Melba (The Journal of Machine Learning for Biomedical Imaging) is a web-based journal devoted to the free and unrestricted access of high quality articles in the broad field that bridges machine learning and biomedical imaging. Melba’s aims include:
- to be a top tier publication venue for journal-length articles in its scope;
- to provide recognition of meritorious scholarship;
- to provide curated content that will help researchers keep abreast of rapid developments;
- to incentivize and to support the sharing of resources, including code and data;
- to empower authors with copyrights to their own work;
- to encourage a constructive, rigorous, and streamlined review process that enables authors to improve their articles;
- to offer unrestricted and free access to readers;
- to recognize the efforts of reviewers; and
- to minimize the financial burden on authors by leveraging web-based technologies.
Melba (The Journal of Machine Learning for Biomedical Imaging) invites the submission of previously unpublished journal-length papers that report research developments at the interface of machine learning and biomedical imaging. Papers focusing on innovative methods and/or novel biomedical applications are strongly encouraged to be submitted.Topics of interest include:
- new machine learning techniques that are motivated by or applicable to known problems in biomedical imaging, with strong theoretical justification and appropriate empirical evaluation;
- empirical accounts that evaluate, and/or compare existing machine learning algorithms for biomedical imaging applications of interest;
- presentation of a novel biomedical application involving imaging, where existing machine algorithms offer a key solution;
- description of a new problem formulation for machine learning and/or a new publicly available dataset that is relevant to a biomedical imaging application, and a baseline attempt at solving the problem; and
- well-written surveys of existing work relevant to the scope of the journal.
Overall, Melba aims to publish high-quality research contributions that add value and move science forward, including negative results that can be generalized from. Melba encourages authors to share and support their code and data, with an emphasis on replicability.
Manuscripts must communicate their ideas and findings in the English language, in a concise and complete manner. Therefore papers should be carefully proofread and polished by authors. If these criteria are not met, this can be grounds for rejecting the paper without a review.
Melba adheres to standard practices in modern scientific publication. This means that all claims need to be articulated precisely and justified through theoretical arguments and/or empirical evidence. Prior literature should be properly acknowledged. Submitted papers need to be original and not published. Significantly extended versions of conference proceedings are allowed to be submitted, however the overlap of content should be less than 50%.