8 applications with artificial intelligence for home diagnosis of moles

Original Source Here

8 applications with artificial intelligence for home diagnosis of moles

Over the past few years, many applications have been created that allow early detection of skin cancer. The smartphone and artificial intelligence help in this.
Some programs send photos to the dermatologist, some provide instant feedback, and others offer useful reminders to self-monitor your skin condition and schedule appointments with your doctor. We decided to test several applications and make up our own opinion about the quality of self-diagnosis.
For the purity of the experiment, the same healthy mole on the left hand will be examined. The device is running on Android 8.1 OS. Evaluation criteria: ease of use, the ability to save research, compatibility with the device. So, let’s go!

Scanoma
This app analyzes moles using AI. The principle of operation is simple – we photograph the mole through the app, give a couple of seconds for processing and get the result. The next step is to send an image of the mole to a dermatologist registered in the partner program. An examination and a diagnosis based on a photo by a doctor costs $24. The interesting thing is that the application gave the same result both on photos with a mole and without it.
You can not view the scan history in the application, but all photos are saved in a separate, automatically created folder.
easy to use – 5 stars;
ability to save research – no;
compatibility with the device-yes.

SmartSkin
The app analyzes moles using AI. Simple interface-we upload a photo with a mole, wait a couple of seconds and get the result.
And the result is as follows: a healthy nevus is 70% normal and 26% melanoma. The app also allows you to save measurements by first marking it on the eerie outline of the human body.
easy to use – 5 stars;
ability to save research-yes;
compatibility with the device-yes.

Model Dermatology
An application that uses artificial intelligence and Wikipedia articles for a versatile approach to mole examination.
Everything works simply – a photo is taken through the app and a few general questions are asked, which you don’t have to answer. Next, the photo with the mole is processed and the result is given.
The photo app tries to identify possible problems and pulls up relevant articles from Wikipedia. The result of the examination – the mole is more than half not specific.
ease of use – 5 stars;
ability to save research – no;
compatibility with the device-yes.

UMSkinCheck
This app was created by researchers from the University of Michigan (UM). The program allows you to pass a self-test of skin cancer of the entire body, as well as create and track the history of moles, growths and lesions.
No artificial intelligence and high-quality photos. You yourself answer the questions compiled according to the ABCDE rule and get a natural result.
In fact, this is an interpretation of the advice-take a piece of paper, draw a mole once a month and then compare the drawings with each other.
easy to use – 3 stars;
ability to save research-yes;
compatibility with the device-yes.

SkinVision
The app’s website claims that it helps in the early detection of melanoma. SkinVision uses deep analysis of skin photos and helps in the early detection of skin cancer. The images are processed using a machine learning algorithm that filters the image layers based on a simple, complex, and more abstract function. This uses a technology called “convolutional neural network” (CNN), and the calculation itself takes less than a minute. And although SkinVision is supported by the scientific council of dermatologists, not all doctors share the optimism. According to a dermatologist from Texas, even an app with outstanding support from scientists has limitations. Dr. Friedmann does not consider it necessary to completely abandon such programs, but advises patients to treat the results with cautious skepticism.
Also interesting is that the app is not supported on Androids from the US.🤷
The prices of the application are quite biting – for the examination of one mole, you need to pay $8.5. If you pay for a year, the examination of moles will cost $61.
ease of use-unknown;
ability to save research-yes;
compatibility with the device-no.

Miiskin
This app uses a map of moles for skin analysis and focuses as a tool for digital dermatoscopy at home. The website of the application claims that with the help of Miiskin, you can detect everything, even the most insignificant changes on the skin that are not subject to the usual look. It’s all about improved digital photography of high quality and the ability to photograph a large area of the body at once.
The developers have made it possible for ordinary consumers to use this technology. The site reports that any owner of a modern iPhone with iOS 10 and later or Android 4.4 and higher can use Miiskin. Meanwhile, Google Play Market gives the result – the app is not suitable for your device.
Miiskin has a 30-day free version, after which it will cost $5.65 a month to monitor moles through the app, according to a support employee’s response.
However, as the developers write in their own blog, the technology of examination using AI is not approved by the FDA and you can rely on such applications solely at your own risk.
ease of use-unknown;
ability to save research-yes;
compatibility with the device-no.

The following 2 applications do not initially make an independent analysis using AI, but are an addition to the devices for skin examination. We will discuss them in more detail.

MoleScope
An application that combines scanoma with sending photos of moles to the doctor and Miiskin with high-quality digital images. However, here the developers went further, and in fact, MoleScope is a magnifying glass that connects to the phone, and the app sends photos to a dermatologist for online verification.
The MoleScope app itself will not analyze or diagnose your moles, but you will be kindly asked to use the ABCDE rule to independently track suspicious moles. The site also provides information that the app works on different versions of Android, but…
It is worth noting that this app is also not supported in Canada.
ease of use-unknown;
ability to save research-yes;
compatibility with the device-no.

NOTA mole tracker
This application comes with a device that examines moles using impedance, that is, it measures the resistance of the cells of the mole to a weak electrical pulse. The same technology is used in fitness trackers for counting calories.

Remark: In 2018, the bioimpedance method was certified by the FDA as a method for non-invasive (without removing a mole) detection of skin cancer.
The device’s website also states that the measurements are safe and suitable for children, the elderly, and pregnant women.
So, download the app, go through the registration procedure and
get to the chatbot. We answer a few questions and go to the screens with the body map, where you can select a mole and enter the size of the nevus. Then we get to the screen with the device connected, and, if you believe the photos from the site, on the following screens we will be able to see the result of measuring the mole.
The description of the device says that it takes 20 seconds to examine one mole, and all measurements are stored on a cloud server. So if you replace your phone, all your moles will remain intact.
Easy to use – 5 stars;
ability to save research-yes;
compatibility with the device-yes.

Instead of output

In a recent 2019 study, researchers at the University of Pittsburgh analyzed four smartphone apps claiming to be able to detect skin cancer. The exact names of the programs are hidden, they pass under ordinal numbers. Three apps used algorithms to notify the owner of the risk of skin cancer, and a fourth app sent photos to a dermatologist.
Unsurprisingly, the researchers found the fourth app to be the most accurate. The other three apps were found to incorrectly classify a large number of skin lesions. However, one of them is missing in almost 30% of melanomas that were identified them as lesions with a low risk to health.
The advantages of applications based on AI is that they raise people’s awareness about the importance of self-diagnosis of moles and can be the first step in the campaign to the doctor, but to rely “on the eye” even a highly intelligent, yet it’s still early.
The situation is different with a device with bioimpedance technology, which was approved by the FDA as a method of non-invasive detection of skin cancer. With such a home diagnosis, you really can not be afraid to miss the moment of malignancy of the mole.
And what to use for home diagnostics of moles-it’s up to you to decide!

AI/ML

Trending AI/ML Article Identified & Digested via Granola by Ramsey Elbasheer; a Machine-Driven RSS Bot

%d bloggers like this: