Using a set of machine learning diagnostic models to determine a diagnosis based on a skin tone of a patient
Inventors
Swart, Elliot • Alivisatos, Elektra Efstratiou • Ferrante, Joseph • Asai, Elizabeth
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12014828-B2
Publication Date
2024-06-18
Expiration Date
Abstract
Systems and methods are disclosed herein for determining a diagnosis based on a base skin tone of a patient. In an embodiment, the system receives a base skin tone image of a patient, generates a calibrated base skin tone image by calibrating the base skin tone image using a reference calibration profile, and determines a base skin tone of the patient based on the calibrated base skin tone image. The system receives a concern image of a portion of the patient's skin, and selects a set of machine learning diagnostic models from a plurality of sets of candidate machine learning diagnostic models based on the base skin tone of the patient, each of the sets of candidate machine learning diagnostic models trained to receive the concern image and output a diagnosis of a condition of the patient.
Core Innovation
A method and computer program product determine a diagnosis based on a patient’s base skin tone by receiving a base skin tone image and generating a calibrated base skin tone image by calibrating the base skin tone image using a reference calibration profile. The method determines a base skin tone of the patient based on the calibrated base skin tone image and uses that base skin tone to drive downstream model selection.
The method receives a concern image of a portion of the patient’s skin and selects a set of machine learning diagnostic models from a plurality of sets of candidate machine learning diagnostic models based on the base skin tone of the patient. Each candidate set is trained to receive the concern image and output a diagnosis of a condition of the patient.
The method inputs the base skin tone and the concern image into at least one classifier of the selected set of machine learning diagnostic models and receives, as output from the at least one classifier, information from which the diagnosis is derived. The disclosed system includes calibration and image-quality/suitability gating and selection strategies for the set of machine learning diagnostic models based on the patient’s base skin tone to produce a diagnosis.
Claims Coverage
The independent claims are clm-00001, clm-00010, and clm-00017. Across these independent claims, the core inventive structure combines image- or biometric-driven patient attribute determination, selection of sets of machine learning diagnostic models based on that attribute, and producing diagnosis-related output using at least one classifier selected from the chosen set(s).
Calibrated base skin tone determination for diagnosis selection
Receiving a base skin tone image of a patient; generating a calibrated base skin tone image by calibrating the base skin tone image using a reference calibration profile; determining a base skin tone of the patient based on the calibrated base skin tone image.
Attribute-based selection of machine learning diagnostic model sets
Selecting a set of machine learning diagnostic models from a plurality of sets of candidate machine learning diagnostic models based on the base skin tone of the patient, each of the sets of candidate machine learning diagnostic models trained to receive the concern image and output a diagnosis of a condition of the patient.
Classifier input with base skin tone and concern image to derive diagnosis output
Inputting the base skin tone and the concern image into at least one classifier of the selected set of machine learning diagnostic models; receiving, as output from the at least one classifier, information from which the diagnosis is derived.
Computer program product implementing attribute-based calibrated base skin tone diagnosis
A non-transitory computer-readable storage medium containing computer program code for receiving a base skin tone image of a patient; generating a calibrated base skin tone image by calibrating the base skin tone image using a reference calibration profile; determining a base skin tone of the patient based on the calibrated base skin tone image; receiving a concern image of a portion of the patient's skin; selecting a set of machine learning diagnostic models from a plurality of sets of candidate machine learning diagnostic models based on the base skin tone of the patient; inputting the base skin tone and the concern image into at least one classifier of the selected set of machine learning diagnostic models; and receiving, as output from the at least one classifier, information from which the diagnosis is derived.
Attribute-to-model-set diagnosis using attribute and concern image for organ depiction
Receiving biometric data of at least a portion of a patient; inputting the biometric data into a classifier trained to obtain an attribute of the patient; receiving, as output from the classifier, the attribute of the patient; and selecting a set of machine learning diagnostic models from a plurality of sets of candidate machine learning diagnostic models based on the attribute of the patient, wherein each of the sets of candidate machine learning diagnostic models are trained to receive a concern image depicting an organ of the patient in addition to the attribute, and to output the diagnosis of the condition of the patient based on both the attribute and the concern image.
The independent claims cover diagnosing a patient by first determining an attribute using calibrated imaging or a trained classifier, then selecting from multiple sets of machine learning diagnostic models based on that attribute, and finally generating diagnosis-related output using at least one classifier that receives the concern image together with the attribute.
Stated Advantages
Selects sets of machine-learning diagnostic models based on the patient’s base skin tone to produce a diagnosis.
Uses calibrated base skin tone determination based on a reference calibration profile to support diagnosis.
Receives output from at least one classifier from which diagnosis information is derived.
Documented Applications
Automated skin-diagnosis based on a base skin tone image and a concern image of skin, including selection of machine-learning diagnostic models based on base skin tone.
Diagnosis using biometric data to obtain an attribute, and selecting machine-learning diagnostic model sets that use both the attribute and a concern image depicting an organ of the patient.
Interested in licensing this patent?