# US 12,178,549 B2 · Optoacoustic image analysis method and system for automatically estimating lesion traits

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

- Number: US 12,178,549 B2
- Document ID: US12178549B2
- Status: Granted
- Kind: B2
- Application number: 16/445,765
- Filed: 2019-06-19
- Published: 2020-04-16
- Granted: 2024-12-31
- Earliest claimed: 2018-10-11
- Assignees: Seno Medical Instruments, Inc.
- Applicants: Seno Medical Instruments, Inc.
- Inventors: Bryan Clingman; Sandra G. Dykes
- Web page: https://patentgrep.com/patent/US12178549B2

## CPC

- A61B 5/0095: {by applying light and detecting acoustic waves, i.e. photoacoustic measurements}
- A61B 5/4312: {Breast evaluation or disorder diagnosis (optical mammography A61B5/0091)}
- A61B 5/4887: {Locating particular structures in or on the body}
- A61B 5/7267: {involving training the classification device}
- A61B 5/7275: {Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor}
- G06F 18/2411: based on the proximity to a decision surface, e.g. support vector machines
- G06F 18/24147: {Distances to closest patterns, e.g. nearest neighbour classification}
- G06F 18/24155: {Bayesian classification}
- G06F 18/24323: {Tree-organised classifiers}
- G06N 20/00: Machine learning
- G06N 20/10: using kernel methods, e.g. support vector machines [SVM]
- G06N 20/20: Ensemble learning
- G06V 10/764: using classification, e.g. of video objects
- 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
- A61B 5/7425: {Displaying combinations of multiple images regardless of image source, e.g. displaying a reference anatomical image with a live image}
- G06V 2201/03: Recognition of patterns in medical or anatomical images
- G16H 30/40: for processing medical images, e.g. editing
- G16H 50/70: for mining of medical data, e.g. analysing previous cases of other patients
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

A system and method for performing optoacoustic classification prediction is provided. The method utilizes one or more processors in connection with receiving OA feature scores in connection with OA images collected from a patient examination for a volume of interest. The volume of interest includes a lesion. The method applies the OA feature scores to a classification model to obtain a predictive result indicative of a trait of the lesion.

## Claims

### Claim 1

A method for performing optoacoustic (OA) classification prediction, comprising:
utilizing one or more processors for, receiving OA feature scores associated with OA images collected from a patient examination for a volume of interest, the volume of interest including a lesion;
applying the OA feature scores to a classification model to obtain a predictive result indicative of a trait of the lesion, wherein the classification model was built utilizing a predictive machine learning classifier and a data set that includes OA feature scores, malignant labels and benign labels for corresponding lesions in the data set; and
outputting the predictive result.

### Claim 2 (depends on 1)

The method of claim 1, further comprising receiving non-OA feature scores associated with non-OA images collected from the patient examination for the volume of interest, and applying the non-OA feature scores, in combination with the OA feature scores, to the classification model to obtain the predictive result.

### Claim 3 (depends on 1)

The method of claim 1, wherein the non-OA feature score relates to one or more of the following ultrasound features: 1) US Shape Score, 2) US Internal Texture, 3) US Sound Transmission, 4) US Capsular or Boundary Zone, 5) US Peripheral Zone, 6) Patient Age, 7) Mammogram-BIRADS, 8) Lesion Size, or 9) Lesion Posterior Depth.

### Claim 4 (depends on 1)

The method of claim 1, wherein the predictive result is i) indicative of a likelihood that the lesion is in a malignant class or benign class, ii) indicative of a care path decision; and/or iii) a likelihood of malignancy (LOM) designator that the lesion is in the malignant class or benign class.

### Claim 5 (depends on 4)

The method of claim 4, wherein the LOM designator represents a mean confidence interval and wherein the predictive result further comprises a confidence interval range.

### Claim 6 (depends on 5)

The method of claim 5, wherein the classification models represent decision trees that comprise decision points, branches and lesion traits, the applying operation comprising testing the OA feature scores at the decision points and branching through the decision trees based on the testing until reaching one of the lesion traits.

### Claim 7 (depends on 1)

The method of claim 1, wherein the applying operation comprises applying the OA feature scores to an ensemble of classification models, at least a portion of the classification models outputting a corresponding predictive result, the method further comprising combining the predictive results to form a composite predictive result indicative of the likelihood that the lesion is in the malignant class.

### Claim 8

A system for performing optoacoustic (OA) classification prediction, comprising:
memory configured to store program instructions;
one or more processors that, when executing the program instructions, are configured to:
receive OA feature scores associated with OA images collected from a patient examination for a volume of interest, the volume of interest including a lesion;
apply the OA feature scores to a classification model to obtain a predictive result indicative of a likelihood that the lesion is in a malignant class or benign class, wherein the classification model was built utilizing a predictive machine learning classifier and a data set that includes OA feature scores, malignant labels and benign labels for corresponding lesions in the data set; and
output the predictive result.

### Claim 9 (depends on 8)

The system of claim 8, wherein the one or more processors are further configured to receive non-OA feature scores associated with non-OA images collected from the patient examination for the volume of interest, and apply the non-OA feature scores, in combination with the OA feature scores, to the classification model to obtain the predictive result.

### Claim 10 (depends on 8)

The system of claim 8, wherein the predictive result comprises a likelihood of malignancy (LOM) designator that the lesion is in the malignant class or benign class.

### Claim 11 (depends on 10)

The system of claim 10, wherein the LOM designator represents a mean confidence interval and wherein the predictive result further comprises a confidence interval range.

### Claim 12 (depends on 8)

The system of claim 8, wherein the one or more processors are further configured to apply the OA feature scores to an ensemble of classification models, at least a portion of the classification models outputting a corresponding predictive result, the processors further configured to combine the predictive results to form a composite predictive result indicative of the likelihood that the lesion is in the malignant class.

### Claim 13 (depends on 8)

The system of claim 8, wherein the classification models represent decision trees that comprise decision points, branches and lesion traits, the applying operation comprising testing the OA feature scores at the decision points and branching through the decision trees based on the testing until reaching one of the lesion traits.