Systems and methods for virtual clinical trials

Abstract

The technology disclosed relates to a system and method for assigning participants to groups in a clinical trial. The system includes a federated server configured with group assignability data specifying a plurality of groups assignable to participants in a clinical trial and group distribution data specifying distribution of the participants into groups. The groups include at least one placebo group and one or more treatment groups. The system includes an intervention server configured to generate group encryption keys for encrypting the group assignability data. The system includes edge devices of each of the participants. The edge devices are in communication with the federated server.

1 . A system, comprising:
 a federated server configured with group assignability data specifying a plurality of groups assignable to participants in a clinical trial, and group distribution data specifying distribution of the participants into groups in the plurality of groups, the groups including at least one placebo group and one or more treatment groups;   an intervention server configured to generate group encryption keys for encrypting the group assignability data, the group encryption keys including a public key and a private key;   the federated server configured to generate respective group distribution thresholds for each of the groups based on the group distribution data, to encrypt the group assignability data using the public key and respective random initialization vectors for each of the participants to generate respective encrypted group assignability data for each of the participants, to split the group distribution thresholds into a federated server share of the group distribution thresholds and an edge device share of the group distribution thresholds, to split the encrypted group assignability data into a federated server share of the encrypted group assignability data and an edge device share of the encrypted group assignability data, and to generate respective federated server assignments of each of the participants to one of the groups using the federated server share of the group distribution thresholds and the federated server share of the encrypted group assignability data; and   respective edge devices of each of the participants, the respective edge devices in communication with the federated server, and configured to generate respective edge device assignments of each of the participants to one of the groups using the edge device share of the group distribution thresholds and the edge device share of the encrypted group assignability data, and to respectively combine the federated server assignments and the edge device assignments to produce respective encrypted group assignments of each of the participants to one of the groups.   
     
     
         2 . The system of  claim 1 , further comprising an initializer server in communication with the federated server and the respective edge devices, and configured to generate Beaver multiplication triples and respective random floating-point numbers for each of the participants. 
     
     
         3 . The system of  claim 2 , wherein the random floating-point numbers are between zero and one. 
     
     
         4 . The system of  claim 3 , wherein the initializer server is further configured to convert the random floating-point numbers into random fixed-precision numbers. 
     
     
         5 . The system of  claim 2 , wherein the initializer server is further configured with bit decomposition protocol to split the Beaver multiplication triples into a federated server share of the Beaver multiplication triples and an edge device share of the Beaver multiplication triples. 
     
     
         6 . The system of  claim 4 , wherein the initializer server is further configured with bit decomposition protocol to split the random fixed-precision numbers into a federated server share of the random fixed-precision numbers and an edge device share of the random fixed-precision numbers. 
     
     
         7 . The system of  claim 5 , wherein the initializer server is further configured to deliver the federated server share of the Beaver multiplication triples to the federated server, and to deliver the edge device share of the Beaver multiplication triples to the respective edge devices. 
     
     
         8 . The system of  claim 6 , wherein the initializer server is further configured to deliver the federated server share of the random fixed-precision numbers to the federated server, and to deliver the edge device share of the random fixed-precision numbers to the respective edge devices. 
     
     
         9 . The system of  claim 8 , wherein the federated server is further configured to generate the federated server assignments using the federated server share of the Beaver multiplication triples and the federated server share of the random fixed-precision numbers. 
     
     
         10 . The system of  claim 8 , wherein the respective edge devices are further configured to generate the edge device assignments using the edge device share of the Beaver multiplication triples and the edge device share of the random fixed-precision numbers. 
     
     
         11 . The system of  claim 1 , wherein each of the group distribution thresholds includes an upper distribution bound and a lower distribution bound. 
     
     
         12 . The system of  claim 11 , wherein the federated server is further configured to convert, for each of the groups, the upper distribution bound into an upper bound fixed-precision number, and the lower distribution bound into a lower bound fixed-precision number. 
     
