3D printing allows rapid fabrication of complex objects from digital designs. One 3D-printing process, direct laser writing, polymerises a light-sensitive material by steering a focused laser beam through the shape of the object to be created. The highest-resolution direct laser writing systems use a femtosecond laser, steered using mechanised stages or galvanometer-controlled mirrors, to effect two-photon polymerisation. Here we report a new high-resolution direct laser writing system that employs a resonant mirror scanner to achieve a significant increase in printing speed over current methods while maintaining resolution on the order of a micron. This printer is based on a software modification to a commercially available resonant-scanning two-photon microscope. We demonstrate the complete process chain from hardware configuration and control software to the printing of objects of approximately 400 × 400 × 350 μm, and validate performance with objective benchmarks. Released under an open-source license, this work makes micron-scale 3D printing available at little or no cost to the large community of two-photon microscope users, and paves the way toward widespread availability of precision-printed devices.

Objective . Most preparations for making neural recordings degrade over time and eventually fail due to insertion trauma and reactive tissue response. The magnitudes of these responses are thought to be related to the electrode size (specifically, the cross-sectional area), the relative stiffness of the electrode, and the degree of tissue tolerance for the material. Flexible carbon fiber ultra-microelectrodes have a much smaller cross-section than traditional electrodes and low tissue reactivity, and thus may enable improved longevity of neural recordings in the central and peripheral nervous systems. Only two carbon fiber array designs have been described previously, each with limited channel densities due to limitations of the fabrication processes or interconnect strategies. Here, we describe a method for assembling carbon fiber electrodes on a flexible polyimide substrate that is expected to facilitate the construction of high-density recording and stimulating arrays. Approach . Individual carbon fibers were aligned using an alignment tool that was 3D-printed with sub-micron resolution using direct laser writing. Indium deposition on the carbon fibers, followed by low-temperature microsoldering, provided a robust and reliable method of electrical connection to the polyimide interconnect. Main results . Spontaneous multiunit activity and stimulation-evoked compound responses with SNR >10 and >120, respectively, were recorded from a small (125 µ m) peripheral nerve. We also improved the typically poor charge injection capacity of small diameter carbon fibers by electrodepositing 100 nm-thick iridium oxide films, making the carbon fiber arrays usable for electrical stimulation as well as recording. Significance . Our innovations in fabrication technique pave the way for further miniaturization of carbon fiber ultra-microelectrode arrays. We believe these advances to be key steps to enable a shift from labor intensive, manual assembly to a more automated manufacturing process.

Objective . The vision of bioelectronic medicine is to treat disease by modulating the signaling of visceral nerves near various end organs. In small animal models, the nerves of interest can have small diameters and limited surgical access. New high-resolution methods for building nerve interfaces are desirable. In this study, we present a novel nerve interface and demonstrate its use for stimulation and recording in small nerves. Approach . We design and fabricate micro-scale electrode-laden nanoclips capable of interfacing with nerves as small as 50 µ m in diameter. The nanoclips are fabricated using a direct laser writing technique with a resolution of 200 nm. The resolution of the printing process allows for incorporation of a number of innovations such as trapdoors to secure the device to the nerve, and quick-release mounts that facilitate keyhole surgery, obviating the need for forceps. The nanoclip can be built around various electrode materials; here we use carbon nanotube fibers for minimally invasive tethering. Main results . We present data from stimulation-evoked responses of the tracheal syringeal (hypoglossal) nerve of the zebra finch, as well as quantification of nerve functionality at various time points post implant, demonstrating that the nanoclip is compatible with healthy nerve activity over sub-chronic timescales. Significance . Our nerve interface addresses key challenges in interfacing with small nerves in the peripheral nervous system. Its small size, ability to remain on the nerve over sub-chronic timescales, and ease of implantation, make it a promising tool for future use in the treatment of disease.

The song of the adult male zebra finch is strikingly stereotyped. Efforts to understand motor output, pattern generation, and learning have taken advantage of this consistency by investigating the bird’s ability to modify specific parts of song under external cues, and by examining timing relationships between neural activity and vocal output. Such experiments require that precise moments during song be identified in real time as the bird sings. Various syllable-detection methods exist, but many require special hardware, software, and know-how, and details on their implementation and performance are scarce. We present an accurate, versatile, and fast syllable detector that can control hardware at precisely timed moments during zebra finch song. Many moments during song can be isolated and detected with false negative and false positive rates well under 1% and 0.005% respectively. The detector can run on a stock Mac Mini with triggering delay of less than a millisecond and a jitter of σ ≈ 2 milliseconds.

Networks consisting of stationary data sources and sinks linked by mobile ferries allow high-latency communication over large distances. Network performance depends on the trajectory of the ferry, which usually should aim to take the shortest path that allows it to exchange all waiting data with each node. An accurate model of the system would permit optimal trajectory planning, but such a model is hard to discover. Consequently, extant ferry flight planners make simplistic assumptions about data rates, ignoring signal reflections and occlusions from aircraft and ground, noise sources, and aircraft dynamics. We present an approach based on reinforcement learning, which allows a ferry to learn to improve its performance while in service, obviating considerable effort in system identification. We compare hand-coded and learned trajectories through radio fields with a complex data rate model, and show that learning trajectory planners can quickly discover good flight paths, outperforming a standard handcoded policy.