@shree_778gp@hraness It’s not difficult to have ideas. I’ve 28+ apps in the last 1.5months. Here’s your idea generator. Google pain points select any industry. Make app. Then really make the app not bs.
Thermal energy capture system from waste water heat. Prototype, systems run with MATLAB and reengineered over the last year. With some new tools was able to create a preliminary exploded model down to the nuts and bolts over 7000 individual elements. Still needs electrical, sensors, and a ton of other layers of intricacy. Currently have 35 individual ai roles across the production and engineering working on aspects of standards and governance, engineering, system design procurement processes, risk and other aspects to bring it to a higher level. As well as a consultant that is highly regarded in the industry that has been in talks.
A few updates. Some of the heart procedures and dye. Ultrasound is actually a really challenging feature to get right. Can see the real time flow rates that interact. Has all the electrical, but not sure I showed that. Every one of the 10.6m voxels that are part of the model are coded with density, fluid dynamics, electrical, and other aspects of anatomy.
It would be really cool if someone did one with 5 cameras. 6 would be cool, but would have to be on some sort of transport platform or falling or jumping, maybe like at a trampoline park setup the 6th in between two trampolines and they are jumping across the gap or someone from a diving board and a GoPro is on something stable as they jump off. Not sure how the center imaging would work with this sort of angular volumetric crossover, but maybe with opacity features on the overlay or as you turn the voxels the side with most angular surface area becomes the overarching shadowy 4th dimension. It assumes a 4th dimensional being sees us from one angle of time, which then means that for any real angle the being would still see someone parallax to another similar viewing capability of 4th dimension in their reality. If there were a 360 camera and the object was no long boxy it might look like a cylinder with angular adjustment which automatically changes the slicing view point across the surfaces.
@grok@sskmbk@grok What’s the significance of a successful consistent transmission? How difficult is it to get this to be reliable? What are the challenges?
YOLO object recognition for all the notes connected to the tablet camera. Create some rules for reading left to right too to bottom. Convert to timing and the buttons on a virtual keyboard. Motion scrolling in layers. If you put layers for the different audio tones it will add or play them all overtop of each other and you could even adjust the timing for each individually. Now take this same concept, reverse engineer a song where the sound is a visual equalizer have the audio then convert to layers as the different frequencies/instruments are converted to notes. So an audio clip could be listened to converted to visuals and then turned into specific notes.
It does have toggle-able parts for 11 systems. I’m working on the fluid dynamics to get those more correctly functioning. It has all the chemistries, cbc, medications used, that change the dynamics based on their interactions also. Has a lot more than a 1-2 min video could ever really touch on. I really need to sit down with a doctor to really iron out the simulation more to real experiences. The TEER has 4 clips and placement features, similar to a real one. Retracing my original builds to make it license-able. If you check out my profile you can see some of the model features and where it is currently. It’s all attached to ct and electrical profiles, but there’s a lot of work to get things more accurate. Ar/vr is already implemented, but not tested or validated and domacs and other loadables that adjust the shapes and sizes of the 626k vertices and 1.26m triangles and 10.5m voxels are also built in. It doesn’t allow PHI or identifiable features to be added intentionally or exports. Just for learning and simulation. I have three balloon procedures: balloon angioplasty & stenting, balloon valvuloplasty, and intra-aortic balloon pump(iabp) that I’m working on modeling also. There’s 12 categories of realism that have to be figured into this as well as all the findings for each. Then, extensive coding and complex engineering work to get it to actually run this. Each procedure has 100’s of pages of analysis for how it should work to simulate it properly. Then all the dependencies have to be mapped across the full modeling so all the features in menus coordinate the features. Takes a lot of heirarchy maps and reworking of interactive algorithms. Not many systems do this, even the best systems available are missing chemical degradation formulas accurately modeled. With it being a trimmed down version operating in a webgl/mobile/tablet environment it has to make some optimizations to be able to handle the gpu for all these calculations. So been a real medical engineering experience for sure. Plus, I’m in outside my field from Medical Laboratory science so hopefully I can get it right and accurate. It’s supposed to help people. That’s the real goal.