Advantages

Enabling Prediction of Discharge Eligibility:
- Previously, discharge eligibility was determined by measuring radiation dose at the time of discharge; this technology enables prediction at the time of diagnosis.

Improved Operational Efficiency in Radiation Therapy Units:
- Utilizing these predictive results helps optimize individual patient admission plans and bed allocation.

Background and Technology

In recent years, the field of theranostics—which aims to integrate diagnosis and treatment—has been rapidly gaining traction. Patients undergoing nuclear medicine therapy must remain in a radiation therapy ward until their radiation levels fall below the discharge threshold; however, these wards are costly to establish and maintain, and there is a chronic shortage of them. Under current operations, the decision to discharge a patient is based on survey meter readings taken at the scheduled discharge time, and no method for predicting this in advance has been established. Therefore, it would be beneficial to predict discharge eligibility for each patient in advance to optimize bed utilization. Furthermore, while previous studies have reported dose prediction models using survey meter readings, they faced challenges such as operator-dependent errors and instability in predicting unknown data.
This technology calculates treatment doses using Monte Carlo simulation based on patient-specific source distributions and body geometry obtained from diagnostic SPECT/CT images. This enables the prediction of discharge eligibility on a specified date at the time of diagnosis, thereby supporting the efficient operation of radiation therapy wards.

Data

- A dose prediction model for the day following treatment was developed using ¹¹¹In-Pentetreotide SPECT/CT images in 16 patients undergoing ¹⁷⁷Lu-DOTATATE therapy. Compared to previous studies, the model exhibited lower RMSE (root mean square error) and MAE (mean absolute error) values, indicating improved predictive accuracy.

- Stable dose prediction performance for unknown data was confirmed, suggesting the model’s potential for use in determining whether patients can be discharged the following day.

Expectations

TECH MANAGE is currently seeking companies interested in implementing or jointly developing this technology on behalf of The University of Osaka. We hope that this technology will be utilized as a value-added feature for medical devices or as a bed management solution. We can arrange meetings with the researchers and disclose unpublished data upon signing a non-disclosure agreement with the university. Please feel free to contact us.

Principal Investigator

Kentaro Ouchi (Radiation Control Office, Radiology Department, The University of Osaka Hospital, Japan)

Patents and Publications

Patent:
- Applied and pending (unpublished).