Summary Paragraph The fast-growing field of bioelectronic medicine aims to develop engineered systems that relieve clinical conditions through stimulation of the peripheral nervous system (PNS) 1 – 5 . Technologies of this type rely largely on electrical stimulation to provide neuromodulation of organ function or pain. One example is sacral nerve stimulation to treat overactive bladder, urinary incontinence and interstitial cystitis/bladder pain syndrome 4 , 6 , 7 . Conventional, continuous stimulation protocols, however, cause discomfort and pain, particularly when treating symptoms that can be intermittent in nature (e.g. sudden urinary urgency) 8 . Direct physical coupling of electrodes to the nerve can lead to injury and inflammation 9 – 11 . Furthermore, typical therapeutic stimulators target large nerve bundles that innervate multiple structures, resulting in a lack of organ specificity. This paper introduces a miniaturized bio-optoelectronic implant that avoids these limitations, via the use of (1) an optical stimulation interface that exploits microscale inorganic light emitting diodes (μ-ILEDs) to activate opsins, (2) a soft, precision biophysical sensor system that allows continuous measurements of organ function, and (3) a control module and data analytics approach that allows coordinated, closed-loop operation of the system to eliminate pathological behaviors as they occur in real-time. In an example reported here, a soft strain gauge yields real-time information on bladder function. Data analytics algorithms identify pathological behavior, and automated, closed-loop optogenetic neuromodulation of bladder sensory afferents normalize bladder function in the context of acute cystitis. This all-optical scheme for neuromodulation offers chronic stability and the potential for cell-type-specific stimulation.

No abstract

Extensive planting of crops genetically engineered to produce insecticidal proteins from the bacterium Bacillus thuringiensis (Bt) has suppressed some major pests, reduced insecticide sprays, enhanced pest control by natural enemies, and increased grower profits. However, rapid evolution of resistance in pests is reducing these benefits. Better understanding of the genetic basis of resistance to Bt crops is urgently needed to monitor, delay, and counter pest resistance. We discovered that a point mutation in a previously unknown tetraspanin gene in the cotton bollworm (Helicoverpa armigera), a devastating global pest, confers dominant resistance to Cry1Ac, the sole Bt protein produced by transgenic cotton planted in China. We found the mutation using a genome-wide association study, followed by fine-scale genetic mapping and DNA sequence comparisons between resistant and susceptible strains. CRISPR/Cas9 knockout of the tetraspanin gene restored susceptibility to a resistant strain, whereas inserting the mutation conferred 125-fold resistance in a susceptible strain. DNA screening of moths captured from 23 field sites in six provinces of northern China revealed a 100-fold increase in the frequency of this mutation, from 0.001 in 2006 to 0.10 in 2016. The correspondence between the observed trajectory of the mutation and the trajectory predicted from simulation modeling shows that the dominance of the mutation accelerated adaptation. Proactive identification and tracking of the tetraspanin mutation demonstrate the potential for genomic analysis, gene editing, and molecular monitoring to improve management of resistance.

Convolutional neural networks (CNNs) have been the de facto standard for nowadays 3D medical image segmentation. The convolutional operations used in these networks, however, inevitably have limitations in modeling the long-range dependency due to their inductive bias of locality and weight sharing. Although Transformer was born to address this issue, it suffers from extreme computational and spatial complexities in processing high-resolution 3D feature maps. In this paper, we propose a novel framework that efficiently bridges a Convolutional neural network and a Transformer (CoTr) for accurate 3D medical image segmentation. Under this framework, the CNN is constructed to extract feature representations and an efficient deformable Transformer (DeTrans) is built to model the long-range dependency on the extracted feature maps. Different from the vanilla Transformer which treats all image positions equally, our DeTrans pays attention only to a small set of key positions by introducing the deformable self-attention mechanism. Thus, the computational and spatial complexities of DeTrans have been greatly reduced, making it possible to process the multi-scale and highresolution feature maps, which are usually of paramount importance for image segmentation. We conduct an extensive evaluation on the Multi-Atlas Labeling Beyond the Cranial Vault (BCV) dataset that covers 11 major human organs. The results indicate that our CoTr leads to a substantial performance improvement over other CNN-based, transformerbased, and hybrid methods on the 3D multi-organ segmentation task.

The accurate identification of malignant lung nodules on chest CT is critical for the early detection of lung cancer, which also offers patients the best chance of cure. Deep learning methods have recently been successfully introduced to computer vision problems, although substantial challenges remain in the detection of malignant nodules due to the lack of large training datasets. In this paper, we propose a multi-view knowledge-based collaborative (MV-KBC) deep model to separate malignant from benign nodules using limited chest CT data. Our model learns 3D lung nodule characteristics by decomposing a 3D nodule into nine fixed views. For each view, we construct a knowledge-based collaborative (KBC) submodel, where three types of image patches are designed to fine-tune three pre-trained ResNet-50 networks that characterize the nodules' overall appearance, voxel and shape heterogeneity, respectively. We jointly use the nine KBC submodels to classify lung nodules with an adaptive weighting scheme learned during the error back propagation, which enables the MV-KBC model to be trained in an end-to-end manner. The penalty loss function is used for better reduction of the false negative rate with a minimal effect on the overall performance of the MV-KBC model. We tested our method on the benchmark LIDC-IDRI dataset and compared it to five state-of-the-art classification approaches. Our results show that the MV-KBC model achieved an accuracy of 91.60% for lung nodule classification with an AUC of 95.70%. These results are markedly superior to the state-of-the-art approaches.