Hardware Podcasts

149. Reverse Engineering The Brain With AI

The Next Byte Podcast · 21 Nov 2023 · 24m

AI in practice: 2 real uses of AI described

What this episode covers

Hosts Daniel and Forbode discuss MIT researchers' development of 'Smart EM,' an AI-integrated scanning electron microscope that accelerates connectomics by autonomously identifying regions of interest in brain tissue. They explain how the system reduces the time to map 100,000 neurons from over a decade to under three months, drawing parallels to human visual attention and Elon Musk’s first-principles approach to material science. The hosts also read a listener shout-out from AI architect Bruno Marquier and announce an upcoming interview episode.

AI in practice

Where this conversation describes AI being used to build or deliver a product. Part of the AI in practice tracker.

Engineering & simulationReduced mapping time for 100,000 neurons from over a decade to under 3 months (40x speedup)

The AI model controls the electron microscope to identify regions of interest, focusing on complex areas and skipping routine ones, which optimizes the image acquisition process.

“it basically the machine learning helps control the microscope and helps the microscope to understand and focus on what are the important parts of the image. What are the parts that are worth taking a photo of and doing further study on, and which parts are routine or not interesting, and which parts can we basically discard.”

Speaker 2 · Hear it at 7:34

Engineering & simulationSmart EM (machine learning algorithm)

The algorithm slows down the scanning speed when it detects regions of interest to capture higher-quality images, ensuring data quality is maintained while accelerating the overall acquisition process.

“once it finds regions of interest, it slows down the scanning so that it can have the best possible image before moving on. So that you're not, I guess, not only are you expediting this but you're not losing quality in your analysis either.”

Speaker 1 · Hear it at 11:02

Most relevant to

large enterprises

Topics

Mentioned

Companies
MIT, Mouser Electronics, Tesla, Wevolver
Products
Smart EM
Also
Elon Musk, Bruno Marquier

Related episodes from other shows

Other shows on the same ground, closest first.

  • EE Times Current · 43m

    Intel’s Hamster | Conferences To Go | Apollo 13 in Real Time

    Intel researcher Junko Yoshida explains the company's neuromorphic Loihi chip system, scaled to 768 chips and 100 million neurons (comparable to a naked mole rat), which uses spiking neural networks for low-power edge computing. Amir Sherman of Arrow Electronics discusses 'Embedded to Go,' a virtual conference featuring partners like STM and ON Semiconductor, offering free development boards to attendees. Ben Feist, creator of the 'Apollo 13 in Real Time' website, details how he synchronized historical NASA audio and video to recreate the mission's timeline, highlighting the engineering rigor of the flight controllers.

    AI & machine learningMicrocontrollers & dev platforms
    5 chapters

  • How I Built This · 35m

    When your headphones listen to you with Ramses Alcaide of Neurable

    Ramses Alcaide, co-founder and CEO of Neurable, explains how his company developed Anten, a pair of headphones that use non-invasive brain-computer interface (BCI) technology to track focus and prevent burnout. Alcaide details the transition from his PhD research on EEG for ALS patients to creating a consumer product that integrates AI to boost signal-to-noise ratios from ear-level sensors. He outlines Neurable's roadmap, which includes upcoming in-earbuds and partnerships with OEMs to embed BCI into various headworn devices for both consumer productivity and enterprise safety applications.

    Embedded firmwareStarting a product company
    9 chapters

  • Hackaday Podcast · 1h 02m

    Ep 291: Walking in Space, Lead in the Earth, and Atoms under the DIY MIcroscope

    Hosts Elliot Williams and Al Williams review the week's top hacks, including a DIY scanning tunneling microscope built by Mech Panda that images atoms, and a deep dive into power optimization for the Raspberry Pi Zero by Manaworm. They also discuss a 32-IMU sensor array by Will Wong designed for high-precision dead reckoning, and a hardware hack by Treble that enables NVMe SSDs in a ThinkPad 13. The episode concludes with long-form articles on the history and dangers of spacewalks (EVA) and the industrial mining of lead.

    Prototyping & iterationBatteries & low-power design
    7 chapters

  • Hackaday Podcast · 1h 08m

    Ep 188: Zapping Cockroaches, Tricking AIs, Antique 3D Scanning, and Grinding Chips to QFN

    Hackaday Editor-in-Chief Elliot Williams and Staff Writer Dan Maloney review a week of hardware hacks, including a laser-based cockroach neutralization system, a microscope built from DVD pickup sleds, and the manual conversion of DIP chips to QFN packages for a handheld NES. They also discuss prompt injection vulnerabilities in GPT-3, the history of the Q1 microcomputer, and sources for harvesting lithium-ion batteries.

    Mechanical engineeringAI & machine learning
    9 chapters

  • Hackaday Podcast · 54m

    Ep028: Brain Skepticism Turned Up to 11, Web Browsing in '69, Verilog For 7400 Logic

    Hackaday editors Mike Szczys and Elliot Williams discuss the technical realities and hype surrounding Neuralink's brain-computer interface, analyzing the gap between their surgical robot demo and actual neuroscience. They also cover the 40th anniversary of the Sony Walkman, arguing that its success was driven by the design of its headphones rather than the player itself. The episode highlights several hardware hacks, including synthesizing 7400-series logic from Verilog, a subtractive 3D printing process using MDF, and a rotary phone hack that controls web browsers.

    Debugging & failure analysisPrototyping & iteration
    8 chapters

The summary, chapters and topics are generated from the episode’s audio by this directory, not written by The Next Byte Podcast. The episode itself belongs to its publisher.