Medical device location and tracking system
Methods and systems for automatically ascertaining physical location information about a plurality of medical device controllers and tracking changes in the physical locations of ones of the respective medical device controllers. One or more machine learning modules infer the physical locations from computer network messages received from the medical device controllers, including information about computer network components that are proximate ones of the medical device controllers.
1 . A medical device controller location and tracking system, the system comprising:
a first electronic data store comprising a plurality of first entries, wherein each first entry contains information that associates a medical device controller with a physical location;
a second electronic data store comprising a plurality of second entries, wherein each second entry contains information that associates a computer network device with a physical location;
a plurality of medical device controllers, wherein each medical device controller comprises a user interface configured to receive information, from a human user, identifying an entered physical location of the respective medical device controller, and wherein each user interface comprises a display screen configured to display operational data for a heart pump communicatively coupled to the respective medical device controller;
a plurality of computer network devices;
a monitoring station configured to display physical locations associated with one or more of the plurality of medical device controllers; and
a server comprising a plurality of machine learning modules, the server configured to:
receive messages, via a computer network, from the plurality of medical device controllers, wherein each message includes: (a) an identifier of the respective medical device controller, (b) an identifier of one of the plurality of computer network devices through which the respective message traveled en route to the server, and (c) an image of at least some of the contents displayed on the display screen of the respective medical device controller;
for each of the received messages, learn a physical location of the respective medical device controller with the plurality of machine learning modules by:
extracting at least some of the operational data for the heart pump communicatively coupled to the respective medical device controller from the respective image;
determining whether an entry in the first electronic data store associates the respective medical device controller with a physical location;
determining whether an entry in the second electronic data store associates the respective one of the plurality of computer network devices with a physical location;
if (a) an entry in the first electronic data store is found that associates the respective medical device controller with a physical location and (b) no entry in the second electronic data store is found that associates the respective one of the plurality of computer network devices with a physical location, then storing, with the plurality of machine learning modules, a new second entry in the second electronic data store, wherein the new second entry associates the respective one of the plurality of computer network devices with the physical location associated with the respective medical device controller in the first electronic data store;
if (a) no entry in the first electronic data store is found that associates the respective medical device controller with a physical location and (b) an entry in the second electronic data store is found that associates the respective one of the plurality of computer network devices with a physical location, then storing, with the plurality of machine learning modules, a new first entry in the first electronic data store, wherein the new first entry associates the respective medical device controller with the physical location associated with the respective one of the plurality of computer network devices in the second electronic data store; and
if (a) an entry in the first electronic data store is found that associates the respective medical device controller with a physical location, (b) an entry in the second electronic data store is found that associates the respective one of the plurality of computer network devices with a physical location, and (c) a comparison indicates that the physical location associated with the respective medical device controller in the first electronic data store differs from the physical location associated with the respective one of the plurality of computer network devices in the second electronic data store, then revising, with the plurality of machine learning modules, the entry in the first electronic data store to associate the respective medical device controller with the physical location associated with the respective one of the plurality of computer network devices in the second electronic data store;
transmit, via the computer network, to the monitoring station, a physical location of one of the plurality of medical device controllers that was learned by the plurality of machine learning modules; and
transmit, via the computer network, to the monitoring station, an alarm based on the operational data extracted from one of the received messages, wherein the one received message was received from the one medical device controller, and wherein the alarm prompts an adjustment to a speed setting or a motor current value of the heart pump communicatively coupled to the one medical device controller.
2 . The system of claim 1 , wherein, for each received message, the identifier of the respective medical device controller comprises a serial number, and wherein, for each of the received messages, the server is further configured to extract the serial number from the respective image.
3 . The system of claim 2 , wherein, for each received message, the identifier of the respective one of the plurality of computer network devices comprises an Internet protocol (IP) address, a media access control (MAC) addresses, or a service set identifier (SSID).
4 . The system of claim 1 , wherein at least one of the plurality of first entries in the first electronic data store was created in response to a human user providing a physical location of one of the plurality of medical device controllers through the user interface of that medical device controller.
5 . The system of claim 1 , wherein, for each received message, the respective image comprises a serial number, a placement signal, a speed setting, a motor current value, a blood flow rate, and one or more graphs.
6 . The system of claim 5 , wherein the respective image of the one received message further comprises an error message.
7 . The system of claim 1 , wherein the operational data extracted from the one received message comprises a heart signal level, a battery temperature, a blood flow rate, or an indication of plumbing integrity.
8 . The system of claim 1 , wherein the alarm prompts an adjustment to the speed setting of the heart pump communicatively coupled to the one medical device controller.
