Method and apparatus for detecting diverted drugs
A method and apparatus for detecting suspicious activities surrounding the management of controlled substances in a medical facility is described herein, where the activities that may indicate the diversion of controlled drugs are flagged for further review. The activities are detected by invigilating application layer network packets, related to controlled substances, on the medical facility network, and processing these packets with a rules engine and machine learning generated rules to make a determination if the circumstances surrounding the packets indicate the diversion of controlled substances.
1. A method of detecting diversion of controlled substances, the method comprising:
promiscuously invigilating application layer network packets on a network, filtering out all network traffic except application layer network packets;
reviewing said application layer network packets for material related to the controlled substances;
if the application layer network packets are related to the controlled substances;
executing a rules engine on contents of the application layer network packets to determine if the application layer network packets contains evidence of a possible controlled substance diversion;
if the application layer network packets contains the evidence of the possible controlled substance diversion;
sending the evidence of the possible controlled substance diversion for further review.
2. The method of claim 1 further comprising storing the contents of the application layer network packets in a database.
3. The method of claim 1 further comprising executing rules created by a machine learning algorithm on the application layer network packets to determine if the application layer network packets contains the evidence of the possible controlled substance diversion.
4. The method of claim 3 wherein the machine-learning algorithm utilizes a distributed algorithm operating on features located on different devices.
5. The method of claim 3 further comprising generating rules for the rules engine using a machine learning data set.
6. The method of claim 1 wherein the rules engine compares a patient's toxicology data with the controlled substances dispensed to the patient.
7. The method of claim 1 wherein the rules engine compares a patient's biological data with an expected impact of the controlled substances on the patient.
8. The method of claim 1 wherein the rules engine compares the controlled substances that are dispensed to a patient to a prescription for the controlled substances for the patient.
9. The method of claim 8 wherein the rules engine compares a picture of the controlled substance as the controlled substance is being dispensed to a picture of the controlled substance listed in the prescription.
10. The method of claim 1 wherein the rules engine compares a location of a patient with a location where the controlled substance is dispensed for said patient.
11. An apparatus for monitoring network packets on a network, the apparatus comprising:
a network interface device, said network interface device configured to promiscuously invigilate all traffic on the network, where the network interface device filters out all network traffic except application layer packets;
a database for storage of information from the application layer packets;
a processor unit connected to the database and the network interface device;
wherein the processor unit executes an algorithm to parse the application layer packets and filter out packets that relate to controlled substances, sending the application layer packets related to the controlled substances to a rules engine and a machine learning rules set;
wherein if the rules engine or the machine learning rules set determine that the application layer packet relate to the controlled substances, the application layer packet related to the controlled substances is sent for further review.
12. The apparatus of claim 11 wherein the network interface device is a network switch.
13. The apparatus of claim 11 wherein the network interface device is integrated into a server with the processor unit.
14. The apparatus of claim 11 further comprising a plurality of network interface devices that monitor a plurality of networks.
15. The apparatus of claim 11 wherein the rules engine compares a patient's toxicology data with the controlled substances dispensed to the patient.
16. The apparatus of claim 11 wherein the rules engine compares a patient's biological data with an expected impact of the controlled substances on the patient.
17. The apparatus of claim 11 wherein the rules engine compares the controlled substances that are dispensed to a patient to a prescription for the controlled substances for the patient.
18. The apparatus of claim 17 wherein the rules engine compares a picture of the controlled substance as it is being dispensed to a picture of the controlled substance listed in the prescription.
19. The apparatus of claim 11 wherein the rules engine compares a location of a patient with a location where the controlled substance is dispensed for said patient.
20. An apparatus for monitoring network packets on a network, the apparatus comprising:
a network interface means, said network interface means configured to promiscuously invigilate all traffic on the network, where the network interface means filters out all network traffic except application layer packets;
a database for storage of information from the application layer packets;
a processor unit connected to the database and the network interface means;
wherein the processor unit executes an algorithm to parse the application layer packets and filter out packets that relate to controlled substances, sending the application layer packets related to the controlled substances to a rules engine means and a machine learning rules means;
wherein if the rules engine means or the machine learning rules means determine that the application layer packets relate to the controlled substances, the application layer packets related to the controlled substances is sent for further review.