The ability to identify, diagnose and classify tumours as early as possible can have an immeasurable impact on the opportunities presented in treating and curing cancer cases. The difference between benign and malignant tumours and every subsequent decision and action of the cancer care team is extremely significant, so the ability to identify a benign or a malignant tumour early is imperative and will go on to influence the treatment and monitoring cascade in almost every way. The ability to make this identification in a non-invasive manner through the implementation of imaging techniques such as computed tomography (CT) or magnetic resonance imaging (MRI) is greatly beneficial, as the potential risks associated with surgeries and invasive extraction of biopsies are completely circumvented. However, cancer diagnoses are far from black and white, and can differ in a magnitude of ways, from the tissues and organs they affect, to the types of cells they originated from, to the positive and negative statuses of hormone production - making it no easy task to differentiate malignant tumours from benign via imaging procedures.
By evaluating scans of tumours for characteristics which include the presence of low-signal septation, variation in the perceived pattern of tissue homogeneity, and the definition and shapes of distinguishable tissues enables pre-operative assumptions to be made regarding the likelihood of malignancy of tumours. With advancements in artificial intelligence (AI) and deep learning neural networks, these processes and decisions could be made automatically or even near instantaneously, allowing preliminary decisions to be made which can be backed up with further clinical parameters or biopsy results.
For this reason we aim to bring together research articles in the field which outline how radiologists can differentiate malignant from benign tumours, either in isolation or through conferring with other clinicians in the cancer care team and involving clinical measurements and parameters. We also welcome submissions outlining how AI and deep learning can enhance or even automate these processes to the benefit of cancer care teams. Through submitted research articles we aim to provide a foundation of knowledge which can be used as the basis for the production of nomograms which outline the procedures to be followed by radiologists in the identification and classification of tumours.
Important Note: Manuscripts consisting solely of bioinformatics, computational analysis, or predictions of public databases which are not accompanied by validation (independent cohort or biological validation in vitro or in vivo) will not be accepted in any of the sections of Frontiers in Oncology.
The ability to identify, diagnose and classify tumours as early as possible can have an immeasurable impact on the opportunities presented in treating and curing cancer cases. The difference between benign and malignant tumours and every subsequent decision and action of the cancer care team is extremely significant, so the ability to identify a benign or a malignant tumour early is imperative and will go on to influence the treatment and monitoring cascade in almost every way. The ability to make this identification in a non-invasive manner through the implementation of imaging techniques such as computed tomography (CT) or magnetic resonance imaging (MRI) is greatly beneficial, as the potential risks associated with surgeries and invasive extraction of biopsies are completely circumvented. However, cancer diagnoses are far from black and white, and can differ in a magnitude of ways, from the tissues and organs they affect, to the types of cells they originated from, to the positive and negative statuses of hormone production - making it no easy task to differentiate malignant tumours from benign via imaging procedures.
By evaluating scans of tumours for characteristics which include the presence of low-signal septation, variation in the perceived pattern of tissue homogeneity, and the definition and shapes of distinguishable tissues enables pre-operative assumptions to be made regarding the likelihood of malignancy of tumours. With advancements in artificial intelligence (AI) and deep learning neural networks, these processes and decisions could be made automatically or even near instantaneously, allowing preliminary decisions to be made which can be backed up with further clinical parameters or biopsy results.
For this reason we aim to bring together research articles in the field which outline how radiologists can differentiate malignant from benign tumours, either in isolation or through conferring with other clinicians in the cancer care team and involving clinical measurements and parameters. We also welcome submissions outlining how AI and deep learning can enhance or even automate these processes to the benefit of cancer care teams. Through submitted research articles we aim to provide a foundation of knowledge which can be used as the basis for the production of nomograms which outline the procedures to be followed by radiologists in the identification and classification of tumours.
Important Note: Manuscripts consisting solely of bioinformatics, computational analysis, or predictions of public databases which are not accompanied by validation (independent cohort or biological validation in vitro or in vivo) will not be accepted in any of the sections of Frontiers in Oncology.