# US 12,597,501 B2 · Intradialytic analysis method and analysis apparatus for dialysis > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,597,501 B2 - Document ID: US12597501B2 - Status: Granted - Kind: B2 - Application number: 17/530,484 - Filed: 2021-11-19 - Published: 2023-05-04 - Granted: 2026-04-07 - Earliest claimed: 2021-10-28 - Assignees: Wistron Corporation - Applicants: Wistron Corporation - Inventors: Chih-Hsin Yeh; Chih-Yi Chien; Chih-Feng Kuo; Ting-Hui Lee - Web page: https://patentgrep.com/patent/US12597501B2 ## CPC - G16H 20/40: relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture - A61M 1/1613: {Profiling or modelling of patient or predicted treatment evolution or outcome} - A61M 2205/52: with memories providing a history of measured variating parameters of apparatus or patient - G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems - G16H 50/80: for detecting, monitoring or modelling epidemics or pandemics, e.g. flu - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA - A61M: DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY (introducing media into or onto the bodies of animals A61D7/00; means for inserting tampons A61F13/26; devices for administering food or medicines orally A61J; containers for collecting, storing or administering blood or medical fluids A61J1/05); DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY (surgery A61B; chemical aspects of surgical articles A61L); DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR {(Electrotherapy, e.g. producing anaesthesia by the use of alternating or intermittent currents A61N1/36021)} ## Abstract An intradialytic analysis method and an analysis apparatus for dialysis are provided. One or more input features are obtained. The input features include variance relation between current data and previous data of an operating parameter related to a dialysis machine and data related to a tester. Future data is predicted according to the input features through one or more prediction models. The future data includes blood pressure information and a predicted result of intradialytic hypotension at a future time point. Therefore, intradialytic hypotension can be predicted with high accuracy. ## Claims ### Claim 1 An intradialytic analysis method, comprising: obtaining at least one input feature, wherein the at least one input feature comprises variance relation between current data and previous data of an operating parameter related to a dialysis machine and data related to a tester; predicting future data according to the at least one input feature through at least one prediction model, wherein the future data comprises blood pressure information and a predicted result of intradialytic hypotension at a future time point, the at least one prediction model comprises a plurality of first classification models and a plurality of second classification models, and the step of predicting the future data according to the at least one input feature through the at least one prediction model comprises: determining final future data according to the future data predicted by the first and second prediction models, wherein the final future data comprises the blood pressure information and the predicted result of the intradialytic hypotension at the future time point, the blood pressure information indicates systolic blood pressure, and the step of determining the final future data according to the future data predicted by the prediction models comprises: determining the blood pressure information in the future data predicted by the first classification models; and determining the predicted result of the intradialytic hypotension in the future data predicted by the second classification models according to the at least one input feature and the blood pressure information, resampling a plurality of training data used by the at least one prediction model for predicting the predicted result of the intradialytic hypotension according to the future data and corresponding actual data, comprising: adjusting a number of at least one positive sample and at least one negative sample in the training data according to a number ratio, wherein the number ratio is a ratio of an expected number of the at least one positive sample and the at least one negative sample, the at least one positive sample is a sample related to the intradialytic hypotension occurring in the actual data, and the at least one negative sample is a sample related to the intradialytic hypotension not occurring in the actual data; and training a new prediction model according to resampled training data. ### Claim 2 (depends on 1) The intradialytic analysis method according to claim 1, wherein the step of determining the final future data according to the future data predicted by the prediction models comprises: determining a statistical indicator of the future data predicted by the prediction models based on ensemble learning, wherein machine learning algorithms used by the prediction models are different. ### Claim 3 (depends on 1) The intradialytic analysis method according to claim 1, wherein the variance relation comprises at least one of an ultrafiltration rate change, a conductivity change, a dialysate temperature change, a blood flow change, a dehydration rate difference, a venous pressure difference, a dialysis time difference, a dialysate flow difference, a systolic blood pressure difference, and a pulse difference, the variance relation is changes in data between t and t−1 or between t−1 and t−2, t is a current time point, and t−1 and t−2 are previous time points. ### Claim 4 (depends on 1) The intradialytic analysis method according to claim 1, wherein the previous data comprises at least one of blood pressure information before a current