Cases of delayed presentation of breast cancer underscore intricate barriers impeding access to health care. Among those affected are caregivers, who often face unique barriers. Advanced cases occur despite progress in early breast cancer diagnosis, presenting an emotional paradox for breast imagers. This article addresses some challenges encountered by female caregivers while exploring strategies to reduce the health care gap for this vulnerable population.
IV contrast media improve the diagnostic power of radiology examinations. These media include gadolinium-based contrast media and iron-oxide nanoparticles for MRI, iodinated contrast material for CT, microbubbles for ultrasound, and radiopharmaceuticals for nuclear medicine. As for all medications, contrast media carry risks, which may be heightened in the conditions of pregnancy and lactation. Radiologists must understand the potential risks from contrast media exposure to the pregnant patient, fetus, and nursing infant, as well as understand these administrations' impact on examinations' clinical utility. This article reviews the available literature on these media, along with key regulatory bodies' and professional societies' current recommendations for their use, in the settings of pregnancy and lactation. This knowledge should help radiologists make well-reasoned risk-benefit analyses regarding contrast media administration and provide informed consent discussions with pregnant and nursing patients in whom contrast media administration is being considered. This information and analysis can also assist facilities in designing policies and standard operating procedures of possible clinical benefit to the pregnant patient, fetus, or nursing infant, balancing contrast media exposure considerations versus augmented diagnostic capabilities.
Interpretive artificial intelligence (AI) tools are poised to change the future of radiology. However, certain pitfalls may pose particular challenges for optimal AI interpretative performance. These include anatomic variants, age-related changes, postoperative changes, medical devices, image artifacts, lack of integration of prior and concurrent imaging examinations and clinical information, as well as the satisfaction-of-search effect. Model training and development should account for such pitfalls, to minimize errors and optimize interpretation accuracy. More broadly, AI algorithms should be exposed to diverse and complex training data sets, to yield a holistic interpretation that considers all relevant information beyond the individual examination. Successful clinical deployment of AI tools will require that radiologist end-users recognize these pitfalls and other limitations of the available models. Furthermore, developers should incorporate explainable AI techniques (e.g., heat maps) into their tools, to improve radiologists' understanding of model outputs and to enable radiologists to provide feedback for guiding continuous learning and iterative refinement. In this article, we provide an overview of common pitfalls that radiologists may encounter when using interpretive AI products in daily practice. We present how such pitfalls lead to AI errors and offer potential strategies that AI developers may use for their mitigation.
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：