9 . The system of claim 1 , wherein the alarm prompts an adjustment to the motor current value of the heart pump communicatively coupled to the one medical device controller.
10 . The system of claim 1 , wherein the alarm further prompts an adjustment to a position of the heart pump communicatively coupled to the one medical device controller.
11 . The system of claim 1 , wherein the alarm further prompts a restart of the heart pump communicatively coupled to the one medical device controller.
12 . The system of claim 1 , wherein the alarm further prompts a dispatch of medical personnel to the physical location of the one medical device controller that was learned by the plurality of machine learning modules.
13 . A method comprising:
receiving messages, at one or more servers, via a computer network, from a plurality of medical device controllers, wherein each medical device controller comprises a user interface configured to receive information, from a human user, identifying an entered physical location of the respective medical device controller, wherein each user interface comprises a display screen configured to display operational data for a heart pump communicatively coupled to the respective medical device controller, and wherein each message includes: (a) an identifier of the respective medical device controller, (b) an identifier of one of a plurality of computer network devices through which the respective message traveled en route to the one or more servers, and (c) an image of at least some of the contents displayed on the display screen of the respective medical device controller;
for each of the received messages, extracting at least some of the operational data for the heart pump communicatively coupled to the respective medical device controller from the respective image;
for each of the received messages, learning a physical location of the respective medical device with a plurality of machine learning modules by:
determining whether an entry in a first electronic data store associates the respective medical device controller with a physical location, wherein the first electronic data store comprises a plurality of first entries, and wherein each first entry contains information that associates a medical device controller with a physical location;
determining whether an entry in a second electronic data store associates the respective one of the plurality of computer network devices with a physical location, wherein the second electronic data store comprises a plurality of second entries, and wherein each second entry contains information that associates a computer network device with a physical location;
if (a) an entry in the first electronic data store is found that associates the respective medical device controller with a physical location and (b) no entry in the second electronic data store is found that associates the respective one of the plurality of computer network devices with a physical location, then storing, with the plurality of machine learning modules, a new second entry in the second electronic data store, wherein the new second entry associates the respective one of the plurality of computer network devices with the physical location associated with the respective medical device controller in the first electronic data store;
if (a) no entry in the first electronic data store is found that associates the respective medical device controller with a physical location and (b) an entry in the second electronic data store is found that associates the respective one of the plurality of computer network devices with a physical location, then storing, with the plurality of machine learning modules, a new first entry in the first electronic data store, wherein the new first entry associates the respective medical device controller with the physical location associated with the respective one of the plurality of computer network devices in the second electronic data store; and
if (a) an entry in the first electronic data store is found that associates the respective medical device controller with a physical location, (b) an entry in the second electronic data store is found that associates the respective one of the plurality of computer network devices with a physical location, and (c) a comparison indicates that the physical location associated with the respective medical device controller in the first electronic data store differs from the physical location associated with the respective one of the plurality of computer network devices in the second electronic data store, then revising, with the plurality of machine learning modules, the entry in the first electronic data store to associate the respective medical device controller with the physical location associated with the respective one of the plurality of computer network devices in the second electronic data store;
transmitting, by the one or more servers, via the computer network, to a monitoring station, a physical location of one of the plurality of medical device controllers that was learned by the plurality of machine learning modules; and
adjusting, with the one medical device controller, a speed setting or a motor current value of the heart pump communicatively coupled to the one medical device controller, wherein the adjusting of the speed setting or the motor current value is responsive to an alarm being triggered by the operational data extracted from one of the received messages, and wherein the one received message was received from the one medical device controller.
14 . The method of claim 13 , wherein, for each received message, the respective image comprises a serial number, a placement signal, a speed setting, a motor current value, a blood flow rate, and one or more graphs.
15 . The method of claim 14 , wherein the respective image of the one received message further comprises an error message.
16 . The method of claim 13 , wherein the speed setting of the heart pump communicatively coupled to the one medical device controller is adjusted.
17 . The method of claim 13 , wherein the motor current value of the heart pump communicatively coupled to the one medical device controller is adjusted.
18 . The method of claim 13 , further comprising:
adjusting a position of the heart pump communicatively coupled to the one medical device controller, wherein the adjusting of the position is responsive to the alarm being triggered.
19 . The method of claim 13 , further comprising:
restarting the heart pump communicatively coupled to the one medical device controller, wherein the restarting is responsive to the alarm being triggered.
20 . The method of claim 13 , further comprising:
dispatching medical personnel to the physical location of the one medical device controller that was learned by the plurality of machine learning modules, wherein the dispatching is responsive to the alarm being triggered.