Patent US5572028A successfully refreshed from Google Patents.
US5572028A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Multi-element dosimetry system using neural network
Assignee: Thermo Fisher Scientific Inc
•
Filing: 1994-10-20
•
Granted: 1996-11-05
•
Expires: 2014-10-20
Abstract
Abstract A dosimetry system and method characterized by use a plurality of radiation sensitive elements to monitor exposure to a radiation field composed of one or more types of radiation at one or more different energies; reading the radiation sensitive elements in a reader after irradiation by the radiation field to obtain element outputs; and supplying the element outputs to a trained neural network computer apparatus wherein the element outputs are analyzed to provide an output indicative of the radiation field.
Inventors
- 👤 Marko Moscovitch
- 👤 Joseph E. Rotunda
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-04 10:58
CPC Classifications
G; G01; G01T; G01T1/00; G01T1/02; G01T1/10; G01T1/11
Claims Analysis
Total Claims: 26
Claim 1
Independent Claim
93 words
A dosimetry method comprising the steps of: using a plurality of radiation sensitive elements to monitor exposure to a radiation field composed of one or more types of radiation at one or more different energies; reading the radiation sensitive elements after irradiation by the radiation field to obtain element outputs; and analyzing the element outputs to determine the radiation field to which the radiation sensitive elements have been exposed, said analyzing step including supplying the element outputs to a trained neural network computer apparatus to provide an output indicative of the radiation field.
Claim 2
Dependent Claim
93 words
A dosimetry method comprising the steps of: using a plurality of radiation sensitive elements to monitor exposure to a radiation field composed of one or more types of radiation at one or more different energies; reading the radiation sensitive elements after irradiation by the radiation field to obtain element outputs; and analyzing the element outputs to determine the radiation field to which the radiation sensitive elements have been exposed, said analyzing step including supplying the element outputs to a trained neural network computer apparatus to provide an output indicative of the radiation field.
Claim 3
Dependent Claim
36 words
A dosimetry method as set forth in claim 1, wherein said neural computer network is composed of at least two layers of nodes connected by a plurality of links each having associated therewith a weighting coefficient
Claim 4
Dependent Claim
24 words
A dosimetry method as set forth in claim 2, wherein the input elements are functionally linked to a succeeding layer of said neural network.
Claim 5
Dependent Claim
27 words
A dosimetry method as set forth in claim 2, wherein the input elements are functionally linked by logarithmic functions to a succeeding layer of said neural network.
Claim 6
Dependent Claim
25 words
A dosimetry method as set forth in claim 2, wherein said element outputs are supplied to said trained neural network as ratios of one another.
Claim 7
Dependent Claim
37 words
A dosimetry method as set forth in claim 1, wherein said output is a correction factor that may be applied to a selected one of said element outputs to obtain a value for a respective dose category.
Claim 8
Dependent Claim
31 words
A dosimetry method as set forth in claim 6, wherein the dose category is selected from the group consisting of deep dose, shallow dose, eye dose, and photon and beta components.
Claim 9
Dependent Claim
37 words
A dosimetry method as set forth in claim 1, further comprising the step of training the neural network by using a computer implemented learning algorithm supplied with a plurality of input-output pairs to obtain the weighting coefficients.
Claim 10
Dependent Claim
37 words
A dosimetry method as set forth in claim 8, including the step of testing the input to see if it is outside the limits of the set of the input-output pairs used to train the neural network.
Claim 11
Dependent Claim
33 words
A dosimetry method as set forth in claim 9, including the step of forcing an input outside the limits of the training set of input-output pairs to the limits of the training set.
Claim 12
Dependent Claim
25 words
A dosimetry method as set forth in claim 1, including the step of displaying the output indicative of the radiation field on a real-time basis.
