Artificial intelligence system reduces false-positive findings in the interpretation of breast ultrasound exams.

Though consistently shown to detect mammographically occult cancers, breast ultrasound has been noted to have high false-positive rates. In this work, we present an AI system that achieves radiologist-level accuracy in identifying breast cancer in ultrasound images. Developed on 288,767 exams, consisting of 5,442,907 B-mode and Color Doppler images, the AI achieves an area under the receiver operating characteristic curve (AUROC) of 0.976 on a test set consisting of 44,755 exams. In a retrospective reader study, the AI achieves a higher AUROC than the average of ten board-certified breast radiologists (AUROC: 0.962 AI, 0.924 ± 0.02 radiologists). With the help of the AI, radiologists decrease their false positive rates by 37.3% and reduce requested biopsies by 27.8%, while maintaining the same level of sensitivity. This highlights the potential of AI in improving the accuracy, consistency, and efficiency of breast ultrasound diagnosis.

Prophylaxis of hemodialysis graft thrombosis with fish oil: double-blind, randomized, prospective trial.

Diets enriched with fish oil may favorably affect the vascular perturbations underlying synthetic graft thrombosis. Therefore, these studies were designed to test the hypothesis that diets enriched with fish oil would decrease the incidence of thrombosis in newly constructed polytetrafluorethylene grafts. A double-blind, randomized trial was conducted. Twenty-four patients were randomized to receive 4000 mg of fish oil or 4000 mg of control oil. Both preparations were enriched with antioxidants and deodorized with peppermint. Patients began therapy within 2 wk after graft placement and were monitored for 12 mo or until thrombosis developed. With a permuted-block randomization schedule, 12 patients received fish oil and 12 patients received control oil. The primary patency rates at 365 d were 14.9% for the control group and 75.6% for the fish oil-treated group. Survival analysis revealed a significant difference between fish oil-treated and untreated patients (P < 0.03, Mantel-Cox test), with a power of 90%. Moreover, analysis of covariables, including age of > or =50 yr, gender, race, body weight, diabetes mellitus, bleeding times, and lipid profiles, indicated that this effect occurred principally as a result of fish oil administration. Importantly, fish oil treatment also decreased venous outflow resistance and systemic BP, compared with control values. Fish oils possess unique biologic properties that favorably affect the incidence of polytetrafluorethylene graft thrombosis, and they thus represent a potential treatment strategy for the prevention of access thrombosis.