Magnetoencephalography (MEG) allows the non-invasive detection of interictal epileptiform discharges (IEDs). Clinical MEG analysis in epileptic patients traditionally relies on the visual identification of IEDs, which is time consuming and partially subjective. Automatic, data-driven detection methods exist but show limited performance. Still, the rise of deep learning (DL)-with its ability to reproduce human-like abilities-could revolutionize clinical MEG practice. Here, we developed and validated STIED, a simple yet powerful supervised DL algorithm combining two convolutional neural networks with temporal (1D time-course) and spatial (2D topography) features of MEG signals inspired from current clinical guidelines. Our DL model enabled both temporal and spatial localization of IEDs in patients suffering from focal epilepsy with frequent and high amplitude spikes (FE group), with high-performance metrics-accuracy, specificity, and sensitivity all exceeding 85%-when learning from spatiotemporal features of IEDs. This performance can be attributed to our handling of input data, which mimics established clinical MEG practice. Reverse engineering further revealed that STIED encodes fine spatiotemporal features of IEDs rather than their mere amplitude. The model trained on the FE group also showed promising results when applied to a separate group of presurgical patients with different types of refractory focal epilepsy, though further work is needed to distinguish IEDs from physiological transients. This study paves the way of incorporating STIED and DL algorithms into the routine clinical MEG evaluation of epilepsy.

Currently, the generative model has garnered considerable attention due to its application in addressing the challenge of scarcity of abnormal samples in the industrial Internet of Things (IoT). However, challenges persist regarding the edge deployment of generative models and the optimization of joint edge AI-generated content (AIGC) tasks. In this paper, we focus on the edge optimization of AIGC task execution and propose GMEL, a generative model-driven industrial AIGC collaborative edge learning framework. This framework aims to facilitate efficient few-shot learning by leveraging realistic sample synthesis and edge-based optimization capabilities. First, a multi-task AIGC computational offloading model is presented to ensure the efficient execution of heterogeneous AIGC tasks on edge servers. Then, we propose an attention-enhanced multi-agent reinforcement learning (AMARL) algorithm aimed at refining offloading policies within the IoT system, thereby supporting generative model-driven edge learning. Finally, our experimental results demonstrate the effectiveness of the proposed algorithm in optimizing the total system latency of the edge-based AIGC task completion.

In electroencephalogram (EEG) recordings, the presence of interictal epileptiform discharges (IEDs) serves as a critical biomarker for seizures or seizure-like events.Detecting IEDs can be difficult; even highly trained experts disagree on the same sample. As a result, specialists have turned to machine-learning models for assistance. However, many existing models are black boxes and do not provide any human-interpretable reasoning for their decisions. In high-stakes medical applications, it is critical to have interpretable models so that experts can validate the reasoning of the model before making important diagnoses. We introduce ProtoEEGNet, a model that achieves state-of-the-art accuracy for IED detection while additionally providing an interpretable justification for its classifications. Specifically, it can reason that one EEG looks similar to another ''prototypical'' EEG that is known to contain an IED. ProtoEEGNet can therefore help medical professionals effectively detect IEDs while maintaining a transparent decision-making process.

As reported by the Informationsdienst Wissenschaft, Continental intends to use its collaboration with the DFKI both to optimize its internal processes and to advance the mobility of the future, especially concepts for autonomous vehicles. To this end, the company and the research institute want to set up a research laboratory for intelligent technologies (FIT) at DFKI's Kaiserslautern site. Continental employees will work together with the scientists of the DFKI on basic research, but also tackle specific problems. For Continental, DFKI is the most recent of a number of cooperation partners, including technology companies Nvidia and Baidu, Oxford University, Darmstadt Technical University and the Indian Institute of Technology Madras (India). Continental has already begun to develop a platform for highly automated driving with the AI processor manufacturer Nvidia.

India has ambitions to fire up its artificial intelligence capabilities -- but experts say that it's unlikely to catch up with the U.S. and China, which are fiercely competing to be the world leader in the field. An Indian government-appointed task force has released a comprehensive plan with recommendations to boost the AI sector in the country for at least the next five years -- from developing AI technologies and infrastructure, to data usage and research. The task force, appointed by India's Ministry of Commerce and Industry, proposes that the government work with the private sector to develop technologies, with a focus on smart cities and the country's power and water infrastructure. It recommends a network of infrastructure -- a testing facility, and six centers focusing on research in generating AI technologies, such as robotics, autonomous trucks and advanced financial technology. A data center could be set up to "develop an autonomous AI machine that can work on multiple data streams in real time," the plan said. Calling data the "fuel that powers AI," the report said data marketplaces and exchanges could allow the "free flow of data."

More than a decade after the improvised explosive device became the scourge of the wars in Iraq and Afghanistan, the Pentagon is battling another relatively rudimentary device that threatens to wreak havoc on American troops: the drone. Largely a preoccupation of hobbyists and experimenting companies, the vehicles are beginning to become a menace on the battlefield, where their benign commercial capabilities have been transformed into lethal weapons and intelligence tools. Instead of delivering packages, some have been configured to drop explosives. Instead of inspecting telecommunications towers, others train their cameras to monitor troops and pick targets. Instead of spraying crops, they could spread toxic gas, commanders worry.

The operation to liberate the ISIS Syrian stronghold of Raqqa has entered its third month, and while the U.S. and its partners have largely depleted the enemy ranks - but lethal danger lurks throughout the city. There are about 1,500 ISIS fighters left in Raqqa, a big reduction from around 5,000 less than two months ago, according to Col. Ryan Dillon, spokesman for Operation Inherent Resolve – the U.S.-led coalition tasked to destroy ISIS in Iraq and Syria. But Raqqa is still teeming with landmines and booby traps, many set by fleeing jihadists. "Eighty percent of the engagement the Syrian Democratic Forces (SDF) has had has been with IEDs, whether they be vehicle-born IEDs, inside houses, static vehicles and even IEDs planted inside corpses," Dillion told Fox News. "Those have been the proponents of how ISIS is fighting in Raqqa so far."

Imagine you're an evil genius in the style of a James Bond villain. You've got a hundred million dollars or so burning a hole in your pocket, and you're looking to cause some destruction. You want to know your options. Greg Allen (@Gregory_C_Allen) is a George Leadership Fellow at Harvard Kennedy School and Harvard Business School. He previously worked on space and robotics issues at the White House Office of Science and Technology Policy.

The US Air Force is spending $15m on a mysterious drone killing system from an Israeli firm. The contract for'counter-unmanned aerial systems' will supply 21 kits, which are believed to be earmarked for dealing with the growing threat of drones from ISIS. However, details of the kits and how they will work have not been revealed, although it is believed to be a modified version of the firm's existing'drone shield' The deal is with ELTA North America, a U.S. subsidiary of Israeli Aerospace Industries which does produce a'drone buster' called Drone Shield, pictured here. It is believed the new system mixes scanning systems with a system to disable drones mid flight, or cause them to return to their base, allowing them to be tracked. According to Army documents, 'ELTA North America Inc., Annapolis Junction, Maryland, has been awarded a $15,553,483 firm-fixed-price letter contract for counter-unmanned aerial systems.

While bomb disposal robots have become a a common sight on land, the US Navy is now hoping to use them underwater as well. It hopes the robo-hands will be able to search harbours, piers and even ships for IEDs. Once found, the hands are so dexterous they will be able to safety diffuse the device and remove it, the researchers hope. The'underwater dual manipulator system' has two robot arms, and will be put onto an underwater drone to look for, and defuse IEDs. The robot arms will be attached to a US Navy underwater drone.