continual learning methods are validated in settings where task boundaries are clearly defined and task identity information is available during training and testing. We explore how such methods perform in a task-agnostic setting that more closely resembles dynamic clinical environments with gradual population shifts. We propose ODEx, a holistic solution that combines out-of-distribution detection with continual learning techniques. Validation on two scenarios of hippocampus segmentation shows that our proposed method reliably maintains performance on earlier tasks without losing plasticity. |
| Freie Schlagworte: | Continual learning, Lifelong learning, Distribution shift |
| Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Graphisch-Interaktive Systeme |
| Hinterlegungsdatum: | 15 Jun 2023 07:43 |
| Letzte Änderung: | 27 Feb 2024 12:55 |
| PPN: | 515843296 |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum