Breast cancer is among the most prevalent and lethal tumors globally. Approximately 20% of all breast cancers are classified as triple-negative (TNBC). Triple-negative breast cancer (TNBC) is generally linked to a worse prognosis compared to other breast cancer subtypes. Conventional cytotoxic chemotherapy is the standard treatment due to its aggressiveness and resistance to hormone therapy; nevertheless, this approach is not consistently effective, and a significant proportion of patients experience recurrence. Recently, immunotherapy has been employed in certain populations with TNBC, demonstrating encouraging outcomes. Regrettably, immunotherapy is applicable to only a small subset of patients, and the responses in metastatic triple-negative breast cancer have been rather moderate compared to other cancer types. This situation demonstrates the necessity for the development of effective biomarkers to stratify and personalize patient care.Recent advancements in artificial intelligence (AI) have generated heightened interest in its application for medical purposes, particularly in enhancing clinical decision-making. Numerous studies have employed AI with diagnostic medical imaging, particularly in radiography and digitized histopathology tissue samples, with the objective of extracting disease-specific information that is challenging for the human eye to assess. These studies have shown that the examination of such pictures in the context of TNBC holds significant promise for (1) risk stratification of patients to identify those at higher likelihood of disease recurrence or mortality and (2) predicting pathological full response. This publication provides an overview of AI and its integration with radiology and histopathology imaging to develop prognostic and predictive methods for TNBC. We delineate cutting-edge methodologies in the literature and examine the opportunities and challenges associated with the advancement and clinical implementation of AI algorithms. This includes discerning patients who may benefit from specific treatments, such as adjuvant chemotherapy, versus those who should be redirected to alternative therapies, uncovering potential disparities among populations, and identifying disease subtypes.
