Prediction of the Pathological Response to Neoadjuvant Chemotherapy in Breast Cancer Patients With MRI-Radiomics: A Systematic Review and Meta-analysis

Radiomic Features Applied to Contrast Enhancement Spectral Mammography: Possibility to Predict Breast Cancer Molecular Subtypes in a Non-Invasive Manner

“…The application of radiomics to clinical practice, in particular the correlation with molecular biomarkers, represents one of the most interesting challenges for radiologists and clinicians in recent years [ 15 , 16 , 17 , 18 ]. The possibility of providing increasingly personalized and precise diagnostic and therapeutic pathways can be greatly assisted by the intrinsic radiological image information that radiomics is able to provide: radiomics is able to support the clinical decision-making process of the patient’s management.…”

How Radiomics Can Improve Breast Cancer Diagnosis and Treatment

“…All these percutaneous procedures also allow the placement of a harmless lesion marker, which is useful for the recognition of the target lesion during surgery or after neoadjuvant therapy (NAT—chemotherapy performed before surgery). The NAT may lead to resolution of the tumour without detectable mass, and lesion markers may be helpful to confirm complete treatment response [ 16 , 18 , 19 ].…”

“…The recently developed AI algorithms on vast amounts of imaging data has led to satisfactory models for the application of radiomics in breast cancer care, and even the patients approve the introduction of AI in clinical practice although only as a support to radiologist, and not in substitution thereof [ 78 ]. Indeed, radiomics already showed valuable applications in breast imaging practice: it may distinguish between malignant and benign lesions, assess the tumour subtype and its grade, assess the molecular expressions, and predict response to therapy and the risk of recurrence [ 5 , 9 , 16 , 17 , 18 , 20 ]. With the ability to infer the molecular profile of the tumour, a specific mutation or genotype, or even defining treatment possibilities and prognosis in BC patients, radiomics data may substitute physical breast biopsies in the near future [ 9 , 17 , 20 ].…”

“…Continued exponential growth of medical imaging has led to an advancement in radiomics, which provides countless quantitative biomarkers extracted from modern diagnostic images, including detailed tumoral characterization of BC [ 9 ]. In particular, recent technological advances in the field of artificial intelligence (AI) applied to image analysis, via software based on machine learning (ML) and deep learning (DL), hold promise in addressing medical challenges in cancer detection, treatment assessment, prediction of treatment response and monitoring of disease progression [ 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 ]. In the current breast imaging practice, evaluation of BC is largely qualitative, including subjective evaluations such as tumour morphology, enhancement curves, and anatomic relationship to the surrounding tissues.…”

“…Therefore, radiomics is an emerging field of extreme interest, dealing with quantitative evaluation of images and extraction of designated features [ 21 , 22 , 23 ]. Data derived from a radiomics investigation, such as intensity, shape, texture, and wavelength [ 22 , 24 , 25 , 26 , 27 ], can be input to ML or DL algorithms, providing information to differentiate malignant and benign tumours, assess cancer genetics, predict treatment response, and contribute to more robust models that combine multidisciplinary information [ 11 , 16 , 22 , 28 , 29 , 30 ].…”