dialysis operation and statistical information of a previous dialysis operation. ### Claim 5 (depends on 1) The intradialytic analysis method according to claim 1, further comprising: predicting the future data again through the new prediction model. ### Claim 6 (depends on 1) The intradialytic analysis method according to claim 1 wherein at least one of the at least one positive sample and the at least one negative sample is copied; and/or at least one of the at least one positive sample and the at least one negative sample is deleted. ### Claim 7 (depends on 1) The intradialytic analysis method according to claim 1, further comprising: emitting a warning notification according to the blood pressure information or the predicted result of the intradialytic hypotension at the future time point. ### Claim 8 (depends on 1) The intradialytic analysis method according to claim 1, further comprising: converting the operating parameter of the dialysis machine and/or the data related to the tester into the at least one input feature, wherein the at least one input feature conforms to an input format of the at least one prediction model. ### Claim 9 An analysis apparatus for dialysis, comprising: a storage, configured to store a code; and a processor, coupled to the storage and configured to load and execute the code to: obtain at least one input feature, wherein the at least one input feature comprises variance relation between current data and previous data of an operating parameter related to a dialysis machine and data related to a tester; and predict future data according to the at least one input feature through at least one prediction model, wherein the future data comprises blood pressure information and a predicted result of intradialytic hypotension at a future time point, the at least one prediction model comprises a plurality of first classification models and a plurality of second classification models, and the processor is further configured to: determine final future data according to the future data predicted by the first and second prediction models, wherein the final future data comprises the blood pressure information and the predicted result of the intradialytic hypotension at the future time point, and the blood pressure information indicates systolic blood pressure; determine the blood pressure information in the future data predicted by the first classification models; determine the predicted result of the intradialytic hypotension in the future data predicted by the second classification models according to the at least one input feature and the blood pressure information, resample a plurality of training data used by the at least one prediction model for predicting the predicted result of the intradialytic hypotension according to the future data and corresponding actual data, comprising: adjust a number of at least one positive sample and at least one negative sample in the training data according to a number ratio, wherein the number ratio is a ratio of an expected number of the at least one positive sample and the at least one negative sample, the at least one positive sample is a sample related to the intradialytic hypotension occurring in the actual data, and the at least one negative sample is a sample related to the intradialytic hypotension not occurring in the actual data; and train a new prediction model according to resampled training data. ### Claim 10 (depends on 9) The analysis apparatus for dialysis according to claim 9, wherein the processor is further configured to: determine a statistical indicator of the future data predicted by the prediction models based on ensemble learning, wherein machine learning algorithms used by the prediction models are different. ### Claim 11 (depends on 9) The analysis apparatus for dialysis according to claim 9, wherein the variance relation comprises at least one of an ultrafiltration rate change, a conductivity change, a dialysate temperature change, a blood flow change, a dehydration rate difference, a venous pressure difference, a dialysis time difference, a dialysate flow difference, a systolic blood pressure difference, and a pulse difference, the variance relation is changes in data between t and t−1 or between t−1 and t−2, t is a current time point, and t−1 and t−2 are previous time points. ### Claim 12 (depends on 9) The analysis apparatus for dialysis according to claim 9, wherein the previous data comprises at least one of blood pressure information before a current dialysis operation and statistical information of a previous dialysis operation. ### Claim 13 (depends on 9) The analysis apparatus for dialysis according to claim 9, wherein the processor is further configured to: predict the future data again through the new prediction model. ### Claim 14 (depends on 9) The analysis apparatus for dialysis according to claim 9, wherein at least one of the at least one positive sample and the at least one negative sample is copied; and/or at least one of the at least one positive sample and the at least one negative sample is deleted. ### Claim 15 (depends on 9) The analysis apparatus for dialysis according to claim 9, further comprising: a prompt apparatus, coupled to the processor, wherein the processor is further configured to: emit a warning notification according to the predicted result of the intradialytic hypotension through the prompt apparatus. ### Claim 16 (depends on 9) The analysis apparatus for dialysis according to claim 9, wherein the processor is further configured to: convert the operating parameter of the dialysis machine and/or the data related to the tester into the at least one input feature, wherein the at least one input feature conforms to an input format of the at least one prediction model.