Deep learning has been shown to be able to recognize data patterns better than humans in specific circumstances or contexts. In parallel, quantum computing has demonstrated to be able to output complex wave functions with a few number of gate operations, which could generate distributions that are hard for a classical computer to produce. Here we propose a hybrid quantum-classical convolutional neural network (QCCNN), inspired by convolutional neural networks (CNNs) but adapted to quantum computing to enhance the feature mapping process. QCCNN is friendly to currently noisy intermediate-scale quantum computers, in terms of both number of qubits as well as circuit's depths, while retaining important features of classical CNN, such as nonlinearity and scalability. We also present a framework to automatically compute the gradients of hybrid quantum-classical loss functions which could be directly applied to other hybrid quantum-classical algorithms. We demonstrate the potential of this architecture by applying it to a Tetris dataset, and show that QCCNN can accomplish classification tasks with learning accuracy surpassing that of classical CNN with the same structure.

No abstract

No abstract

Tumor-selective contrast agents have the potential to aid in the diagnosis and treatment of cancer using noninvasive imaging modalities such as magnetic resonance imaging (MRI). Such contrast agents can consist of magnetic nanoparticles incorporating functionalities that respond to cues specific to tumor environments. Genetically engineering magnetotactic bacteria to display peptides has been investigated as a means to produce contrast agents that combine the robust image contrast effects of magnetosomes with the transgenic-targeting peptides displayed on their surface. This work reports the first use of magnetic nanoparticles that display genetically encoded pH low insertion peptide (pHLIP), a long peptide intended to enhance MRI contrast by targeting the extracellular acidity associated with the tumors. To demonstrate the modularity of this versatile platform to incorporate diverse targeting ligands by genetic engineering, we also incorporated the cyclic αv integrin-binding peptide iRGD into separate magnetosomes. Specifically, we investigate their potential for enhanced binding and tumor imaging both in vitro and in vivo. Our experiments indicate that these tailored magnetosomes retain their magnetic properties, making them well suited as T2 contrast agents, while exhibiting an increased binding compared to the binding in wild-type magnetosomes.

Objectives. To examine what changes are caused in the activity of the vastus medialis oblique (VMO) and vastus lateralis (VL) at the time of sling-based exercises in patients with patellofemoral pain syndrome (PFPS) and compare the muscular activations in patients with PFPS among the sling-based exercises. Methods. This was a cross-over study. Sling-based open and closed kinetic knee extension and hip adduction exercises were designed for PFPS, and electromyography was applied to record maximal voluntary contraction during the exercises. The VMO and VL activations and VMO : VL ratios for the three exercises were analyzed and compared. Results. Thirty male (age = 21.19 ± 0.68 y) and 30 female (age = 21.12 ± 0.74 y) patients with PFPS were recruited. VMO activations during the sling-based open and closed kinetic knee extension exercises were significantly higher (P = 0.04 and P = 0.001) than those during hip adduction exercises and VMO : VL ratio for the sling-based closed kinetic knee extension and hip adduction exercises approximated to 1. Conclusions. The sling-based closed kinetic knee extension exercise produced the highest VMO activation. It also had an appropriate VMO : VL ratio similar to sling-based hip adduction exercise and had beneficial effects on PFPS.