Identification of diseases with portable medical diagnostics instruments that utilize artificial intelligence
Inventors
Moretti, Luke Michael • Digiore, Andrew
Assignees
Publication Number
US-12193746-B1
Publication Date
2025-01-14
Expiration Date
2041-03-24
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
A handheld, portable devices with integrated artificial intelligence (AI) configured to assess a patient's body part to detect a disease and methods of operating such devices are disclosed. In some cases, a device can be a retina camera configured to assess a patient's retina and, by using an on-board AI retinal disease detection system, provide real-time analysis and diagnosis of the patient's retina. Easy and comfortable visualization of the patient's retina can be facilitated using such retina camera, which can be placed over the patient's eye, display the retina image on a high-resolution display, analyze a captured image by the on-board AI system, and provide determination of presence of a disease.
Core Innovation
The invention relates to handheld, portable medical diagnostic devices with integrated artificial intelligence (AI) designed to assess a patient's body part and detect diseases. A specific embodiment is a retina camera capable of evaluating a patient’s retina and using an on-board AI system for real-time analysis and diagnosis of retinal diseases. The device includes a high-resolution display for visualizing the retina, the ability to analyze captured images with the AI system, and provides immediate determination of disease presence.
The problem addressed is the limitation of existing fundus cameras, which often lack portability, on-board AI capabilities, or require external devices and network connectivity to analyze retinal images and produce diagnostic information. This reliance on connectivity can interrupt clinical workflow and is a particular problem in settings where technological infrastructure may be lacking or unreliable. There is a need for improved ophthalmoscopy and retinal image analysis that is self-contained, easy to use, and capable of high-quality diagnosis without cloud-dependency.
The core innovation is a portable, self-contained retinal camera incorporating imaging optics, a display, and an electronic processing circuitry equipped with machine learning models for onboard analysis. The device can irradiate the eye, capture high-quality images, process them with an optimized AI model trained to detect a plurality of retinal diseases, and display diagnostic results in real-time without network connectivity. The system achieves on-device performance through techniques such as shallow neural networks, lightweight model architectures, pruning, and quantization, allowing complex AI models to function efficiently on limited hardware resources.
Claims Coverage
There are three independent claims covering key inventive features of the retinal camera, portable medical diagnostics instrument, and the method of diagnosing using such devices.
Retinal camera with on-board AI disease detection
A retinal camera comprised of: - A handheld housing with a handle. - Integrated power source, display, light source, imaging optics, and image detector all supported by the housing and powered by the power source. - Electronic processing circuitry in the housing that: - Generates at least one image from signals provided by the image detector. - Stores the image in memory. - Interfaces with at least one graphics processing unit (GPU) to process the image using a machine learning model trained to identify multiple retinal diseases. - Provides a determination of the detected retinal disease on the display.
Portable medical diagnostics instrument with AI-assisted multi-disease detection
A portable medical diagnostics instrument comprised of: - A housing that supports a light source for irradiating a patient’s body part, imaging optics, an image detector for capturing images and converting them to electrical signals, and electronic processing circuitry. - The electronic processing circuitry: - Generates at least one image of the body part from the signals. - Stores the image in memory. - Processes the image with a machine learning model trained to identify multiple diseases. - Provides a determination of presence of at least one disease. - A power source to provide power to the components.
Diagnosis method using portable retinal camera and machine learning model
A method for diagnosing a patient using a portable retinal camera which includes: 1. Generating at least one image of the eye by irradiating it with light and sensing reflected light. 2. Processing the image with a machine learning model trained to identify multiple retinal diseases. 3. Providing a determination of the presence of at least one retinal disease. - The method is executed under control of electronic processing circuitry supported by the portable retinal camera.
The claims broadly cover a portable diagnostic platform integrating image capture, on-device machine learning analysis, and real-time disease detection and reporting, specifically for retinal and other body part assessments.
Stated Advantages
Provides on-board, real-time analysis and diagnosis of diseases without the need for network connectivity or external computing devices.
Facilitates easy, efficient, and comfortable visualization and assessment of the patient's retina or other body parts.
Improves accuracy and usability compared to devices that rely on manual analysis or require cloud-based processing.
Maintains patient privacy and data security by processing all information on the device without data transfer for analysis.
Enables use in locations where technological infrastructure and network connectivity are unreliable or lacking.
Allows for better data capture and analysis, enhancing diagnostic sensitivity and specificity.
Can be used in physician offices, clinics, emergency departments, hospitals, telemedicine settings, or at home.
Supports high-quality retinal viewing and image capturing with intuitive, easy-to-use operation.
Documented Applications
Assessment, real-time analysis, and diagnosis of the retina to detect retinal diseases such as diabetic retinopathy.
Use as an otoscope to assess a patient’s ear and provide AI-based disease detection of the ear.
Use as a dermatology scope to assess a patient’s skin and provide AI-based disease detection of the skin.
Deployment in medical settings including hospitals, clinics, emergency departments, telemedicine, and at-home use.
Interested in licensing this patent?