Notice of Publication of Article on Software as a Medical Device for Medical Support of Maintenance Hemodialysis

We work with physicians and researchers at universities and medical institutions to research and develop diverse modalities (pharmaceuticals, medical devices, and software as a medical device (SaMD) utilizing artificial intelligence (AI)) in the medical field to solve medical issues.

We are pleased to announce that the paper on our SaMD equipped with the AI to support maintenance hemodialysis treatment has been published in a review article in the first issue of the “Japanese Journal of Artificial Intelligence in Nephrology and Blood Purification”.

[SaMD for Medical Support of Maintenance Hemodialysis].

Patients with chronic kidney failure undergo maintenance hemodialysis three times a week for life to remove water and waste products to replace the crippled kidneys. The number of patients exceeds 330,000 and the medical cost exceeds 1 trillion yen. Insufficient dehydration causes heart failure, hypertension, and other cardiopulmonary dysfunctions, while excessive dehydration causes hypotension during dialysis, resulting in adverse events such as bad mood and loss of consciousness. A dialysis hospital typically manages several dozen patients with one physician and several nurses and clinical engineers. Therefore, human resources are inadequate at the hospitals and, the occurrence of adverse events during dialysis is an important medical challenge because it consumes scarce human resources and adversely affects the life expectancy of patients.

To realize safe and secure hemodialysis, we have developed the SaMD equipped with the AI that predicts appropriate total amount of water removal, and blood pressure decrease during dialysis, in collaboration with Tohoku University and other university hospitals, private dialysis clinics, NEC Corporation (NEC), and NEC Solution Innovator, Ltd. By learning the empirical knowledge of experienced nephrologists (utilizing the records of about 2,800 patients with 725,619 dialysis sessions), the SaMD can predict the target total amount of water removal on the day of dialysis with an error of just a glass of water (200 ml), and can also predict the occurrence of blood pressure drop (below 20 mmHg) during dialysis with an accuracy of AUC 0.91 before the dialysis starts.

Dialysis is highly specialized medical treatment, and it is difficult for non-specialist physician to set the appropriate amount of water removal as experienced nephrologists. However, the number of experienced nephrologists is not sufficient, and non-specialist physicians are often engaged in dialysis treatment in rural areas or at night, and in many dialysis facilities, non-specialist physicians, experienced nurses, and clinical engineers assist in setting the amount of water removal under the direction of a few experienced nephrologists. This SaMD can reduce the burden of medical personnel involved in dialysis treatment with fewer human resources, enabling safe and secure dialysis treatment.

The patent of the AI was applied in October 2022; and the project was adopted by AMED under the program of Research on Development of New Medical Devices in February 2023, with Tohoku University as the representative research organization. The pre-development consultation with PMDA was completed in May 2023, and the clinical performance study will be conducted for the regulatory approval.

For more information, please visit :