hh.sePublications
Change search
Link to record
Permanent link

Direct link
BETA
Publications (3 of 3) Show all publications
Polymeri, E., Sadik, M., Kaboteh, R., Borrelli, P., Enqvist, O., Ulén, J., . . . Edenbrandt, L. (2019). Deep learning-based quantification of PET/CT prostate gland uptake: association with overall survival. Clinical Physiology and Functional Imaging
Open this publication in new window or tab >>Deep learning-based quantification of PET/CT prostate gland uptake: association with overall survival
Show others...
2019 (English)In: Clinical Physiology and Functional Imaging, ISSN 1475-0961, E-ISSN 1475-097XArticle in journal (Refereed) In press
Abstract [en]

Aim: To validate a deep-learning (DL) algorithm for automated quantification of prostate cancer on positron emission tomography/computed tomography (PET/CT) and explore the potential of PET/CT measurements as prognostic biomarkers. Material and methods: Training of the DL-algorithm regarding prostate volume was performed on manually segmented CT images in 100 patients. Validation of the DL-algorithm was carried out in 45 patients with biopsy-proven hormone-naïve prostate cancer. The automated measurements of prostate volume were compared with manual measurements made independently by two observers. PET/CT measurements of tumour burden based on volume and SUV of abnormal voxels were calculated automatically. Voxels in the co-registered 18F-choline PET images above a standardized uptake value (SUV) of 2·65, and corresponding to the prostate as defined by the automated segmentation in the CT images, were defined as abnormal. Validation of abnormal voxels was performed by manual segmentation of radiotracer uptake. Agreement between algorithm and observers regarding prostate volume was analysed by Sørensen-Dice index (SDI). Associations between automatically based PET/CT biomarkers and age, prostate-specific antigen (PSA), Gleason score as well as overall survival were evaluated by a univariate Cox regression model. Results: The SDI between the automated and the manual volume segmentations was 0·78 and 0·79, respectively. Automated PET/CT measures reflecting total lesion uptake and the relation between volume of abnormal voxels and total prostate volume were significantly associated with overall survival (P = 0·02), whereas age, PSA, and Gleason score were not. Conclusion: Automated PET/CT biomarkers showed good agreement to manual measurements and were significantly associated with overall survival. © 2019 The Authors. Clinical Physiology and Functional Imaging published by John Wiley & Sons Ltd on behalf of Scandinavian Society of Clinical Physiology and Nuclear Medicine

Place, publisher, year, edition, pages
Blackwell Publishing, 2019
Keywords
artificial intelligence, convolutional neural network, objective quantification, prostatic neoplasms
National Category
Medical Image Processing Radiology, Nuclear Medicine and Medical Imaging Medical Laboratory and Measurements Technologies Cardiac and Cardiovascular Systems
Identifiers
urn:nbn:se:hh:diva-41541 (URN)10.1111/cpf.12611 (DOI)2-s2.0-85076741263 (Scopus ID)
Available from: 2020-02-03 Created: 2020-02-03 Last updated: 2020-02-03Bibliographically approved
Abiri, N., Linse, B., Edén, P. & Ohlsson, M. (2019). Establishing strong imputation performance of a denoising autoencoder in a wide range of missing data problems. Neurocomputing, 65, 137-146
Open this publication in new window or tab >>Establishing strong imputation performance of a denoising autoencoder in a wide range of missing data problems
2019 (English)In: Neurocomputing, ISSN 0925-2312, E-ISSN 1872-8286, Vol. 65, p. 137-146Article in journal (Refereed) Published
Abstract [en]

Dealing with missing data in data analysis is inevitable. Although powerful imputation methods that address this problem exist, there is still much room for improvement. In this study, we examined single imputation based on deep autoencoders, motivated by the apparent success of deep learning to efficiently extract useful dataset features. We have developed a consistent framework for both training and imputation. Moreover, we benchmarked the results against state-of-the-art imputation methods on different data sizes and characteristics. The work was not limited to the one-type variable dataset; we also imputed missing data with multi-type variables, e.g., a combination of binary, categorical, and continuous attributes. To evaluate the imputation methods, we randomly corrupted the complete data, with varying degrees of corruption, and then compared the imputed and original values. In all experiments, the developed autoencoder obtained the smallest error for all ranges of initial data corruption. © 2019 Elsevier B.V.

Place, publisher, year, edition, pages
Amsterdam: Elsevier, 2019
Keywords
Deep learning, Autoencoder, Imputation, Missing data
National Category
Other Computer and Information Science
Identifiers
urn:nbn:se:hh:diva-41245 (URN)10.1016/j.neucom.2019.07.065 (DOI)000484072600014 ()2-s2.0-85069939556 (Scopus ID)
Funder
Swedish Foundation for Strategic Research
Available from: 2019-12-13 Created: 2019-12-13 Last updated: 2019-12-13Bibliographically approved
Najmeh, A. & Ohlsson, M. (2019). Variational auto-encoders with Student’s t-prior. In: ESANN 2019 Proceedings, 27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning: Bruges – 24-26 April 2019. Paper presented at 27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2019), Bruges, Belgium, April 24-26, 2019 (pp. 415-420). Bruges: ESANN
Open this publication in new window or tab >>Variational auto-encoders with Student’s t-prior
2019 (English)In: ESANN 2019 Proceedings, 27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning: Bruges – 24-26 April 2019, Bruges: ESANN , 2019, p. 415-420Conference paper, Published paper (Refereed)
Abstract [en]

We propose a new structure for the variational auto-encoders (VAEs) prior, with the weakly informative multivariate Student’s t-distribution. In the proposed model all distribution parameters are trained, thereby allowing for a more robust approximation of the underlying data distribution. We used Fashion-MNIST data in two experiments to compare the proposed VAEs with the standard Gaussian priors. Both experiments showed a better reconstruction of the images with VAEs using Student’s t-prior distribution. © 2019 ESANN (i6doc.com). All rights reserved.

Place, publisher, year, edition, pages
Bruges: ESANN, 2019
National Category
Other Computer and Information Science
Identifiers
urn:nbn:se:hh:diva-41247 (URN)2-s2.0-85071324436 (Scopus ID)978-287-587-065-0 (ISBN)
Conference
27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2019), Bruges, Belgium, April 24-26, 2019
Available from: 2019-12-13 Created: 2019-12-13 Last updated: 2019-12-19Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0003-1145-4297

Search in DiVA

Show all publications