Month: March 2017

Automated diagnosis

“I think that if you work as a radiologist you are like Wile E. Coyote in the cartoon. You’re already over the edge of the cliff, but you haven’t yet looked down. There’s no ground underneath.” Deep-learning systems for breast and heart imaging have already been developed commercially. “It’s just completely obvious that in 5 years deep learning is going to do better than radiologists. It might be 10 years. I said this at a hospital. It did not go down too well.”

2022-10-05: The deep learning dividend for medicine

Today’s report on AI of retinal vessel images to help predict the risk of heart attack and stroke, from 65k UK Biobank participants, reinforces a growing body of evidence that deep neural networks can be trained to “interpret” medical images far beyond what was anticipated. Add that finding to last week’s multinational study of deep learning of retinal photos to detect Alzheimer’s disease with good accuracy. AI models have been shown to be quite useful for detecting eye diseases, such as diabetic retinopathy. But this is about the indirects, the not so obvious. That work has now extended to detection of kidney disease, control of blood glucose and blood pressure, hepatobiliary disease, a previous study on predicting heart attack, close correlation of the retinal vessels with the heart (coronary) artery calcium score


2023-07-31: Misdiagnosis is one of the biggest causes of death, yet doctors think they’re better than AI

~800k Americans are permanently disabled or die each year from diagnostic medical errors. “Our results demonstrate that, unless the documented mistakes can be corrected, the optimal solution involves assigning cases either to humans or to AI, but rarely to a human assisted by AI.”

State of motion capture

motion capture has come a long way

We present the first end-to-end solution to create high-quality free-viewpoint video encoded as a compact data stream. Our system records performances using a dense set of RGB and IR video cameras, generates dynamic textured surfaces, and compresses these to a streamable 3D video format. 4 technical advances contribute to high fidelity and robustness: multimodal multi-view stereo fusing RGB, IR, and silhouette information; adaptive meshing guided by automatic detection of perceptually salient areas; mesh tracking to create temporally coherent subsequences; and encoding of tracked textured meshes as an MPEG video stream. Quantitative experiments demonstrate geometric accuracy, texture fidelity, and encoding efficiency. We release several datasets with calibrated inputs and processed results to foster future research.

Mythic

Mythic can do an 8-bit multiply and add in a single transistor

2020-10-17: AI Analog Compute

Mythic is the first and only company that have been able to implement a deep learning model like ResNet 50 in a non-digital architecture: > 50 layers, 1000 fps, 3W total, 9->2ms latency, 8 TOPS/W in 40nm silicon. 10x cost advantage over digital chips.

2023-04-11: Commercialization takes a long time

Mythic’s analog chip uses less power by storing neural weights not in SRAM but in flash memory, which doesn’t consume power to retain its state. And the flash memory is embedded in a processing chip, a configuration Mythic calls “compute-in-memory.” Instead of consuming a lot of power moving millions of bytes back and forth between memory and a CPU (as a digital computer does), some processing is done locally. Mythic’s success on that front has been variable: The company ran out of cash and raised $13 million in new funding and appointed a new CEO.
I asked him whether the state of analog computing today could be compared to that of quantum computing 25 years ago. Could it follow a similar path of development, from fringe consideration to common (and well-funded) acceptance?

It would take a fraction of the time. “We have our experimental results. It has proven itself. If there is a group that wants to make it user-friendly, within 1 year we could have it.” And at this point he is willing to provide analog computer boards to interested researchers, who can use them with Achour’s compiler.