Artificial Intelligence in Oncology: Improving Imaging Diagnostics and Treatment

Dear Colleagues,

The emergence of artificial intelligence has irreversibly affected numerous medical fields, among them—oncology.

Oncology itself has always been the specialty harboring cutting-edge technologies and solutions. Not surprisingly, rapid advances in the field of machine learning and artificial intelligence are fueling the development of computational oncology, radiomics and other areas where mathematical models supported by large amounts of data can assist oncologists.   

Throughout the past few decades, we have witnessed an incredible, ever-increasing amount of data that are recorded, analyzed and retrieved. Moreover, mining huge datasets uncovers certain correlations and interrelationships that, otherwise, would have gone unnoticed. 

This change pertains to the handling of datasets (fusion of variables originating from distinct realms of medical records, e.g., images, texts, signals, metadata, searching for relevant studies), mining and analysis (AI-based algorithms), as well as the discovery of novel relations, insights and knowledge.

Initially, machine learning models helped in the processing of vast datasets. However, the raw data had to be converted to higher-level features defined based on expert knowledge.

Subsequently, due to their higher computation power and much larger datasets, deep learning models became popularized, wherein datasets comprising various components (speech, text, images) would be fed into models as the raw data and transformed to higher-level features by the algorithm. This has constituted a tremendous leap forward.

While knowledge was initially transferred from the domain expert to the model, the development of foundation models has led to an increasing transfer of data representations learned by one generic model to models that perform specific tasks. Foundation models are capable of absorbing various data from different modalities in their raw form. The power of foundation models is their ability to discover novel functionalities, which stems from transfer learning.

The development of the model must be conducted in a sustainable and responsible way, respecting the fundamental principle of “first, do no harm”. All systems produce some predictions/outputs based on the inputs provided. However, the decision-making tree is not exposed and is, therefore, unverifiable to a user. This aforementioned new branch of AI allows questions to be asked of the decision-making process, thus granting full interaction between a user and a model. This constitutes a unique opportunity to learn from the AI system.

In this Special Issue, the goal is to demonstrate:

1. How AI algorithms have permeated oncology, starting from diagnosis, both pathologically and with the use of a variety of imaging techniques;
2. What AI’s power is for predicting treatment results (more relevant predictors discovery);
3. What the AI’s role is in monitoring treatment outcomes, utilizing extractable prognostic factors.

There is no threat that AI will eliminate physicians one day. However, it is likely that physicians who are not involved in the active use of AI methodology will be replaced by those who use it.

Prof. Mariusz Adamek
Prof. Przemyslaw Biecek
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. International Journal of Environmental Research and Public Health 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 2500 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.

• computational oncology
• computational histology
• computational radiology
• radiomics
• explainable artificial intelligence
• human oriented artificial intelligence