Remote medical examination
A platform, tips, and otoscope systems are described herein that can aid with evaluation of human ears (specifically children's ears), diagnose middle ear disease, and suggest appropriate treatments. The platform can serve as an end-to-end evaluation, treatment, and delivery of treatment to a user without requiring an office visit, and improves the accuracy of such systems and ease-of-use.
1 . A non-transitory computer-readable storage medium comprising computer-readable instructions that, when executed by one or more processors of a computing device, cause the one or more processors to:
present a medical history questionnaire to a user and receive answers to the medical history questionnaire pertaining to a predetermined status of a middle ear;
based on the answers to the medical history questionnaire pertaining to the predetermined status of the middle ear, prompt the user to take at least one image of an ear canal that includes the middle ear and/or the tympanic membrane;
autonomously classify the middle ear using at least the at least one image taken by the user to indicate a status of the middle ear;
when the at least one image taken by the user is autonomously classified as middle ear infection, output a recommendation to consult with a healthcare provider; and
when the at least one image taken by the user is autonomously classified as normal ear, output a recommendation for no further action needed or treating-at-home.
2 . The non-transitory computer-readable storage medium of claim 1 , wherein executing the computer-readable instructions by the one or more processors to output the recommendation to consult with the healthcare provider when the at least one image taken by the user is autonomously classified as middle ear infection comprises outputting at least both of a middle ear infection diagnosis to the healthcare provider and a recommendation for an antibiotic prescription.
3 . The non-transitory computer-readable storage medium of claim 2 , wherein the healthcare provider is a telemedicine healthcare provider.
4 . The non-transitory computer-readable storage medium of claim 1 , wherein the computer-readable instructions, when executed by one or more processors, further cause the one or more processors to: when the at least one image taken by the user is autonomously classified as middle ear infection, send an antibiotic prescription to a pharmacy.
5 . The non-transitory computer-readable storage medium of claim 1 , wherein the computer-readable instructions, when executed by one or more processors, further cause the one or more processors to: when the at least one image taken by the user does not depict at least a portion of a tympanic membrane, output an insufficient image capture classification.
6 . The non-transitory computer-readable storage medium of claim 5 , wherein the computer-readable instructions, when executed by one or more processors, further cause the one or more processors to: when the at least one image taken by the user does not depict at least a portion of a tympanic membrane, output a recommendation to consult with a healthcare provider along with the insufficient image capture classification.
7 . The non-transitory computer-readable storage medium of claim 1 , wherein the computer-readable instructions comprise a machine learning software component that has been trained at least in part with: (i) one or more images of a tympanic membrane prior to a myringotomy, and (ii) one or more post-myringotomy surgical findings.
8 . The non-transitory computer-readable storage medium of claim 7 , wherein the machine learning software component is trained at least in part with: (i) the one or more images of a tympanic membrane prior to a myringotomy, and (ii) the one or more post-myringotomy surgical findings to configure the machine learning software component to autonomously classify the middle ear using at least the at least one image taken by the user to indicate the status of the middle ear.
9 . The non-transitory computer-readable storage medium of claim 1 , wherein the computer-readable instructions comprise a machine learning software component that has been trained at least in part with images of a same resolution as the at least one image taken by the user.
10 . The non-transitory computer-readable storage medium of claim 1 , wherein the computer-readable instructions comprise a machine learning software component that has been trained at least in part with images that include a rim of an at-home otoscope tip.
11 . A method comprising:
receiving from a user, by one or more programmable processors, answers to a medical history questionnaire pertaining to a predetermined status of a middle ear;
based on the answers from the user to the medical history questionnaire pertaining to the predetermined status of the middle ear, prompting, by the one or more programmable processors, the user to take at least one image of an ear canal that includes the middle ear and/or the tympanic membrane;
autonomously classifying, by the one or more programmable processors, the middle ear using at least the at least one image taken by the user to indicate the a status of the middle ear;
when the at least one image taken by the user is autonomously classified as middle ear infection, outputting, by the one or more programmable processors, a recommendation to consult with a healthcare provider; and
when the at least one image taken by the user is autonomously classified as normal ear, outputting, by the one or more programmable processors, a recommendation for no further action needed or treating-at-home.
12 . The method of claim 11 , wherein outputting, by the one or more programmable processors, the recommendation to consult with a healthcare provider when the at least one image taken by the user is autonomously classified as middle ear infection comprises outputting at least both of a middle ear infection diagnosis to the healthcare provider and a recommendation for an antibiotic prescription.
13 . The method of claim 12 , wherein outputting at least both of the middle ear infection diagnosis to the healthcare provider and the recommendation for the antibiotic prescription comprises outputting at least both of the middle ear infection diagnosis and the recommendation for the antibiotic prescription to a telemedicine healthcare provider.
14 . The method of claim 11 , further comprising:
when the at least one image taken by the user is autonomously classified as middle ear infection, outputting an antibiotic prescription to a pharmacy.
15 . The method of claim 11 , further comprising:
when the at least one image taken by the user does not depict at least a portion of a tympanic membrane, outputting an insufficient image capture classification.
16 . The method of claim 15 , further comprising:
when the at least one image taken by the user does not depict at least a portion of a tympanic membrane, outputting a recommendation to consult with a healthcare provider along with the insufficient image capture classification.
17 . The method of claim 11 , wherein autonomously classifying, by the one or more programmable processors, the middle ear using at least the at least one image taken by the user to indicate the status of the middle ear comprises using a machine learning software component that has been trained at least in part with: (i) one or more images of a tympanic membrane prior to a myringotomy, and (ii) one or more post-myringotomy surgical findings.
18 . The method of claim 17 , wherein the machine learning software component autonomously classifies the middle ear using an association between the at least one image taken by the user to indicate the status of the middle ear and both of: (i) the one or more images of a tympanic membrane prior to a myringotomy, and (ii) the one or more post-myringotomy surgical findings.
19 . The method of claim 11 , wherein autonomously classifying, by the one or more programmable processors, the middle ear using at least the at least one image taken by the user to indicate the status of the middle ear comprises using a machine learning software component that has been trained at least in part with images of a same resolution as the at least one image taken by the user.
20 . The method of claim 11 , wherein autonomously classifying, by the one or more programmable processors, the middle ear using at least the at least one image taken by the user to indicate the status of the middle ear comprises using a machine learning software component that has been trained at least in part with images that include a rim of an at-home otoscope tip.