Microcalcifications (MCs) are routinely used to detect breast cancer in mammography. Little is known, however, about their materials properties and associated organic matrix, or their correlation to breast cancer prognosis. We combine histopathology, Raman microscopy, and electron microscopy to image MCs within snap-frozen human breast tissue and generate micron-scale resolution correlative maps of crystalline phase, trace metals, particle morphology, and organic matrix chemical signatures within high grade ductal carcinoma in situ (DCIS) and invasive cancer. We reveal the heterogeneity of mineral-matrix pairings, including punctate apatitic particles (<2 µm) with associated trace elements (e.g., F, Na, and unexpectedly Al) distributed within the necrotic cores of DCIS, and both apatite and spheroidal whitlockite particles in invasive cancer within a matrix containing spectroscopic signatures of collagen, non-collagen proteins, cholesterol, carotenoids, and DNA. Among the three DCIS samples, we identify key similarities in MC morphology and distribution, supporting a dystrophic mineralization pathway. This multimodal methodology lays the groundwork for establishing MC heterogeneity in the context of breast cancer biology, and could dramatically improve current prognostic models.

Vertical cavity surface emitting lasers (VCSELs) emitting at 850 nm have experienced explosive growth in the past decade because of their many attractive optical features and incredibly low-cost manufacturability. This review reviews the foundations for GaAs-based VCSEL technology as well as the materials and device challenges to extend the operating wavelength to both shorter and longer wavelengths. We discuss some of the applications that are enabled by the integration of VCSELs with both active and passive semiconductor elements for telecommunications, both in vivo and in vitro biosensing, high-density optical storage and imaging at wavelengths much less than the diffraction limit of light.

No abstract

From machine learning (ML) and computer vision to robotics and natural language processing, the application of data science and artificial intelligence (AI) is expected to transform health care (Celi et al. 2019). While the rapid development of technological capabilities offers paths toward new discoveries and large-scale analysis, numerous critical ethical issues have been identified, spanning privacy, data protection, transparency and explainability, responsibility, and bias.Last year, for instance, a commercial prediction algorithm affecting millions of patients was shown to exhibit significant racial bias, dramatically underestimating the health needs of Black patients (Obermeyer et al. 2019). Trained using health care cost as the proxy for the need for more comprehensive care, the algorithm had been designed specifically to exclude race as a feature, in an attempt to avoid bias-but cost was clearly not a race-neutral measure of health care need. Studies have repeatedly illuminated racial disparities in the provision of primary care services: Black patients incur approximately US$1800 less in medical costs per year compared to white patients with the same number of chronic conditions, and are less likely to be identified as high-risk for complex care in the future. But even if another proxy, such as probability of death, had been used to train the algorithm, would it have led to a "better" algorithm and fair patient outcomes? At present, a key evaluation metric for machine learning in health care applications is accuracy. To inspect an algorithm for bias, an additional step is often undertaken to measure performance across different subpopulations, aiming for consistent accuracy across race, gender, country, and other categories where disparities exist. But just because an algorithm is deemed accurate does not mean it will support fairness in health care applications. In an ideal world, only individual patient health and disease factors would determine-and guide prediction of-clinical outcomes. However, studies have repeatedly demonstrated that this is far from the case. For example, mortality from critical illness has been shown to be higher in ✉

Abstract-Intrinsic Optical Signal (IOS) imaging is a widely accepted technique for imaging brain activity. We propose an integrated device consisting of interleaved arrays of gallium arsenide (GaAs) based semiconductor light sources and detectors operating at telecommunications wavelengths in the near-infrared. Such a device will allow for long-term, minimally invasive monitoring of neural activity in freely behaving subjects, and will enable the use of structured illumination patterns to improve system performance. In this work we describe the proposed system and show that nearinfrared IOS imaging at wavelengths compatible with semiconductor devices can produce physiologically significant images in mice, even through skull. I. INTRODUCTIONptical imaging of neural activity is a widely accepted technique for imaging brain function in the field of neuroscience research, and has been used to study the cerebral cortex for nearly two decades [1]. Maps of brain activity are obtained by monitoring intensity changes in back-scattered light, called Intrinsic Optical Signals (IOS) that correspond to fluctuations in blood oxygenation and volume associated with neural activity. Current imaging systems typically employ benchtop equipment including lamps and CCD cameras to study animals using visible light. Such systems require the use of anesthetized or immobilized subjects with craniotomies, which imposes limitations on the Thomas T. Lee is with the