Deep reinforcement learning and convolutional autoencoders for anomaly detection of congenital inner ear malformations in clinical CT images

Detection of abnormalities within the inner ear is a challenging task even for experienced clinicians. In this study, we propose an automated method for automatic abnormality detection to provide support for the diagnosis and clinical management of various otological disorders. We propose a framewor...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Computerized medical imaging and graphics Ročník 113; s. 102343
Hlavní autoři:	López Diez, Paula, Sundgaard, Josefine Vilsbøll, Margeta, Jan, Diab, Khassan, Patou, François, Paulsen, Rasmus R.
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	United States Elsevier Ltd 01.04.2024
Témata:	Anomaly detection Cochlear implant Computed tomography Congenital malformation Deep reinforcement learning Internal Medicine Other Cochlear implant Deep reinforcement learning Computed tomography Congenital malformation Anomaly detection
ISSN:	0895-6111, 1879-0771, 1879-0771
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	Detection of abnormalities within the inner ear is a challenging task even for experienced clinicians. In this study, we propose an automated method for automatic abnormality detection to provide support for the diagnosis and clinical management of various otological disorders. We propose a framework for inner ear abnormality detection based on deep reinforcement learning for landmark detection which is trained uniquely in normative data. In our approach, we derive two abnormality measurements: Dimage and Uimage. The first measurement, Dimage, is based on the variability of the predicted configuration of a well-defined set of landmarks in a subspace formed by the point distribution model of the location of those landmarks in normative data. We create this subspace using Procrustes shape alignment and Principal Component Analysis projection. The second measurement, Uimage, represents the degree of hesitation of the agents when approaching the final location of the landmarks and is based on the distribution of the predicted Q-values of the model for the last ten states. Finally, we unify these measurements in a combined anomaly measurement called Cimage. We compare our method’s performance with a 3D convolutional autoencoder technique for abnormality detection using the patch-based mean squared error between the original and the generated image as a basis for classifying abnormal versus normal anatomies. We compare both approaches and show that our method, based on deep reinforcement learning, shows better detection performance for abnormal anatomies on both an artificial and a real clinical CT dataset of various inner ear malformations with an increase of 11.2% of the area under the ROC curve. Our method also shows more robustness against the heterogeneous quality of the images in our dataset. •Congenital inner ear malformation detection is key for better development of children with severe hearing loss.•Anomaly detection with automated detection of congenital malformations of the inner ear.•Landmark localization with deep reinforcement learning for anomaly detection in CT images.•Trained only in normative data which is more accessible and easier to collect.•Evaluate and compare our method with a 3D convolutional autoencoder in two clinical datasets of artificial and real congenital inner ear malformations.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
ISSN:	0895-6111 1879-0771 1879-0771
DOI:	10.1016/j.compmedimag.2024.102343