Thomas Oxley? Endovascular? Looks like they finally commercialised the stentrode! I remember seeing the initial trials of this back in the day. Basically fu

Thomas Oxley? Endovascular? Looks like they finally commercialised the stentrode! I remember seeing the initial trials of this back in the day. Basically functions like a slightly worse ECoG sensor (since the signal still gets filtered by the dura), but much easier to install. Think best case performance of about 80 binary inputs per minute, or a really noisy cursor input. Honestly, without the improvement in algorithms they were hoping for, there's not a snowball's chance in hell of this playing Doom within 5 years, let alone one. There's still fundamental unsolved problems (like the fact that ~20% of people simply can't Motor Imagery BCIs no matter how hard they try, and a massive chunk of those who can produce incredibly noisy signals), but new sensors are always a step in the right direction (as opposed to new algorithms, which, in this field, are almost always complete bullshit.)