Bio-inspired Deep Learning Image and Signal Processing Pipelines in Medical Oncology: From Radiomics to Pathomics for non-invasive oncological early diagnosis and outcome prediction - a PeerJ Collection


Modern approaches to the quantitative analysis of medical images (CT-scans, MRI, PET, SPECT, etc..) and biomedical signals (Photoplethysmography (PPG), Electrocardiography (ECG), ElectroEncephaloGraphy (EEG), etc..) have recently opened up exciting avenues of research in the healthcare field. The term Radiomics was recently coined in this area to describe methods that use algorithms and statistical methods to extract large amounts of data from medical images. In the oncology field, both recent Deep Learning techniques and Radiomics approaches have significantly improved both diagnosis and cancer treatment.

Similarly, the term "Pathomics" has been coined, which refers to the use of modern techniques of deep learning and advanced mathematical modeling for the characterization of immuno-histological images aimed at predicting the patient's prognosis and of the response to a specific medical treatment (chemotherapy, immunotherapy and so on).

The objective of this Collection is to collect scientific contributions that confirm the advantages inherent in the application of modern techniques of Radiomics, Pathomics and Deep Learning in the analysis of medical images and biomedical signals in the field of oncology. More in detail, the target of this Collection is: a) to provide a comprehensive overview of the most recent advanced methods for estimating oncological treatment outcome from analysis of biomedical signals and images of the patients; b) to investigate multimodal analysis of biomedical data, signals and images for cancer early diagnosis; and c) to present and report new applications of Radiomics and Pathomics.

The expected topics include, but are not limited to the following:

• Recent advances of Radiomics/Pathomics in medical oncology;

• Cancer dynamic modelling from medical images/signals;

• Motion magnification in medical imaging;

• Fractal and chaos in medical images and biomedical signals;

• Biomedical signals for oncological characterization;

• Recent advances of Deep Learning in medical imaging for oncological applications;

• Advanced Mathematical modelling of cancer;

• Radiomics algorithms for chemotherapy outcome prediction from medical image analysis (CT-Scans, MRI, SPECT, PET, Echo);

• Radiomics algorithms for chemotherapeutic outcome prediction from biomedical signal analysis (PPG, ECG, EEG, etc..);

• Radiomics algorithms for anti PD-1/PD-L1 immunotherapeutic outcome prediction from medical image analysis (CT-Scans, MRI, SPECT, PET, Echo);

• Radiomics algorithms for chemotherapeutic outcome prediction from biomedical signal analysis (PPG, ECG, EEG, etc..);

• Oncological outcome prediction from advanced analysis of immune-histological tissue images;

• Biomedical signals and medical images data fusion for oncological applications;

• Cancer early diagnosis from biomedical signals and images analysis;

Reviews and surveys of the state-of-the-art are also welcomed.

Also available to other groups