Claim 13
Dependent Claim
98 words
An dosimetry system comprising plurality of radiation sensitive elements for monitoring exposure to a radiation field composed of one or more types of radiation at one or more different energies; means for reading the radiation sensitive elements after irradiation by the radiation field to obtain element outputs; and means for analyzing the element outputs to determine the radiation field to which the radiation sensitive elements have been exposed, said means for analyzing step including a trained neural network computer apparatus to which the element outputs are applied as inputs to provide an output indicative of the radiation field.
Claim 14
Dependent Claim
98 words
An dosimetry system comprising plurality of radiation sensitive elements for monitoring exposure to a radiation field composed of one or more types of radiation at one or more different energies; means for reading the radiation sensitive elements after irradiation by the radiation field to obtain element outputs; and means for analyzing the element outputs to determine the radiation field to which the radiation sensitive elements have been exposed, said means for analyzing step including a trained neural network computer apparatus to which the element outputs are applied as inputs to provide an output indicative of the radiation field.
Claim 15
Dependent Claim
38 words
A dosimetry system as set forth in claim 12, wherein said trained neural network computer apparatus is composed of at least two layers of nodes connected by a plurality of links each having associated therewith a weighting coefficient.
Claim 16
Dependent Claim
24 words
A dosimetry system as set forth in claim 13, wherein the input elements are functionally linked to a succeeding layer of said neural network.
Claim 17
Dependent Claim
27 words
A dosimetry system as set forth in claim 13, wherein the input elements are functionally linked by logarithmic functions to a succeeding layer of said neural network.
Claim 18
Dependent Claim
26 words
A dosimetry system as set forth in claim 13, including means for supplying said element outputs to said trained neural network as ratios of one another.
Claim 19
Dependent Claim
40 words
A dosimetry system as set forth in claim 12, wherein said output is a correction factor, and means for applying the correction factor to a selected one of said element outputs to obtain a value for a respective dose category.
Claim 20
Dependent Claim
17 words
A dosimetry system as set forth in claim 17, further comprising output means for displaying said value.
Claim 21
Dependent Claim
32 words
A dosimetry system as set forth in claim 18, wherein the dose category is selected from the group consisting of deep dose, shallow dose, total eye dose, and photon and beta components.
Claim 22
Dependent Claim
36 words
A dosimetry system as set forth in claim 12, wherein said neural network computer apparatus was trained by using a computer implemented learning algorithm supplied with a plurality of input-output pairs to obtain the weighting coefficients.
Claim 23
Dependent Claim
23 words
A dosimetry system as set forth in claim 12, wherein the output indicative of the radiation field is displayed on a real-time basis.
Claim 24
Independent Claim
104 words
A method of developing a dosimetry system for extracting radiation field information from a plurality of radiation sensitive elements used to monitor exposure to a radiation field composed of one or more types of radiation at one or more different energies, comprising the steps of: determining an architecture for a neural computer network composed of at least two layers of nodes connected by a plurality of links each having associated therewith a weighting coefficient, and training the neural network by using a computer implemented learning algorithm supplied with a plurality of input-output pairs derived from processing radiation sensitive elements to obtain the weighting coefficients.
Claim 25
Independent Claim
104 words
A method of developing a dosimetry system for extracting radiation field information from a plurality of radiation sensitive elements used to monitor exposure to a radiation field composed of one or more types of radiation at one or more different energies, comprising the steps of: determining an architecture for a neural computer network composed of at least two layers of nodes connected by a plurality of links each having associated therewith a weighting coefficient, and training the neural network by using a computer implemented learning algorithm supplied with a plurality of input-output pairs derived from processing radiation sensitive elements to obtain the weighting coefficients.
Claim 26
Dependent Claim
16 words
A method as set forth in claim 22, wherein the training step includes using back propagation.
References Cited (Backward Citations)
US3176133A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1999-11-05 | 2000-05-06 | 2000-11-05 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2003-11-05 | 2004-05-06 | 2004-11-05 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2007-11-05 | 2008-05-06 | 2008-11-05 | $7700.00 | Window Closed / Lapsed |