Breast cancer is the most common cancer in women worldwide, and its early detection is critical for improving survival outcomes. As a diagnostic and screening tool, mammography can be less effective owing to the masking effect of fibroglandular tissue, but breast US has good sensitivity even in dense breasts. However, breast US is highly operator dependent, highlighting the need for artificial intelligence (AI)-driven solutions. Unlike other modalities, US is performed using a handheld device that produces a continuous real-time video stream, yielding 12000–48000 frames per examination. This can be significantly challenging for AI development and requires real-time AI inference capabilities. In this review, we classified AI solutions as computer-aided diagnosis and computer-aided detection to facilitate a functional understanding and review commercial software supported by clinical evidence.In addition, to bridge healthcare gaps and enhance patient outcomes in geographically under resourced areas, we propose a novel framework by reviewing the existing AI-based triage workflows including mobile ultrasound.

Purpose: Endoscopic nipple-sparing mastectomy (E-NSM) is a minimally invasive surgical technique that shows good results in patients with breast cancer. The authors compared 3 different types of commercial energy devices to examine their efficacy and safety in E-NSM performed with breast reconstruction. Methods: A total of 36 cases of E-NSM were conducted with either Sonicision (S group, n = 11), Harmonic (H group, n = 6), or Thunderbeat (T group, n = 19). The clinicopathologic factors and postoperative complications, including nipple or skin necrosis and surgical site seroma volume, were evaluated for 3 months after surgery. Results: The surgical duration of E-NSM was significantly shorter in the S group than in the H group (P = 0.043) and T group (P = 0.037). However, the total surgical duration including E-NSM and breast reconstruction, and the total and daily drainage volume of postoperative seroma did not differ significantly among the 3 groups. Even when the energy devices were compared according to their working principle, i.e., ultrasonic (S and H) vs. hybrid (T), the total breast surgery duration and total and daily drainage volume of seroma showed no difference between the 2 groups. Although surgeon satisfaction did not significantly differ when using 3 devices for E-NSM (P = 0.428), surgeon’s fatigue was found to be lowest in the S group, though it was not significant (P = 0.064). Conclusion Any energy device can be safely used for E-NSM with breast reconstruction without causing any major complications. However, cordless ultrasonic energy devices allow greater mobility for the surgeon and, therefore, may shorten surgical time in breast surgery.

Breast cancer is the most common cancer in women worldwide, and its early detection is critical for improving survival outcomes. As a diagnostic and screening tool, mammography can be less effective owing to the masking effect of fibroglandular tissue, but breast US has good sensitivity even in dense breasts. However, breast US is highly operator dependent, highlighting the need for artificial intelligence (AI)-driven solutions. Unlike other modalities, US is performed using a handheld device that produces a continuous real-time video stream, yielding 12000–48000 frames per examination. This can be significantly challenging for AI development and requires real-time AI inference capabilities. In this review, we classified AI solutions as computer-aided diagnosis and computer-aided detection to facilitate a functional understanding and review commercial software supported by clinical evidence.In addition, to bridge healthcare gaps and enhance patient outcomes in geographically under resourced areas, we propose a novel framework by reviewing the existing AI-based triage workflows including mobile ultrasound.

Breast cancer is the most common cancer in women worldwide, and its early detection is critical for improving survival outcomes. As a diagnostic and screening tool, mammography can be less effective owing to the masking effect of fibroglandular tissue, but breast US has good sensitivity even in dense breasts. However, breast US is highly operator dependent, highlighting the need for artificial intelligence (AI)-driven solutions. Unlike other modalities, US is performed using a handheld device that produces a continuous real-time video stream, yielding 12000–48000 frames per examination. This can be significantly challenging for AI development and requires real-time AI inference capabilities. In this review, we classified AI solutions as computer-aided diagnosis and computer-aided detection to facilitate a functional understanding and review commercial software supported by clinical evidence.In addition, to bridge healthcare gaps and enhance patient outcomes in geographically under resourced areas, we propose a novel framework by reviewing the existing AI-based triage workflows including mobile ultrasound.

Recently, several breast surgeons have reported a new method for sentinel lymph node biopsy (SLNB) by using indocyanine green (ICG) with infrared camera. This study aimed to determine whether the lymph nodes (LNs) with ICG uptake are true SLNs and to assess the reliability of using only ICG for SLNB. Data were prospectively collected between April and September 2021. All palpable LNs were fat-trimmed and ordered from high to low signal of the gamma detector. The degree of radioisotope uptake and brightness of ICG staining of the axillary LNs detected with a fluorescent camera were compared and associated factors were analyzed. Discordance was defined as sentinel LNs (SLNs) showing a single uptake of radioisotope or fluorescence of ICG only, or when the orders of uptake and intensity degree were different between the 2 materials. A total of 79 SLNBs were performed on 78 patients with breast cancer. The breast cancer was classified as cTis-2N0-1. The discordance rate was 14/79 (17.7%) overall and 45/270 (16.7%) of the total retrieved axillary LNs. The first SLNs showed the lowest discordance rate of 6.3%, whereas the second and third SLNs showed higher discordance rates of 27.6% and 60.0%, respectively. There were no associated clinicopathologic factors that affected the discordance between uptake of radioisotope and fluorescence intensity of ICG. The use of ICG alone for SLNB may be insufficient because of the high discordance rates between radioisotopes and ICG uptake. However, the first SLN could be cautiously regarded as a true SLN.