http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
윤현조,정형은,안하림,강상율,정성후 대한외과초음파학회 2023 대한외과초음파학회지 Vol.10 No.1
Breast cancer is the most commonly diagnosed cancer in women and its incidence and the mortality associated with it have increased over the years. Early detection of breast cancer via various imaging modalities can significantly improve the prognosis of patients. Ultrasound is a useful imaging tool for breast lesion characterization due to its acceptable diagnostic performance and non-invasive and real-time capabilities. However, one of the major drawbacks of ultrasound imaging is operator dependence. Artificial intelligence (AI), particularly deep learning, is gaining extensive attention for its excellent performance in image recognition. AI can make a quantitative assessment by recognizing imaging information, thereby improving ultrasound performance in the diagnosis of breast cancer lesions. The use of AI for breast ultrasound in clinical practice is extremely important, as it saves time, reduces radiologist fatigue, and compensates for a lack of experience and skill in some cases. This review article discusses the basic technical knowledge required, the algorithms of AI for breast ultrasound, and the application of AI in image identification, segmentation, extraction, and classification. In addition, we also discuss the future perspectives for the application of AI in breast ultrasound.
윤현조,정형은,안하림,강상율,정성후 대한외과초음파학회 2023 대한외과초음파학회지 Vol.10 No.2
Thyroid nodules are a common medical concern with an increasing prevalence due largely to the widespread use of ultrasonography (US). US is the primary diagnostic tool for identifying thyroid nodules, determining their size and characteristics, guiding fine-needle aspiration (FNA), and diagnosing lymph node metastasis. The most consistent US features indicating a malignancy in thyroid nodules include spiculated margins, microcalcifications, a taller-than-wide shape, and pronounced hypoechogenicity. Based on these imaging findings, clinicians must determine which nodules require further evaluation and which can be monitored over time. Despite the central role of thyroid US in managing these nodules, there is no clear consensus regarding which nodules should undergo US-guided FNA, nor is there a standardized terminology for describing the US features. Consequently, medical societies recommend using US risk stratification systems to standardize reporting and simplify clinical decision-making. This review aims to provide a comprehensive overview of the various US risk stratification systems employed for thyroid nodules, allowing readers to gain a better understanding of how nodules are assessed using these different systems. Furthermore, this study examines the efficacy, weaknesses, and potential pitfalls of these US risk stratification systems and discusses future perspectives for their optimal clinical application in managing thyroid nodules.