Calculation of the probability of gradient coil amplifier failure using environment data
The invention provides for a medical instrument ( 100, 300, 500 ) comprising a processor. Execution of machine executable instructions ( 120, 122, 124, 350, 352, 354 ) cause the processor to repeatedly: construct ( 200, 416 ) a measurement vector ( 114, 700 ) comprising multiple data values ( 704 ) using a measurement database ( 116, 122 ), wherein the multiple data values comprise examination room data ( 332 ) descriptive of the environmental conditions of an examination room ( 322 ) of a magnetic resonance imaging system, wherein the multiple data values further comprise technical room data ( 330 ) descriptive of the environmental conditions of a technical room ( 326 ) of the magnetic resonance imaging system; and calculate ( 202, 418 ) a probability ( 706 ) of failure of a gradient coil amplifier ( 312 ) of the magnetic resonance imaging system a predetermined number of days in the future by inputting the measurement vector into a trained neural network program ( 124 ).
1. A medical instrument comprising:
a memory for storing machine executable instructions;
a processor, wherein execution of the machine executable instructions causes the processor to repeatedly:
construct, using a measurement database, a measurement vector comprising multiple data values including examination room data descriptive of environmental conditions of an examination room of a magnetic resonance imaging system and technical room data descriptive of environmental conditions of a technical room of the magnetic resonance imaging system; and
generating a prediction of failure of a gradient coil amplifier of the magnetic resonance imaging system a predetermined number of days in advance by inputting the measurement vector into a trained neural network program to detect fingerprints of failures from the measurement vector.
2. The medical instrument of claim 1 , wherein the medical instrument further comprises the magnetic resonance imaging system, wherein the technical room contains the gradient coil amplifier, wherein the examination room comprises a main magnet of the magnetic resonance imaging system, wherein the technical room comprises a first set of environmental sensors for measuring the technical room data, wherein the examination room further comprises a second set of environmental sensors for measuring the examination room data, wherein execution of the instructions further cause the processor to repeatedly:
acquire the technical room data using the first set of environmental sensors;
record the technical room data in the measurement database;
acquire the examination room data using the second set of environmental sensors; and
record the examination room data in the measurement database.
3. The medical instrument of claim 1 , wherein execution of the instructions causes the processor to train the trained neural network using the measurement database, a historical gradient coil amplifier failure database, or both the measurement database and the historical gradient coil amplifier failure database.
4. The medical instrument of claim 1 , wherein execution of the instructions further causes the processor to store scan parameter data in the measurement database, wherein the scan parameter data is descriptive of the usage of gradient coil amplifier during acquisition of magnetic resonance data, wherein the measurement vector further comprises the scan parameter data.
5. The medical instrument of claim 4 , wherein the scan parameter data comprises any one of the following: number of gradient coil amplifier switching operating, total scan time, average RMS current per scan, peak RMS current per scan, scan protocol, and combinations thereof.
6. The medical instrument of claim 1 , wherein execution of the instructions causes the processor to construct the measurement vector for a predetermined measurement period.
7. The medical instrument of claim 6 , wherein at least one of the multiple data values is descriptive of a measurement value for a specific measurement period different than the predetermined measurement period, wherein execution of the instructions causes the processor to harmonize sampling rates associated with the measurement periods by constructing the least one of the multiple data values by calculating a weighted average.
8. The medical instrument of claim 7 , wherein execution of the instructions causes the processor to construct at least one of the multiple data values by retrieving previous data from the measurement database for a time period that occurred before the specific measurement period.
9. The medical instrument of claim 1 , wherein execution of the instructions further causes the processor to repeatedly:
record failed scan data in the measurement database, wherein the failed scan data is descriptive of one or more occurrence of a failed magnetic resonance image scan by the magnetic resonance imaging system, wherein the measurement vector further comprises the failed scan data in the measurement database, recorded gradient coil amplifier crash data in the measurement database, or both the failed scan data and the recorded gradient coil amplifier crash data in the measurement database, wherein the gradient coil amplifier crash data is descriptive of one or more occurrence of a gradient coil amplifier crash by the gradient coil amplifier, wherein the measurement vector further comprises the gradient coil amplifier crash data.
10. The medical instrument of claim 1 , wherein examination room data comprises any one of the following: examination room temperature, examination room humidity, and combinations thereof.
11. The medical instrument of claim 1 , wherein the technical room data comprises any one of the following: technical room temperature, technical room humidity, gradient coil amplifier coolant temperature, gradient coil amplifier coolant pressure, gradient foil amplifier coolant flow rate, and combinations thereof.
12. The medical instrument of claim 1 , wherein the predetermined number of days is any one of the following: 1 day, 2 days, 3 days, 4 days, 5 days, 6 days, 7 days, greater than 3 days, greater than 4 days, greater than 5 days, greater than 6 days, and greater than 7 days.
13. A computer program product comprising machine executable instructions for execution by a processor controlling a medical instrument, wherein execution of the machine executable instructions cause the processor to repeatedly:
construct, using a measurement database, a measurement vector comprising multiple data values including examination room data descriptive of environmental conditions of an examination room of a magnetic resonance imaging system and technical room data descriptive of environmental conditions of a technical room of the magnetic resonance imaging system;
generating a prediction that warns of failure of a gradient coil amplifier of the magnetic resonance imaging system a predetermined number of days in advance by inputting the measurement vector into a trained neural network program.
14. A method of operating a medical instrument, wherein the method comprises repeatedly:
constructing, using a measurement database, a measurement vector comprising multiple data values including examination room data descriptive of environmental conditions of an examination room of a magnetic resonance imaging system and technical room data descriptive of environmental conditions of a technical room of the magnetic resonance imaging system; and
generating a prediction that warns of failure of a gradient coil amplifier of the magnetic resonance imaging system a predetermined number of days in advance by inputting the measurement vector into a trained neural network program.