Medical device troubleshooting using computational models
Systems and methods for device troubleshooting processes based on computational models are described. In one example, an apparatus may include at least one processor and a memory coupled to the at least one processor. The memory may include instructions that, when executed by the at least one processor, may cause the at least one processor to determine abnormal event information associated with an abnormal event active on the device, and determine, via a computational model: a diagnosis of a device problem based on the abnormal event information, and at least one resolution to resolve the device problem. Other embodiments are described.
1. An apparatus, comprising:
at least one processor; and
a memory coupled to the at least one processor, the memory comprising instructions that, when executed by the at least one processor, cause the at least one processor to:
receive an image associated with a device experiencing an abnormal event,
process the image via an exclusive OR (XOR) process against a plurality of historical images to generate XOR results, wherein each one of the plurality of historical images comprises an associated device problem,
identify one of the plurality of historical images as a match to the image based on the XOR results,
identify diagnosis of the abnormal event based on the one of the plurality of historical images and the associated device problem, and
determine, via a computational model, at least one resolution to resolve the diagnosis of the abnormal event.
2. The apparatus of claim 1 , wherein the instructions, when executed by the at least one processor, further cause the at least one processor to:
train the computational model to determine at least one diagnosis based on historical abnormal event information and historical abnormal event resolution information.
3. The apparatus of claim 1 , wherein the computational model comprises at least one of:
a machine learning (ML) model,
an artificial intelligence (AI) model, or
an expert system.
4. The apparatus of claim 1 , wherein the device comprises a dialysis machine.
5. The apparatus of claim 1 , wherein the computational model is trained to extrapolate at least one diagnosis for undefined abnormal event information.
6. The apparatus of claim 1 , wherein the instructions, when executed by the at least one processor, further cause the at least one processor to:
identify the one of the plurality of historical images as the match to the image based on the XOR results comprising determining whether the XOR results indicate a match for a number of pixels of the image above a threshold number.
7. The apparatus of claim 6 , wherein the threshold number comprises greater than or equal to 80% of the pixels.
8. The apparatus of claim 1 , wherein the image is a screenshot of a graphical user interface (GUI) of the device.
9. A computer-implemented method, comprising, via at least one processor of a computing device:
receiving an image associated with a device with an abnormal event active on the device;
applying an exclusive OR (XOR) process to the image and a plurality of historical images to generate XOR results, wherein each one of the plurality of historical images comprises an associated device problem;
identifying one of the plurality of historical images as a match to the image based on the XOR results;
identifying a diagnosis of the abnormal event based on the one of the plurality of historical images and the associated device problem; and
determining, via a computational model, at least one resolution to resolve the diagnosis of the abnormal event.
10. The method of claim 9 , further comprising:
training the computational model to determine at least one diagnosis based on historical abnormal event information and historical abnormal event resolution information.
11. The method of claim 9 , wherein the computational model comprises at least one of:
a machine learning (ML) model,
an artificial intelligence (AI) model, or
an expert system.
12. The method of claim 9 , wherein the device comprises a dialysis machine.
13. The method of claim 9 , wherein the computational model is trained to extrapolate at least one diagnosis for undefined abnormal event information.
14. The method of claim 9 , wherein identifying the one of the plurality of historical images as the match to the image based on the XOR results comprises:
determining whether the XOR results indicate a match for a number of pixels of the image above a threshold number.
15. The method of claim 14 , wherein the threshold number comprises greater than or equal to 80% of the pixels.
16. The method of claim 9 , wherein the image is a screenshot of a graphical user interface (GUI) of the device.