     
         13 . The system of  claim 12 , wherein the federated server is further configured with bit decomposition protocol to split, for each of the groups, the upper bound fixed-precision number into a first part of the upper bound fixed-precision number and a second part of the upper bound fixed-precision number. 
     
     
         14 . The system of  claim 13 , wherein the federated server is further configured with bit decomposition protocol to split, for each of the groups, the lower bound fixed-precision number into a first part of the lower bound fixed-precision number and a second part of the lower bound fixed-precision number. 
     
     
         15 . The system of  claim 14 , wherein the federated server is further configured to generate, for each of the groups, the federated server share of the group distribution thresholds using the first part of the upper bound fixed-precision number and the first part of the lower bound fixed-precision number and the edge device share of the group distribution thresholds using the second part of the upper bound fixed-precision number and the second part of the lower bound fixed-precision number. 
     
     
         16 . The system of  claim 1 , wherein the intervention server is in communication with the respective edge devices, and further configured to receive the encrypted group assignments from the respective edge devices, and to decrypt the encrypted group assignments using the private key to determine respective group assignments of each of the participants. 
     
     
         17 . The system of  claim 16 , wherein the intervention server is further configured to determine, based on the group assignments, whether the distribution of the participants into the groups, as specified by the group distribution data, is met or not. 
     
     
         18 . The system of  claim 17 , wherein, when the distribution is not met, the intervention server is further configured to cause the federated server and the respective edge devices to respectively regenerate the federated server assignments and the edge device assignments to reproduce the encrypted group assignments. 
     
     
         19 . A method, including:
 receiving group assignability data specifying a plurality of groups assignable to participants in a clinical trial, and group distribution data specifying distribution of the participants into groups in the plurality of groups, the groups including at least one placebo group and one or more treatment groups;   generating group encryption keys for encrypting the group assignability data, the group encryption keys including a public key and a private key;   generating respective group distribution thresholds for each of the groups based on the group distribution data, to encrypt the group assignability data using the public key and respective random initialization vectors for each of the participants to generate respective encrypted group assignability data for each of the participants, to split the group distribution thresholds into a federated server share of the group distribution thresholds and an edge device share of the group distribution thresholds, to split the encrypted group assignability data into a federated server share of the encrypted group assignability data and an edge device share of the encrypted group assignability data, and to generate respective federated server assignments of each of the participants to one of the groups using the federated server share of the group distribution thresholds and the federated server share of the encrypted group assignability data; and   generating respective edge device assignments of each of the participants to one of the groups using the edge device share of the group distribution thresholds and the edge device share of the encrypted group assignability data, and to respectively combine the federated server assignments and the edge device assignments to produce respective encrypted group assignments of each of the participants to one of the groups.   
     
     
         20 . A non-transitory computer readable storage medium impressed with computer program instructions, the instructions, when executed on a processor, implement a method comprising:
 receiving group assignability data specifying a plurality of groups assignable to participants in a clinical trial, and group distribution data specifying distribution of the participants into groups in the plurality of groups, the groups including at least one placebo group and one or more treatment groups;   generating group encryption keys for encrypting the group assignability data, the group encryption keys including a public key and a private key;   generating respective group distribution thresholds for each of the groups based on the group distribution data, to encrypt the group assignability data using the public key and respective random initialization vectors for each of the participants to generate respective encrypted group assignability data for each of the participants, to split the group distribution thresholds into a federated server share of the group distribution thresholds and an edge device share of the group distribution thresholds, to split the encrypted group assignability data into a federated server share of the encrypted group assignability data and an edge device share of the encrypted group assignability data, and to generate respective federated server assignments of each of the participants to one of the groups using the federated server share of the group distribution thresholds and the federated server share of the encrypted group assignability data; and   generating respective edge device assignments of each of the participants to one of the groups using the edge device share of the group distribution thresholds and the edge device share of the encrypted group assignability data, and to respectively combine the federated server assignments and the edge device assignments to produce respective encrypted group assignments of each of the participants to one of the groups.