Self-Supervised Deep Learning for Compressed Sensing-Based Recovery
A special issue of Sensors (ISSN 1424-8220). This special issue belongs to the section "Sensing and Imaging".
Deadline for manuscript submissions: closed (31 July 2023) | Viewed by 302
Special Issue Editor
Special Issue Information
Dear Colleagues,
For many biomedical applications, deep learning (DL)-based recovery methods have shown great potential to improve signal quality and/or accelerate the acquisition process. Recent evidence suggests that DL-based methods can outperform sparsity-driven recovery methods, especially for biomedical imaging. Typically, these DL-based methods rely on supervised learning to train a convolutional neural network (CNN) that recovers signals from noisy and potentially incomplete data. Other supervised learning techniques are inspired by variational optimization methods where an iterative algorithm is unrolled and iterates between data consistency enforcement and CNN application, which provides regularization. Despite the improvements offered by DL-based methods, their extension to applications where training data are scarce remains challenging. For example, collecting densely sampled data from biomedical sensors is not always feasible for dynamic applications. In addition, the test data may have a mismatch with the training database, which can lead to performance degradation. More recently, there has been an increased interest in developing self-supervised methods that require little or no training data. In this Special Issue, we explore self-supervised methods that not only provide state-of-the-art performance but also lower the demand for the training data and thus extend the benefit of DL-based signal recovery methods to broader applications.
Dr. Rizwan Ahmad
Guest Editor
Manuscript Submission Information
Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.
Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Sensors is an international peer-reviewed open access semimonthly journal published by MDPI.
Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.
Keywords
- deep learning
- compressed sensing
- self-supervised learning
- biomedical sensors
- signal acquisition
- image recovery
- biomedical imaging
- medical image analysis
- image reconstruction
