Scalable Deep Learning for Healthcare Analytics

Dear Colleagues,

Healthcare is deemed the most important field due to its contribution to human beneficiary sectors such as hospitals, physicians, nursing homes, diagnostic laboratories, pharmacies, and medical devices. Healthcare sectors acquire a large amount of data about patients, doctors, medicines, diseases, and disease prediction and cure, and much more for various analytics applications. Many datasets with high volume and large variety take the form of big data. Healthcare data is continuously changing through findings and discoveries like new diseases and cures and improved medications. Therefore, scientists and doctors must learn and handle all these concerns deeply and intelligently. Artificial intelligence is one of the best solutions. Among its extensions, the inherent ability of deep learning (DL) to discover correlations from vast quantities of data in an unsupervised fashion has been the main drive for its wide adoption in broad range of medical applications, including MRI scans, CT scans, X-rays, and pathological tests. DL also enables the dynamic discovery of features from data, unlike traditional machine learning approaches, where feature selection remains a challenge.

DL is scalable and upgraded via increasing the data and its quality in the algorithm. Until this time, for disease-specific features, different diagnostic computer programs were coded keeping in view the sequence of predefined assumptions. Such customized programs need to be designed and developed for each body part to identify diseases, but their flexibility and scalability are not properly measurable and hence have never reached widespread clinical adoption due to their oversimplification of reality and poor diagnostic performance. Therefore, as the data size increases exponentially and the deep learning models become more complex, more computing power and memory is required, such as high-performance computing (HPC) resources to train an accurate model in a timely manner. The existing methods are not sufficient to systematically harness such systems/clusters. Therefore, scalability is essential in high-performance computing systems, and is one of the key evaluation criteria of computing systems. There is a need to develop new parallel and distributed algorithms/frameworks for scalable deep learning in healthcare which can accelerate the training process and make it suitable for big data processing and analysis.

This Special Issue aims to provide a forum to qualified and expert researchers, practitioners, and scientific communities to present their innovative and valuable contributions regarding scalable deep leaning in healthcare to overcome health-related issues. It is especially important to develop deep networks to capture normal-appearing lesions which may be neglected by human interpretation. Potential topics include but are not limited to the following:

• Scalable machine learning models, including deep learning algorithms, for extreme-scale systems;
• Enhancing applicability of machine learning in HPC (e.g., feature engineering, usability);
• Learning large models/optimizing hyperparameters (e.g., deep learning, representation learning);
• Facilitating very large ensembles in extreme-scale systems;
• Training machine learning models on large datasets and scientific data;
• Overcoming the problems inherent in processing large datasets (e.g., noisy labels, missing data, scalable input);
• Applications of machine learning utilizing HPC;
• Secure deep learning algorithms for healthcare applications;
• Future research challenges for machine learning on a large scale;
• Large-scale machine learning applications;
• Extreme scale, multicore, GPU accelerators and novel architectures for rethinking scalability
  • Parallel programming models and tools;
  • GPU-, MIC-, and FPGA-based parallel systems, heterogeneous platforms;
  • Extreme-scale systems and applications;
  • Peta-scale and exa-scale workloads;
  • High-performance and high-throughput computing;
  • Fault tolerance in large-scale applications;
  • Near-data processing and data-centric approaches;
• Tools for Big Data
  • New data analytics tools for extreme big data;
  • Distributed architectures/parallel programming models for machine learning/deep learning.

Dr. Sohail Jabbar
Dr. Ahcène Bounceur
Dr. Farhan Ullah
Guest Editors

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. Future Internet is an international peer-reviewed open access monthly 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 1600 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.