Lifeboat Foundation

Object detection has been an important task in the computer vision domain in recent decades. The goal is to detect instances of objects, such as humans, cars, etc., in digital images. Hundreds of methods have been developed to answer a single question: What objects are where?

Traditional methods tried to answer this question by extracting hand-crafted features like edges and corners within the image. Most of these approaches used a sliding-window approach, meaning that they kept checking small parts of the image in different scales to see if any of these parts contained the object they were looking for. This was really time-consuming, and even the slightest change in the object shape, lightning, etc., could have caused the algorithm to miss it.

Then there came the deep learning era. With the increasing capability of computer hardware and the introduction of large-scale datasets, it became possible to exploit the advancement in the deep learning domain to develop a reliable and robust object detection algorithm that could work in an end-to-end manner.

Year 2020 Stroke victims could eventually get dmt infusions where they can recover quickly after a stroke.

N, N-dimethyltryptamine (DMT) is an endogenous ligand of the Sigma 1 receptor (Sig-1R) with documented in vitro cytoprotective properties against hypoxia. Our aim was to demonstrate the in vivo neuroprotective effect of DMT following ischemia-reperfusion injury in the rat brain.

Transient middle cerebral occlusion (MCAO) was induced for 60 min in male Wistar rats using the filament occlusion model under general anaesthesia. Before the removal of the filament the treatment group (n = 10) received an intra-peritoneal (IP) bolus of 1 mg/kg-body weight (bw) DMT dissolved in 1 ml 7% ethanol/saline vehicle, followed by a maintenance dose of 2 mg/Kg-bw/h delivered over 24 h via osmotic minipumps. Controls (n = 10) received a vehicle bolus only. A third group (n = 10) received a Sig-1R antagonist (BD1063, 1 mg/kg-bw bolus +2 mg/kg-bw/h maintenance) in parallel with the DMT. Lesion volume was measured by MRI 24 h following the MCAO. Shortly after imaging the animals were terminated, and the native brains and sera were removed. Four rats were perfusion fixed. Functional recovery was studied in two separate group of pre-trained animals (n = 8–8) using the staircase method for 30 days.

Year 2021 face_with_colon_three

Shift to 7-nanometer process boosts the second-generation chip’s transistor count to a mind boggling 2.6-trillion

Ditching transistors for ferroelectric diodes, this new computer architecture could prove up to 100 times faster than current tech.

Colon cancer is the second deadliest cancer in the US. Early detection is important but finding and diagnosing polyps is difficult. 2 AI-powered endoscopes have been developed in Japan to tackle the problem. One can judge a lesion’s malignancy in 0.4 second with almost 100% accuracy. The other indicates lesions during an exam, even indistinct ones, like a car navigation system. We also introduce fermented Japanese foods that are beneficial for gut health and explain how they should be eaten.

Researchers advance efforts to create groups of robots that could construct almost anything.

When it comes to the manufacturing of commercial aircraft, different parts are manufactured at various locations. Before finally bringing them all together in a central plant and putting the finished aeroplane together, the tail components, the fuselage, and the wings are made at different plants.

Many other large structures, besides aircraft, are also built in sections.

Continue reading “MIT engineers design self-replicating robots capable of assembling giant structures” »

https://www.youtube.com/watch?v=I-dMVwN_qwk

Quick update/ reminder for Neuralink Event.

Twitter: https://twitter.com/ryantanaka3

Continue reading “Livestream: Neuralink Event Expectations” »

Artificial intelligence gets smarter, more capable, and more world-transforming every day. Here’s why that might not be a good thing.

Cornell University researchers have created an interface that allows users to handwrite and sketch within computer code – a challenge to conventional coding, which typically relies on typing.

The pen-based interface, called Notate, lets users of computational, digital notebooks open drawing canvases and handwrite diagrams within lines of traditional, digitized computer code.

Powered by a deep learning model, the interface bridges handwritten and textual programming contexts: notation in the handwritten diagram can reference textual code and vice versa. For instance, Notate recognizes handwritten programming symbols, like “n”, and then links them up to their typewritten equivalents.