Open domain question answering (OpenQA) tasks have been recently attracting more and more attention from the natural language processing (NLP) community. In this work, we present the first free-form multiple-choice OpenQA dataset for solving medical problems, MedQA, collected from the professional medical board exams. It covers three languages: English, simplified Chinese, and traditional Chinese, and contains 12,723, 34,251, and 14,123 questions for the three languages, respectively. We implement both rule-based and popular neural methods by sequentially combining a document retriever and a machine comprehension model. Through experiments, we find that even the current best method can only achieve 36.7%, 42.0%, and 70.1% of test accuracy on the English, traditional Chinese, and simplified Chinese questions, respectively. We expect MedQA to present great challenges to existing OpenQA systems and hope that it can serve as a platform to promote much stronger OpenQA models from the NLP community in the future.

Upconversion nanoparticles (UCNPs) have been widely employed for tumor imaging using magnetic resonance imaging (MRI) and upconversion luminescence (UCL) imaging.

HIGHLIGHTS• A biomimetic nanoprobe was built with cancer cell membrane-coated and Gd 3+ -doped upconversion nanoparticles.• The nanoprobe could be applied to in vivo UCL/MRI/PET multimodality precise imaging and successfully differentiated MDA-MB-231 tumor models through in vivo tri-modality imaging, which may be used for breast cancer molecular classification.ABSTRACT Triple-negative breast cancer (TNBC) is a subtype of breast cancer in which the estrogen receptor and progesterone receptor are not expressed, and human epidermal growth factor receptor 2 is not amplified or overexpressed either, which make the clinical diagnosis and treatment very challenging. Molecular imaging can provide an effective way to diagnose TNBC. Upconversion nanoparticles (UCNPs), are a promising new generation of molecular imaging probes. However, UCNPs still need to be improved for tumor-targeting ability and biocompatibility.This study describes a novel probe based on cancer cell membrane-coated upconversion nanoparticles (CCm-UCNPs), owing to the low immunogenicity and homologous-targeting ability of cancer cell membranes, and modified multifunctional UCNPs. This probe exhibits excellent performance in breast cancer molecular classification and TNBC diagnosis through UCL/MRI/PET tri-modality imaging in vivo. By using this probe, MDA-MB-231 was successfully differentiated between MCF-7 tumor models in vivo. Based on the tumor imaging and molecular classification results, the probe is also expected to be modified for drug delivery in the future, contributing to the treatment of TNBC. The combination of nanoparticles with biomimetic cell membranes has the potential for multiple clinical applications.