Machine Learning (ML) same image different dermatology answer (diagnosis)
Here are two examples of cases that have been verified by dermatologists on the skin diagnosis. However when we upload the images to our machine learning (ML) image recognition algorithm we get different answers when we turn the images around, 90 and 180 respectfully. This is an important observation about our algorithm that we probably do not have sufficient images. However they have all been annotated, so a trick here is to upload the images to our deep convoluted neural network (CNN), with the images turned 90 and 180 degrees.
Get checked and use our FREE artificial intelligence algorithm
Malignant Melanoma was determined by a dermatologist
Get checked and use our FREE artificial intelligence algorithm
We took the original Malignant Melanoma image and rotated it, 90 and 180 degrees.
We ran the original in our AI and it showed, malignant melanoma in 3rd position.
Get checked and use our FREE artificial intelligence algorithm
The Malignant Melanoma image was turned 90
This time Malignant Melanoma is in 2nd position
Get checked and use our FREE artificial intelligence algorithm
Malignant Melanoma image was turned 180 degrees
Now the AI shows a strong correlation with Malignant Melanoma
Get checked and use our FREE artificial intelligence algorithm
SEBORRHEIC KERATOSIS dermatology answer
Get checked and use our FREE artificial intelligence algorithm
SEBORRHEIC KERATOSIS – original image
We ran the original image in our AI and SEBORRHEIC KERATOSIS it showed up in 4th position.
Get checked and use our FREE artificial intelligence algorithm
SEBORRHEIC KERATOSIS image turned 90 degrees
SEBORRHEIC KERATOSIS image turned 90 degrees in the AI, it turns up in 5th position.
SEBORRHEIC KERATOSIS image turned 180 degrees
SEBORRHEIC KERATOSIS image turned 180 degrees in the AI, it turns up in 1st position.
Get checked and use our FREE artificial intelligence algorithm
Ask a Dermatologist
Anonymous, fast and secure!
The Specialist doctor from the University Hospital in Gothenburg, alumnus UC Berkeley. My doctoral dissertation is about Digital Health and I have published 5 scientific articles in teledermatology and artificial intelligence and others.