However the days of leaps in performance of image recognition are over, barring radical innovation in algorithm techniques. The gains in precision, recall and other metrics will be incremental so the emphasis has shifted to other points such as robust data acquisition and learning loops with partners.
While there is a spread in what different algorithms are offering, most are positioned as decision support tools. As a minimum, they need to detect the anomaly of interest. In some cases, they offer detailed pixel-level segmentation. The evolution is now to provide further information and explanations alongside object detection and instant segmentation. Some are even aiming to suggest treatment options, although this is generally further down the line. In short, the goal is to raise the AI complexity beyond object detection.
Furthermore, the algorithms today offer what humans do, but may do so faster and/or better, thus unleashing the automation wave. In the future, with more digitization of patient data, more data fusion can be expected, perhaps enabling AI to offer insights beyond human capability. This could be a game-changer says IDTechEx.
Scale will also be important. This gives more access to data, which increases algorithm accuracy, versatility, and applicability and creates a one-stop-shop proposition. Larger technical teams that can aid the on-site into-work-flow integration process, which in turn boosts installed base and acts a lock-in mechanism.
The report “AI in Medical Diagnostics 2020-2030: Image Recognition, Players, Clinical Applications, Forecasts” is at www.IDTechEx.com/IRAI
Related medical AI articles
- AI BOOST FOR LOW COST MEDICAL ROBOT MAGNETIC LOCATION SYSTEM
- APPLICATIONS PROCESSOR HAS DEDICATED NEURAL PROCESSING UNIT FOR EDGE AI