The University of Miyazaki Hospital faced an increased number of image diagnostic cases with a lack of radiologists to handle the massive amount of data needed per patient. The hospital teamed up with NTT DATA to implement AI into its workflow to reduce the workers’ burden.Read The Case Study
NTT DATA’s AI has been designed specifically to optimize a doctors’ diagnostic efficiency, and gave us hope that they could work toward practical application.
— Dr. Minako Azuma, University of Miyazaki Hospital
As medical scanning devices become more technologically advanced, their importance increases because their use in patient diagnosis increases. However, the number of radiologists at the University of Miyazaki Hospital operating this technology has remained flat or even decreasing.
This results in each radiologist’s workload increasing, putting stress on each worker, and making each patient’s diagnosis take even longer. The right solution would be one that could optimize work while also maintaining the necessary precision needed to detect abnormalities from the scanned images.
NTT DATA integrated their AI into the hospital’s current workflow, which uses a picture archiving and communication system (PACS) to detect diseases in the body. The AI solution was aided by NTT DATA in the U.S., who provided data medical image archives.
These additional images helped refine the AI’s accuracy for detecting abnormalities by machine learning over 19 billion medical CT images covering a wide range of diseases. Radiologists were able to continue their regular workflow while utilizing and applying the AI image diagnosis.
By introducing AI into the existing work process, utilization was streamlined and simplified. The AI technology holds promise for the future, with initial testing showing that it could detect various diseases at a high level of accuracy, no matter who the patient was.
The hope for the future is that such AI will analyze types of lesions in greater detail, classify the disease, and even predict a tumors’ nature. This technology would greatly relieve the burden placed on radiologists as diagnoses would be faster and accurate.