Frequently Ask Questions

This collective of frequently asked questions is meant to assist in understanding what SkinScreen is and how it can assist in detecting and classifying skin lesions through the power of artificial intelligence. If you have any further questions that are non-proprietary in nature please go ahead and contact us.

1. What is SkinScreen?

The SkinScreen application extends human capabilities in the detection and classification of skin lesions in the support value-based healthcare goals. SkinScreen offers the capability to detect malignant and benign skin lesions in real-time through a highly accurate and precise solution. The solution leverages the power of deep learning, a method under artificial intelligence, to allow quicker and more accurate predictions than previously were available. Currently, the detection is manually performed by a dermatologist or technician through a heuristic approach known as ABCDE (Asymmetry, Border Irregularity, Color, Diameter, Evolution).

2. How is SkinScreen different from other AI-powered skin detection solutions or the prior heuristic approach?

SkinScreen offers a number of differences than other solutions in the market:

1. Ensure user privacy - By leveraging the latest MobileNetV2 architecture the AI model is able to run on a user's device and no images need to be uploaded back to SkinScreen's servers unlike other solutions.

2. Detect whether a skin lesion is present - Many AI skin detection solutions do not detect whether a skin lesion is present in the image initially. They rely upon manual intervention by the human user to provide a skin lesion image. For example, if a user provides an image of a giraffe their solutions will classify the image regardless. SkinScreen's sophisticated AI model is able detect whether a skin lesion is present prior to classification.

3. Detect more classes of skin lesions - By detecting 9 common benign and malignant classes of skin lesions we are able provide better feedback for each individual who interfaces with SkinScreen. And we are continuing to expand the number of skin lesion classes that we support.

4. Provide higher accuracy and precision rates - We are leveraging a two-fold approach to accomplish the higher accuracy and precision rates. We first use a one-class classifier to identify whether a skin lesion is present in the image. If so, then we are able to provide back the 3 most likely skin lesion classes and their associated probabilities. Part of this is accomplished through the 180,000 images that we use to train our AI model.

5. Provide real-time feedback - SkinScreen is able to provide back results to the user in under two seconds on average. By leveraging MobileNetV2 architecture which has lower latency and higher accuracy and few proprietary enhancements we are able to notify the user of the results in a timely manner.

6. Provide user-friendly tools - SkinScreen's different platforms are able to assist users in their interactions with the tool. We try to accomplish this through support tools that are imperative in detecting skin lesions regardless of the user's background and skill sets.

3. How can I be trained on using SkinScreen?

You can access our SkinScreen YouTube channel where we post our video-based tutorials on using the tool.

4. How can I access SkinScreen?

Currently, you can access SkinScreen through two different platforms. The first platform is through the use of this website which is available for free. And you can browse for an image one at a time. The second platform is through the use of our separate application programming interface (API) services that are fee-based. Currently, there are two Representational State Transfer (REST) services, batch and streaming, that are available for access. Both services require an SkinScreen API key. These services are intended for third-party entities who need to process large numbers of images in real-time to support their customer base or integrate into their current information system(s). In the near future we plan to release a SkinScreen App that will be available to operate on Android and iOS devices. With the App you will be able to take a picture of a skin lesion or you can browse your Photo library to locate an image of a skin lesion. Additional supporting services will be available on the App as well. This App will also be fee-based.

5. What type of skin lesions can SkinScreen detect?

There are 9 different classes of skin lesions that SkinScreen can detect:

1. Actinic Keratoses

2. Angioma

3. Basal Cell Carcinoma

4. Dermatofibroma

5. Melanocytic nevus

6. Melanoma

7. Seborrheic keratoses

8. Squamous Cell Carcinoma

9. Vascular lesions

6. How is SkinScreen able to detect skin lesions?

SkinScreen leverages 3 forms of Artificial Intelligence within its statistically based algorithms.

1. Neural Networks allow SkinScreen to sift through data in minute detail, which allows the software to learn to recognize patterns that even the most intelligent humans may overlook,

2. Machine Learning allows SkinScreen to respond without human interaction to specific tasks it handles. Machine Learning algorithms use statistics to find patterns in massive amounts of data. They then use those patterns to make predictions on things like what type of skin lesion is in the image.

3. Deep Learning combines both neural networks and machine learning to analyze vast amounts of data contained within images to recognize lesions and make an accurate prediction, based upon it’s ‘Learning” from over 180,000 verified images that were used to train SkinScreen.

7. What are the benefits and value-added that SkinScreen brings to the table?

In today’s business climate, it is imperative for businesses to leverage some form of AI in order to 1) understand the vast amounts of data that are being ingested into their data lakes and 2) make actionable decisions with that knowledge. AI can bring the equivalent of the human cortex—systems that can learn.

SkinScreen provides that capability to the medical community in order to understand, prioritize, and treat individuals based on their skin lesion condition. And SkinScreen will be a great asset for any medical organization to leverage upon.

8. How much accuracy and precision does the tool have?

SkinScreen provides a high-level of accuracy and precision, above 90%, in fact! What does that mean? Well, accuracy is determined with our patent-pending serial chain AI classifiers. Our binary classifier model detects whether a lesion exists. If the binary class classifier model cannot determine this question, then it returns a “Not enough data exists” solution to the user. If one exists, the tool then will proceed using the multi-class classifier, make a classification, and calculate the probability of being correct. The SkinScreen AI models have been “trained on over 180,000 certified images.

9. Why can the precision exceed human performance?

Within the context of artificial intelligence, the algorithms used in defining a curve fitting solution can detect very minute differences between images that are undetectable by human performance. This allows Skinscreen to make classifications that a trained medical professional would not be able to make.

10. Why does SkinScreen tool run at the edge, near the user, versus within the cloud?

The tool was designed to support the user's privacy by running on the their device. Also, by having the pre-trained AI models close to where the data is generated this would avoid network latency issues that are encountered when the AI models reside on the cloud.

11. What is the plan on adding more types of skin lesions to the tool?

There is no technological limit to the number of classes that SkinScreen can detect. The current limiting factor in adding other classes of skin lesions to the tool is based on the availability of validated datasets for these rare type of skin lesions. Once the data becomes available, SkinScreen, LLC. will proceed to train the models to detect and classify these available lesions. Appropriate testing will then occur to ensure a significant level of confidence is present during the classification process.

12. What cloud architecture does SkinScreen use?

Although SkinScreen runs on the user's device when cached down, there are non-AI related support services (e.g., reCaptchaV3) that SkinScreen uses. To host these services in addition to our third-party support API, SkinScreen leverages the Google Cloud platform. We find their platform, services, and experience to be superior for our company's needs.

13. Who owns SkinScreen, LLC?

SkinScreen is a privatedly held, family-owned business.

14. Can I advertise my product or service on the SkinScreen platform?

Yes, we only accept advertisment requests on our web application. Please contact us for further details.

15. What does the SkinScreen trademark and name signify?

The SkinScreen name was carefully thought out and designed. The "A" on top of the "I" signifies 'artificial intelligence' as being embedded. And "Screen" within the name represents shielding or examining of the skin. Together it represents embedding artificial intelligence to use in examining the skin.