# US 11,908,568 B2 · System and methods for radiographic image quality assessment and protocol optimization

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,908,568 B2
- Document ID: US11908568B2
- Status: Granted
- Kind: B2
- Application number: 17/077,413
- Filed: 2020-10-22
- Published: 2022-04-28
- Granted: 2024-02-20
- Earliest claimed: 2020-10-22
- Assignees: Canon Medical Systems Corporation; University Health Network
- Applicants: Canon Medical Systems Corporation; University Health Network
- Inventors: Ting Xia; Zhou Yu; Patrik Rogalla; Bernice Hoppel
- Web page: https://patentgrep.com/patent/US11908568B2

## CPC

- G16H 30/20: for handling medical images, e.g. DICOM, HL7 or PACS
- G06N 3/04: Architecture, e.g. interconnection topology
- G06N 3/045: Combinations of networks
- G06N 3/0455: Auto-encoder networks; Encoder-decoder networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06T 7/0014: {using an image reference approach}
- G16H 30/40: for processing medical images, e.g. editing
- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- G16H 50/30: for calculating health indices; for individual health risk assessment
- G16H 50/50: for simulation or modelling of medical disorders
- G16H 50/70: for mining of medical data, e.g. analysing previous cases of other patients
- G06N 3/048: Activation functions
- G06T 2207/10028: Range image; Depth image; 3D point clouds
- G06T 2207/10081: Computed x-ray tomography [CT]
- G06T 2207/30168: Image quality inspection
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

The present disclosure relates to a method for patient-specific optimization of imaging protocols. According to an embodiment, the present disclosure relates to a method for generating a patient-specific imaging protocol, comprising acquiring scout scan data, the scout scan data including scout scan information and scout scan parameters, generating a simulated image based on the acquired scout scan data, deriving a simulated dose map from the generated simulated image, determining image quality of the generated simulated image by applying machine learning to the generated simulated image, the neural network being trained to generate at least one probabilistic quality representation corresponding to at least one region of the generated simulated image, evaluating the determined image quality relative to a image quality threshold and the derived simulated dose map relative to a dosage threshold, optimizing. based on the evaluating, scan acquisition parameters and image reconstruction parameters, and generating, optimal imaging protocol parameters, wherein the optimal imaging protocol parameters maximize image quality while minimizing radiation exposure.

## Claims

### Claim 1

A method for generating a patient-specific imaging protocol using a neural network having been trained to generate at least one probabilistic quality representation corresponding to at least one region of a generated simulated image, comprising:
receiving, by processing circuitry, scout scan data, the received scout scan data including scout scan information and scout scan parameters;
generating, by the processing circuitry, the generated simulated image based on the received scout scan data; scan acquisition parameters, and image reconstruction parameters;
deriving, by the processing circuitry, a simulated dose map from the received scout scan data and the scan acquisition parameters;
evaluating, by the processing circuitry, a determined image quality of the generated simulated image relative to a predetermined image quality threshold and the derived simulated dose map relative to a predetermined dosage threshold; and
generating, by the processing circuitry and based on the evaluating, imaging protocol parameters based on the scan acquisition parameters and the image reconstruction parameters.

### Claim 2 (depends on 1)

The method according to claim 1, further comprising
generating; by the processing circuitry and based on the evaluating, a subsequent generated simulated image based on the received scout scan data, subsequent scan acquisition parameters, and subsequent image reconstruction parameters,
deriving, by the processing circuitry, a subsequent simulated dose map from the received scout scan data and the subsequent scan acquisition parameters,
evaluating, by the processing circuitry, a determined image quality of the subsequent generated simulated image relative to the predetermined image quality threshold and the subsequent derived simulated dose map relative to the predetermined dosage threshold, and
generating, by the processing circuitry and based on the evaluating, subsequent imaging protocol parameters based on the subsequent scan acquisition parameters and the subsequent image reconstruction parameters, the subsequent imaging protocol parameters increasing image quality while reducing radiation exposure.

### Claim 3 (depends on 1)

The method according to claim 1, wherein the determined image quality of the generated simulated image is determined by applying the trained neural network to the generated simulated image, the trained neural network having been trained on scored reference images.

### Claim 4 (depends on 1)

The method according to claim 1, wherein the at least one probabilistic quality representation is based on one or more image quality properties including resolution, contrast, artifacts, and noise.

### Claim 5 (depends on 1)

The method according to claim 1, wherein the trained neural network is at least one of a disease-specific neural network and a body region-specific neural network.

### Claim 6 (depends on 1)

The method according to claim 1, wherein the scan acquisition parameters include x-ray beam energy and tube current.

### Claim 7 (depends on 1)

The method according to claim 1, wherein the image reconstruction parameters include reconstruction method and reconstruction kernel.

