NLP Logix, under the leadership of Chief Scientist, Matt Berseth, continues to push the boundaries of artificial intelligence in the service of more accurately detecting cancer. In this case, it was a second place finish in the ISIC 2017: Skin Lesion Analysis Towards Melanoma Detection contest, which ended on Wednesday, March 1st.
“This was an excellent contest,” said Matt Berseth, “that gave us the opportunity to again test the infrastructure and team we’ve built here at NLP Logix to quickly train and deploy deep learning models.”
Each year, almost 22 out of every 100,000 men and women will be diagnosed with melanoma of the skin which represents about 1,034,000 in the United States. As with all cancers, it is critical to identify and diagnose a patient early to greatly increase the survival and quality of life of the patient.
“Melanoma is a very image-intensive diagnostic process, so the deep learning team and systems that we have put in place are a very good match,” said Berseth.
As a result of the contest finish, Berseth has been invited to present at the ISBI 2017 conference in Melbourne, Australia on April 18th. Matt will be presenting the technique and results of the competition as well as field questions from the audience.