Neural inference at the frontier of energy, space, and time.

Computing, since its inception, has been processor-centric, with memory separated from compute. Inspired by the organic brain and optimized for inorganic silicon, NorthPole is a neural inference architecture that blurs this boundary by eliminating off-chip memory, intertwining compute with memory on-chip, and appearing externally as an active memory chip. NorthPole is a low-precision, massively parallel, densely interconnected, energy-efficient, and spatial computing architecture with a co-optimized, high-utilization programming model. On the ResNet50 benchmark image classification network, relative to a graphics processing unit (GPU) that uses a comparable 12-nanometer technology process, NorthPole achieves a 25 times higher energy metric of frames per second (FPS) per watt, a 5 times higher space metric of FPS per transistor, and a 22 times lower time metric of latency. Similar results are reported for the Yolo-v4 detection network. NorthPole outperforms all prevalent architectures, even those that use more-advanced technology processes.

Event-Based Vision: A Survey.

Event cameras are bio-inspired sensors that differ from conventional frame cameras: Instead of capturing images at a fixed rate, they asynchronously measure per-pixel brightness changes, and output a stream of events that encode the time, location and sign of the brightness changes. Event cameras offer attractive properties compared to traditional cameras: high temporal resolution (in the order of µs), very high dynamic range (140 dB versus 60 dB), low power consumption, and high pixel bandwidth (on the order of kHz) resulting in reduced motion blur. Hence, event cameras have a large potential for robotics and computer vision in challenging scenarios for traditional cameras, such as low-latency, high speed, and high dynamic range. However, novel methods are required to process the unconventional output of these sensors in order to unlock their potential. This paper provides a comprehensive overview of the emerging field of event-based vision, with a focus on the applications and the algorithms developed to unlock the outstanding properties of event cameras. We present event cameras from their working principle, the actual sensors that are available and the tasks that they have been used for, from low-level vision (feature detection and tracking, optic flow, etc.) to high-level vision (reconstruction, segmentation, recognition). We also discuss the techniques developed to process events, including learning-based techniques, as well as specialized processors for these novel sensors, such as spiking neural networks. Additionally, we highlight the challenges that remain to be tackled and the opportunities that lie ahead in the search for a more efficient, bio-inspired way for machines to perceive and interact with the world.

Development of colorimetric sensor array for diagnosis of tuberculosis through detection of urinary volatile organic compounds.

BACKGROUND: Top priorities for tuberculosis control and elimination include a simple, low-cost screening test using sputum and a non-sputum-based test in patients that do not produce sputum. The aim of this study was to evaluate the performance of a colorimetric sensor array (CSA) test, for analysis of volatile organic compounds in urine, in the diagnosis of pulmonary TB. MATERIAL AND METHODS: Urine samples were collected from individuals suspected of having pulmonary TB in Western Kenya. Reference methods included MGIT culture and/or Xpert MTB/RIF nucleic acid amplification test on sputa. Fresh urine samples were tested with the CSA, with acid and base and without an additive. The CSA were digitally imaged, and the resulting colorimetric response patterns were used for chemometric analysis. Sensitivity, specificity, and negative (NPV) and positive predictive (PPV) values were determined for HIV-positive and HIV-negative patients. RESULTS: In HIV-negative patients, the highest accuracy was obtained in urine samples pre-treated with a base, yielding a sensitivity, specificity, PPV, and NPV of 78.3% (65/83), 69.2% (54/78), 73.0% (n/89) and 75.0% (n/72). The highest sensitivity of 79.5% was achieved using sensor data from all three test conditions at a specificity of 65.4%. In HIV-positive subjects, the sensor performance was substantially lower with sensitivity, specificity, PPV, and NPV ranging from 48.3% to 62.3%, 54.1% to 74.0%, 55.9% to 64.2%, and 60.6% to 64.9%, respectively. CONCLUSION: The CSA fingerprint of urine headspace volatiles showed moderate accuracy in diagnosing TB in HIV-negative patients, but the sensor performance dropped substantially in HIV-coinfected patients. Further development of TB-responsive CSA indicators may improve the accuracy of CSA urine assay.