### Claim 8

An apparatus for generating a patient-specific imaging protocol using a neural network having been trained to generate at least one probabilistic quality representation corresponding to at least one region of a generated simulated image, comprising:
processing circuitry configured to
receive scout scan data, the received scout scan data including scout scan information and scout scan parameters,
generate the generated simulated image based on the received scout scan data, scan acquisition parameters, and image reconstruction parameters,
derive a simulated dose map from the received scout scan data and the scan acquisition parameters,
evaluate a determined image quality of the generated simulated image relative to a predetermined image quality threshold and the derived simulated dose map relative to a predetermined dosage threshold, and
generate, based on the evaluating, imaging protocol parameters based on the scan acquisition parameters and the image reconstruction parameters.

### Claim 9 (depends on 8)

The apparatus according to claim 8, wherein the processing circuitry is further configured to
generate, based on the evaluating, a subsequent generated simulated image based on the received scout scan data, subsequent scan acquisition parameters, and subsequent image reconstruction parameters,
derive a subsequent simulated dose map from the received scout scan data and the subsequent scan acquisition parameters,
evaluate a determined image quality of the subsequent generated simulated image relative to the predetermined image quality threshold and the subsequent derived simulated dose map relative to the predetermined dosage threshold, and
generate, based on the evaluating, subsequent imaging protocol parameters based on the subsequent scan acquisition parameters and the subsequent image reconstruction parameters, the subsequent imaging protocol parameters increasing image quality while reducing radiation exposure.

### Claim 10 (depends on 8)

The apparatus according to claim 8, wherein the processing circuitry is further configured to
apply the trained neural network to the generated simulated image in order to determine the determined image quality of the generated simulated image, the trained neural network having been trained on scored reference images.

### Claim 11 (depends on 8)

The apparatus according to claim 8, wherein the processing circuitry is further configured to
generate, as the determined image quality of the generated simulated image, at least one image quality metric based on the at least one probabilistic quality representation.

### Claim 12 (depends on 8)

The apparatus according to claim 8, wherein the trained neural network is at least one of a disease-specific neural network and a body region-specific neural network.

### Claim 13 (depends on 8)

The apparatus according to claim 8, wherein the scan acquisition parameters include x-ray beam energy and tube current.

### Claim 14 (depends on 8)

The apparatus according to claim 8, wherein the image reconstruction parameters include reconstruction method and reconstruction kernel.

### Claim 15

A non-transitory computer-readable storage medium storing computer-readable instructions that, when executed by a computer, cause the computer to perform a method of generating a patient-specific imaging protocol using a neural network having been trained to generate at least one probabilistic quality representation corresponding to at least one region of a generated simulated image, comprising:
receiving scout scan data, the received scout scan data including scout scan information and scout scan parameters;
generating the generated simulated image based on the received scout scan data, scan acquisition parameters; and image reconstruction parameters;
deriving a simulated dose map from the received scout scan data and the scan acquisition parameters;
evaluating a determined image quality of the generated simulated image relative to a predetermined image quality threshold and the derived simulated dose map relative to a predetermined dosage threshold; and
generating, based on the evaluating, imaging protocol parameters based on the scan acquisition parameters and the image reconstruction parameters.

### Claim 16 (depends on 15)

The non-transitory computer-readable storage medium according to claim 15, the method further comprising
generating, based on the evaluating, a subsequent generated simulated image based on the received scout scan data, subsequent scan acquisition parameters, and subsequent image reconstruction parameters,
deriving a subsequent simulated dose map from the received scout scan data and the subsequent scan acquisition parameters,
evaluating a determined image quality of the subsequent generated simulated image relative to a predetermined image quality threshold and the subsequent derived simulated dose map relative to a predetermined dosage threshold, and
generating, based on the evaluating, subsequent imaging protocol parameters based on the subsequent scan acquisition parameters and the subsequent image reconstruction parameters, the subsequent imaging protocol parameters increasing image quality while reducing radiation exposure.

### Claim 17 (depends on 15)

The non-transitory computer-readable storage medium according to claim 15, wherein the determined image quality of the generated simulated image is determined by
applying the trained neural network to the generated simulated image, the trained neural network having been trained on scored reference images.

### Claim 18 (depends on 15)

The non-transitory computer-readable storage medium according to claim 15, the method further comprising
generating, as the determined image quality, at least one image quality metric based on the at least one probabilistic quality representation.

### Claim 19 (depends on 15)

The non-transitory computer-readable storage medium according to claim 15, wherein the neural network is at least one of a disease-specific neural network and a body region-specific neural network.

### Claim 20 (depends on 15)

The non-transitory computer-readable storage medium according to claim 15, wherein the scan acquisition parameters include x-ray beam energy and tube current.