Systems and methods for predicting dosage
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- UNIVERSITY OF MELBOURNE
- Filing Date
- 2024-04-26
- Publication Date
- 2026-06-16
Smart Images
Figure 2026519402000001_ABST
Abstract
Claims
1. A method for predicting the toxicity of a drug administered to a patient, wherein the method is Accessing at least one computed tomography (CT) torso slice associated with the patient, The process involves using a trained artificial intelligence (AI) segmentation model to perform a labeling process for at least one CT scan, To determine at least one body composition parameter based on at least one labeled CT slice, To receive at least one parameter associated with the aforementioned patient, The patient receives at least one dosage parameter associated with the drug administered to the patient, A method comprising using a trained AI predictive model to generate an output based on the at least one body composition parameter, the at least one demographic parameter, and the at least one dosage parameter, wherein the output corresponds to the predicted toxicity of the drug to the patient.
2. The method according to claim 1, further comprising training the AI prediction model.
3. The method according to claim 1 or 2, wherein the labeled CT slice includes one or all of the labeled regions of muscle, visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), intermuscular / intramuscular adipose tissue (IMAT), bone, and organs.
4. The method according to any one of the prior claims, wherein the body composition parameter is at least one or all of the following: muscle mass, VAT mass, SAT mass, IMAT mass, bone mass, organ mass, muscle radiation density, VAT radiation density, SAT radiation density, IMAT radiation density, bone radiation density, or organ radiation density.
5. The method according to any one of claims 1 to 4, wherein the drug is a chemotherapy drug used to treat one of breast cancer, lung cancer, liver cancer, colorectal cancer, colon cancer, prostate cancer, and pancreatic cancer.
6. The method according to any one of claims 1 to 5, wherein the parameter associated with the patient includes a parameter associated with a treatment being administered to the patient.
7. The method according to any one of claims 1 to 6, wherein the parameter associated with the patient includes a parameter obtained from a blood test.
8. A method for training an artificial intelligence model to evaluate drug dosages that minimize toxic reactions to a drug, wherein the method is Receiving a labeled CT torso slice sample associated with the patient's sample, Based on each labeled CT slice, determine at least one body composition parameter, Identifying a set of predictive parameters for each patient, wherein the predictive parameters include at least one body composition parameter, at least one patient demographic parameter, and at least one dosage parameter. For each combination of the aforementioned prediction parameters, a model fitting process is performed to cause the model to generate a predicted toxicity rank for each patient. For each combination of the prediction parameters, the accuracy of the combination of prediction parameters is determined by comparing the predicted toxicity rank with the measured toxicity rank. A method comprising selecting an optimal combination of prediction parameters, wherein the optimal combination of prediction parameters is the combination of prediction parameters that yields the highest accuracy.
9. The method according to claim 8, wherein the labeled CT slice includes one or all of the labeled regions of muscle, VAT, SAT, IMAT, bone, and organs.
10. The method according to claim 8 or 9, wherein the body composition parameter is at least one or all of the following: muscle mass, VAT mass, SAT mass, IMAT mass, bone mass, organ mass, muscle radiation density, VAT radiation density, SAT radiation density, IMAT radiation density, bone radiation density, or organ radiation density.
11. The method according to any one of claims 8 to 10, further comprising using the trained AI model to label a sample of CT slices to generate a sample of labeled CT slices.
12. The method according to any one of the prior claims, wherein the CT slice is derived from the abdomen.
13. The method according to any one of the prior claims, wherein 2 to 1,000 CT slices are received.
14. A method for minimizing a patient's toxic reaction to a drug, wherein the method is i) Determining the amount of muscle, visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), and intermuscular / intramuscular adipose tissue (IMAT) in at least a portion of the patient's torso, ii) Determining the risk of a toxic reaction to the drug for the patient based on at least i), iii) A method comprising reducing the standard recommended dose of the drug if the patient is determined to be at risk of a toxic reaction.
15. The method according to claim 14, further comprising administering the drug.
16. The method of claim 14 or 15, wherein step i) is carried out using the method of any one of claims 1 to 7, 12 or 13.
17. The method according to any one of claims 14 to 16, wherein the drug is a chemotherapy drug used to treat one of breast cancer, lung cancer, liver cancer, colorectal cancer, colon cancer, prostate cancer, and pancreatic cancer.
18. A non-temporary computer-readable storage medium for storing instructions, wherein when the instructions are executed by a processing device, the processing device causes the processing device to perform the method according to any one of claims 1 to 17.
19. A computer system configured to generate text, wherein the computer system Processor and A computer system comprising the storage medium described in claim 18.