US 2026/0279598 A1 · FEDERATED SURVIVAL ANALYSIS WITH PRIVACY-PRESERVING AGGREGATION

Abstract

A method and systems for updating a federated survival model and generating personalized treatment recommendations. A plurality of local survival modeling nodes are each deployed at healthcare institutions. Each local survival modeling node trains a local survival model on institution-specific patient data. When the updates exceed a survival performance metric beyond a predetermined value, the local survival modeling nodes inject a calibrated noise into the local model update parameters to generate privacy-preserved model parameters to transmit to the federated learning server. The federated learning server generates feature availability vectors to aggregate the privacy-preserved model parameters into updated global model parameters. During the personalized treatment recommendation, the local survival modeling nodes integrate the updated global model parameters with the embedded feature representations of the patient's data. The local survival modeling node generates ranked treatment options for the patient based on a patient-specific survival prediction and local contextual factors.

Claims20

Dependent claims

A method for updating a federated survival model across a plurality of healthcare institutions, the method comprising:

receiving, at a federated learning server, privacy-preserved model parameters from a plurality of local survival modeling nodes, wherein each local survival modeling node is deployed at a respective healthcare institution and is configured to:
train a local survival model to generate local model update parameters using embedded feature representations of institution-specific patient data and local outcome data,
apply differential privacy noise to the local model update parameters to generate the privacy-preserved model parameters, and
transmit the privacy-preserved model parameters to the federated learning server;
generating, by the federated learning server, feature availability vectors based on feature presence at each healthcare institution;
clustering, by the federated learning server, the plurality of healthcare institutions into groups based on similarity of feature presence patterns;
computing, within each cluster, component-wise weighted parameter averages using weights derived from institutional sample size and disease-severity metrics to generate updated global model parameters;
and
distributing the updated global model parameters to the plurality of local survival modeling nodes.

6 more elements

A method for generating personalized treatment recommendations using a federated survival model, the method comprising:

receiving, at a local survival modeling node deployed at a healthcare institution, updated global model parameters from a federated learning server, wherein the updated global model parameters are generated by:
clustering a plurality of healthcare institutions into groups based on feature presence patterns, and
computing, within each cluster, component-wise weighted parameter averages using weights derived from institutional sample size and disease-severity metrics;
receiving, at the local survival modeling node, patient data for a patient;
transforming the patient data into embedded feature representations;
generating, by a personalized inference engine at the local survival modeling node, a patient-specific survival prediction by integrating the updated global model parameters with the embedded feature representations;
processing, by a contextual constraint processor, local contextual factors for the patient, the local contextual factors comprising at least one of cost constraints, insurance coverage, or treatment availability;
and
generating, by a treatment recommendation module, ranked treatment options for the patient based on the patient-specific survival prediction and the local contextual factors.

6 more elements

A decentralized federated learning system, comprising:

a plurality of local survival modeling nodes, each local survival modeling node deployed at a respective healthcare institution and comprising:
a local survival processor;
and
a local survival memory coupled to the local survival processor, the local survival memory storing processor-executable instructions that, when executed, configure the local survival processor for:
training a local survival model on institution-specific patient data;
computing local model update parameters based on local outcome data;
injecting a calibrated noise parameter into the local model update parameters to generate privacy-preserved model parameters;
and
transmitting the privacy-preserved model parameters to a federated learning server;
the federated learning server comprising:
a federated learning processor;
and
a federated learning memory to the federated learning processor, the federated learning memory storing processor-executable instructions that, when executed, configure the federated learning processor for:
receiving the privacy-preserved model parameters from the plurality of local survival modeling nodes;
generating feature availability vectors based on feature presence at each healthcare institution;
clustering the plurality of healthcare institutions into groups based on similarity of feature presence patterns;
executing a severity-weighted aggregation engine configured to compute, within each cluster, component-wise weighted parameter averages using weights derived from institutional sample size and disease-severity metrics to generate updated global model parameters;
and
distributing the updated global model parameters to the plurality of local survival modeling nodes.

16 more elements

Dependencies are inferred from the claim text.

Federated survival analysis with privacy-preserving aggregation

Life of this patent

Federated survival analysis with privacy-preserving aggregation

Life of this patent

Abstract

Claims20