The research was published in an IBM Research Blog post, which essentially means that when a person finds a worrisome spot on their skin, they can simply use their cell phone’s camera to take a picture and send it to an analytics service available to identify and advice on further steps to be taken.
Although this concept has been applied before, it yielded very bad results. Given that, the IBM team claims they are using more powerful tools which will increase the level of computer image analysis accuracy. This improved technology involves Dermascopes, which will be attached in front of the smartphone cameras and help optimise the spot photos sent for analyzation. The second part consists of creating a holistic database of cancerous skin spots with the help of IBM’s machine learning, computer vision and cloud computing.
This useful product has been the result of the efforts of Memorial Sloan-Kettering Cancer Center and the International Skin Imaging Collaboration (ISIC), but they were quick to point out that it will require at least two more years before the method can become reliable and effective.
According to Codella,
“If you have a lesion, a doctor has a variety of tools to diagnose that lesion. We’re looking to augment [those] with another tool to help catch that disease and to even assist general practitioners to be more helpful for their patients. We’re using datasets for our studies that are open and in the public domain. We evaluated on a test set that is publicly available. People can vet our work and confirm and evaluate based on that public data. This is early work. We’re actively working with our partners to determine that point. We’re not there yet.”
The process still requires to be peer reviewed before being applicable practically, and to speed it up the IBM team has already opened an online site where they will publish a 2017 issue of the IBM Journal of Research and Development, reporting on the method’s success or otherwise.
Do you think this app will be a game changer in the detection of skin cancer? Comment below!