AI is crucial for the automation of various tasks ranging from our day to day life to high tech industries such as medical, space, automobiles etc.. AI/DL models are powerful, but at the same time they are resource hungry (i.e., require considerable amount of memory storage and bandwidth), which makes their deployment difficult on mobile devices, or systems with limited resources. Energy consumption is another issue for large sized DL models on mobile devices. These requirements can significantly limit the applications of AI
At AIMED, we are developing AI technologies/tools with light weight AI models. Accuracy, and model training reproducibility of our models are comparable/higher than the state-of-the-art models. Enhanced disease diagnostic accuracy can be achieved for the resource limited devices (such as mobile phone, Nvidia’s Jetson, Raspberry pi) used in self-monitoring, and tele-screening. Potential applications of our models in various fields are limitless.