Privacy preservation for federated learning in health care.

Artificial intelligence (AI) shows potential to improve health care by leveraging data to build models that can inform clinical workflows. However, access to large quantities of diverse data is needed to develop robust generalizable models. Data sharing across institutions is not always feasible due to legal, security, and privacy concerns. Federated learning (FL) allows for multi-institutional training of AI models, obviating data sharing, albeit with different security and privacy concerns. Specifically, insights exchanged during FL can leak information about institutional data. In addition, FL can introduce issues when there is limited trust among the entities performing the compute. With the growing adoption of FL in health care, it is imperative to elucidate the potential risks. We thus summarize privacy-preserving FL literature in this work with special regard to health care. We draw attention to threats and review mitigation approaches. We anticipate this review to become a health-care researcher's guide to security and privacy in FL.

Caffeine and Cisplatin Effectively Targets the Metabolism of a Triple-Negative Breast Cancer Cell Line Assessed via Phasor-FLIM.

Triple-negative tumor cells, a malignant subtype of breast cancer, lack a biologically targeted therapy. Given its DNA repair inhibiting properties, caffeine has been shown to enhance the effectiveness of specific tumor chemotherapies. In this work, we have investigated the effects of caffeine, cisplatin, and a combination of the two as potential treatments in energy metabolism for three cell lines, triple-negative breast cancer (MDA-MB-231), estrogen-receptor lacking breast cancer (MCF7) and breast epithelial cells (MCF10A) using a sensitive label-free approach, phasor-fluorescence lifetime imaging microscopy (phasor-FLIM). We found that solely using caffeine to treat MDA-MB-231 shifts their metabolism towards respiratory-chain phosphorylation with a lower ratio of free to bound NADH, and a similar trend is seen in MCF7. However, MDA-MB-231 cells shifted to a higher ratio of free to bound NADH when cisplatin was added. The combination of cisplatin and caffeine together reduced the survival rate for MDA-MD231 and shifted their energy metabolism to a higher fraction of bound NADH indicative of oxidative phosphorylation. The FLIM and viability results of MCF10A cells demonstrate that the treatments targeted cancer cells over the normal breast tissue. The identification of energy metabolism alteration could open up strategies of improving chemotherapy for malignant breast cancer.