Convolutional networks for fast, energy-efficient neuromorphic computing.

Deep networks are now able to achieve human-level performance on a broad spectrum of recognition tasks. Independently, neuromorphic computing has now demonstrated unprecedented energy-efficiency through a new chip architecture based on spiking neurons, low precision synapses, and a scalable communication network. Here, we demonstrate that neuromorphic computing, despite its novel architectural primitives, can implement deep convolution networks that (i) approach state-of-the-art classification accuracy across eight standard datasets encompassing vision and speech, (ii) perform inference while preserving the hardware's underlying energy-efficiency and high throughput, running on the aforementioned datasets at between 1,200 and 2,600 frames/s and using between 25 and 275 mW (effectively >6,000 frames/s per Watt), and (iii) can be specified and trained using backpropagation with the same ease-of-use as contemporary deep learning. This approach allows the algorithmic power of deep learning to be merged with the efficiency of neuromorphic processors, bringing the promise of embedded, intelligent, brain-inspired computing one step closer.

Artificial brains. A million spiking-neuron integrated circuit with a scalable communication network and interface.

Inspired by the brain's structure, we have developed an efficient, scalable, and flexible non-von Neumann architecture that leverages contemporary silicon technology. To demonstrate, we built a 5.4-billion-transistor chip with 4096 neurosynaptic cores interconnected via an intrachip network that integrates 1 million programmable spiking neurons and 256 million configurable synapses. Chips can be tiled in two dimensions via an interchip communication interface, seamlessly scaling the architecture to a cortexlike sheet of arbitrary size. The architecture is well suited to many applications that use complex neural networks in real time, for example, multiobject detection and classification. With 400-pixel-by-240-pixel video input at 30 frames per second, the chip consumes 63 milliwatts.

Colorimetric sensor array allows fast detection and simultaneous identification of sepsis-causing bacteria in spiked blood culture.

Sepsis is a medical emergency demanding early diagnosis and tailored antimicrobial therapy. Every hour of delay in initiating effective therapy measurably increases patient mortality. Blood culture is currently the reference standard for detecting bloodstream infection, a multistep process which may take one to several days. Here, we report a novel paradigm for earlier detection and the simultaneous identification of pathogens in spiked blood cultures by means of a metabolomic "fingerprint" of the volatile mixture outgassed by the organisms. The colorimetric sensor array provided significantly faster detection of positive blood cultures than a conventional blood culture system (12.1 h versus 14.9 h, P < 0.001) while allowing for the identification of 18 bacterial species with 91.9% overall accuracy within 2 h of growth detection. The colorimetric sensor array also allowed for discrimination between unrelated strains of methicillin-resistant Staphylococcus aureus, indicating that the metabolomic fingerprint has the potential to track nosocomial transmissions. Altogether, the colorimetric sensor array is a promising tool that offers a new paradigm for diagnosing bloodstream infections.

The use of colorimetric sensor arrays to discriminate between pathogenic bacteria.

A colorimetric sensor array is a high-dimensional chemical sensor that is cheap, compact, disposable, robust, and easy to operate, making it a good candidate technology to detect pathogenic bacteria, especially potential bioterrorism agents like Yersinia pestis and Bacillus anthracis which feature on the Center for Disease Control and Prevention's list of potential biothreats. Here, a colorimetric sensor array was used to continuously monitor the volatile metabolites released by bacteria in solid media culture in an Advisory Committee on Dangerous Pathogen Containment Level 3 laboratory. At inoculum concentrations as low as 8 colony-forming units per plate, 4 different bacterial species were identified with 100% accuracy using logistic regression to classify the kinetic profile of sensor responses to culture headspace gas. The sensor array was able to further discriminate between different strains of the same species, including 5 strains of Yersinia pestis and Bacillus anthracis. These preliminary results suggest that disposable colorimetric sensor arrays can be an effective, low-cost tool to identify pathogenic bacteria.