Robust breast cancer detection in mammography and digital breast tomosynthesis using an annotation-efficient deep learning approach.

Breast cancer remains a global challenge, causing over 600,000 deaths in 2018 (ref. 1). To achieve earlier cancer detection, health organizations worldwide recommend screening mammography, which is estimated to decrease breast cancer mortality by 20-40% (refs. 2,3). Despite the clear value of screening mammography, significant false positive and false negative rates along with non-uniformities in expert reader availability leave opportunities for improving quality and access4,5. To address these limitations, there has been much recent interest in applying deep learning to mammography6-18, and these efforts have highlighted two key difficulties: obtaining large amounts of annotated training data and ensuring generalization across populations, acquisition equipment and modalities. Here we present an annotation-efficient deep learning approach that (1) achieves state-of-the-art performance in mammogram classification, (2) successfully extends to digital breast tomosynthesis (DBT; '3D mammography'), (3) detects cancers in clinically negative prior mammograms of patients with cancer, (4) generalizes well to a population with low screening rates and (5) outperforms five out of five full-time breast-imaging specialists with an average increase in sensitivity of 14%. By creating new 'maximum suspicion projection' (MSP) images from DBT data, our progressively trained, multiple-instance learning approach effectively trains on DBT exams using only breast-level labels while maintaining localization-based interpretability. Altogether, our results demonstrate promise towards software that can improve the accuracy of and access to screening mammography worldwide.

Vagus nerve stimulation reduces infarct size in rat focal cerebral ischemia.

BACKGROUND AND PURPOSE: We sought to determine the effect of vagus nerve stimulation (VNS) on infarct size after transient focal cerebral ischemia in rats. METHODS: Ischemia was produced by transient filament occlusion of the right middle cerebral artery. Stimulating electrodes were implanted on the cervical part of the right vagus nerve. Electrical stimulation was initiated 30 min after the induction of ischemia, and delivered for 30s at every 30 min for 3h in experimental group 1 and at every 5 min for 1h in experimental group 2. All the procedures were duplicated but no stimulus was delivered in the control group. Functional deficit was evaluated and animals were killed to determine the infarct size 24h after ischemia. RESULTS: Ischemic lesion volume was smaller in VNS-treated animals as compared with control animals; the relative percentage of contralateral hemispheric volume that underwent infarction was 16.2+/-3.2% in the VNS and 33.0+/-5.0% in the control arms in experimental group 1 (p<0.05). The respective values for experimental group 2 were 19.8+/-0.5% and 37.9+/-2.6% (p<0.05). VNS-treated animals were significantly more likely to have better functional scores at 24h as compared with control animals. The functional score improved by 50% in experimental group 1 and 44% in experimental group 2 (p<0.05 for both groups). CONCLUSION: VNS appears to offer protection against acute ischemic brain injury.