3D Sonic Sensor Skin Diagnostics Using Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current diagnostic tools for skin conditions like melanoma, such as optical coherence tomography and ultrasonography, are expensive and not readily available for early preventive action by the average consumer, leading to unpredictable diagnostics even for experienced dermatologists due to high variability in skin cancer morphology.
Innovation Solution
A computer-implemented method using 3D sonic sensors integrated with smartphones to generate volumetric sonic measurements, processed by machine-learning models for accurate skin condition identification, leveraging beamforming and time-resolution techniques to provide depth information and overcome limitations of lower-resolution transducer arrays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If medical imaging tools such as optical coherence tomography or ultrasonography are used to perform in vivo imaging below the dermis layer, then diagnostic accuracy is improved, but cost and accessibility deteriorate
Solution Approach 1:
The patent creates a simplified copy of professional medical imaging capability using smartphone sensors. The 3D sonic sensor replicates the depth imaging function of expensive ultrasonography equipment, and machine learning models copy the diagnostic expertise of dermatologists, making professional-grade diagnostics accessible through consumer devices.
Solution Approach 2:
The patent replaces complex mechanical imaging systems (optical coherence tomography, ultrasonography equipment) with a smartphone-based system using 3D sonic sensors and computational algorithms. This substitution maintains diagnostic capability while dramatically reducing cost and improving accessibility.
2Ease of operation
If visual analysis of skin lesion surface is used for diagnostics, then ease of operation is improved, but diagnostic reliability deteriorates due to high variability in skin cancer morphology
Solution Approach 1:
The patent transitions from two-dimensional visual surface analysis to three-dimensional subsurface imaging using 3D sonic sensors. This additional dimension reveals depth information and internal structures below the skin surface, providing more reliable diagnostic data while maintaining ease of operation through smartphone-based scanning.
Solution Approach 2:
The patent introduces machine learning models as an intermediary between the raw sonic data and diagnostic conclusions. These models process the complex 3D volumetric data and automatically identify skin cancer features, eliminating the subjectivity and variability inherent in human visual analysis while keeping the system easy to operate.
3Ease of manufacture
If lower resolution transducer arrays are used in consumer devices, then cost and accessibility are improved, but measurement precision deteriorates due to voxel blurring phenomena
Solution Approach 1:
The patent changes the operational parameters of the low-resolution transducer array through beamforming techniques and time-resolution processing. By manipulating the timing and phase of sonic pulses, the system achieves effective resolution beyond the physical limits of the transducer array, overcoming voxel blurring without requiring higher-resolution hardware.
Solution Approach 2:
The patent applies preliminary signal processing operations (beamforming, time-resolution) to the raw sonic data before final analysis. This preliminary action enhances the effective resolution of the data, compensating for the limitations of lower-resolution transducers and reducing voxel blurring effects in the final measurement.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate, cost-effective, and accessible in-field diagnostics of skin conditions like melanoma using consumer-level devices, providing higher clinical accuracy by resolving volume density and overcoming voxel blurring phenomena associated with lower resolution sensors.
Implementation Method 1
generating a three-dimensional (3D) volumetric sonic measurement of a second skin region of the user based at least in part on one or more sonic pulses of the 3D sonic sensor
Implementation Method 2
The machine-learning diagnostic system can provide accurate diagnostics of skin conditions such as skin cancer using sonic fingerprint sensors that generate sonic data that is processed to generate depth information
Data Source
AI summary
Computing systems and methods are provided for detecting skin conditions of humans. A computing device can authenticate a user via a fingerprint scan of a first skin region of the user using a three-dimensional (3D) sonic sensor. The device can generate a three-dimensional (3D) volumetric sonic measurement of a second skin region of the user based at least in part on one or more sonic pulses of the three-dimensional sonic sensor. The device can input data indicative of the 3D volumetric sonic measurement into one or more machine-learning models, generate one or more skin cancer condition identifications associated with the second skin region of the user based on one or more outputs of the one or more machine-learned models, and provide one or more outputs including the one or more skin cancer condition identifications associated with the second skin region of the user.