CAIDE Systems: Opening up Diverse Possibilities in Diagnosis with AI

Though doctors are not very open to using this ‘AI detection system’ currently, as they are unsure of the results it could bring forth, CAIDE Systems is going on to build an entire body detection system. With the buzz that AI has been creating in the recent past, the firm believes that it will be utilized in the near future. In fact, CAIDE Systems is connecting with large hospitals in South Korea to test their systems. However, the firm’s brain disease detection system—a part of m:Studio that already contains numerous trained AI model sets— to review CT images and accurately sort hemorrhages and its subtypes, is out in the market and has already proven to be a success. The AI model is trained through 6,000 patients’ data and 140,000 stroke CT images, giving it the capability to detect and state stroke types and locations accurately. It eliminates all human errors and provides the inference result within a minute.

“We have successfully built our functional tools for both disease detections and effective training, and that is our biggest achievement in technology. Now we are actively communicating with customers to utilize this device,” says Lee. The firm is approaching medical research teams and contacting equipment manufacturing companies to find solutions for their hardware devices— turning them into automatic annotation systems. “As of now we are developing our software and are actively looking to provide excellent quality medical services to all in the near future,” adds Lee.

CAIDE Systems is also planning on selling their AI engine to hardware companies and PACS systems such as GE and Philips. Besides, the firm also intends on enhancing its disease detection system and training its AI models to detect over 500 diseases by 2020. “Our goal is to provide a user-friendly platform for consistent, accurate, and fast diagnosis,” concludes Lee.