MetaOptima Wins the ISIC 2018 Disease Classification Competition!
We are excited to announce that MetaOptima has won first place in the Disease Classification Task of the International Skin Imaging Collaboration (ISIC) 2018 Competition: Skin Lesion Analysis Towards Melanoma Detection!
After many days and nights with countless hours of research (all while dealing with an office move), the entire MetaOptima Family is so proud of the R&D Team’s hard work. In addition to winning first place, we are thrilled to announce that our team also earned second and third place for their other submissions!
The Challenge
The ISIC 2018 challenge consisted of three tasks, the MetaOptima Team focused on the third one “Task 3: Disease Classification”. The objective was to design and build a system that can automatically diagnose an image of a skin lesion as one of the following seven diseases: melanoma, melanocytic nevus, basal cell carcinoma, actinic keratosis, benign keratosis, dermatofibroma, or vascular lesion.
Photo credit: ISIC 2018 Task 3
A set of 10015 images and diagnoses were provided for the training of algorithms, however participants had the option to use any other external data. For the final evaluation, participants made predictions on a held-out set of 1512 images and submitted them to an online system to be scored. The metric used for scoring was the accuracy of the classifier on each of the classes averaged together, referred to as “mean recall” or “balanced accuracy”.
The Impact Of Artificial Intelligence In Dermatology
The outcome of this competition was especially important to the team because it demonstrates how our work positively impacts the healthcare community. By seeing the potential for what we build and how innovation can help empower our doctors to provide better care, our team is inspired to continue their hard work for such a valuable cause.
These results demonstrate clearly that AI offers competent decision support for doctors. The entire MetaOptima Team is grateful for the opportunity to contribute to the development of technologies that will support doctors in the front lines of providing quality patient care. Challenges such as these truly highlight the potential for artificial intelligence in dermatology can be utilized to empower our doctors and care providers to supply patients with better treatment options.
“I’m very proud of what William, Aleksey and myself have accomplished. We spent many hours working towards this task and in last month until the deadline we focused solely on the ISIC challenge. We’re excited to begin the next steps of taking this work from a research prototype to an implementation that dermatologist can begin to access and use. A special thanks goes out to our Technical Team who solved all our issues at light speed.” -- Jordan Yap (Machine Learning Lead)
Next Steps Moving Forward
Excited to contribute to the dermatology ecosystem, every single member of the MetaOptima Team finds purpose in empowering doctors with better tools. Benefits of such improvements include supporting remote physicians where a patients access to expertise in dermatology diagnosis is extremely limited.
By equipping dermatologists with the latest intelligent dermatology software, they are able to provide first-class care for in-need patients while enhancing their patients’ care. Thank you to everyone who has supported us on this amazing journey. We look forward to future opportunities to share the ongoing progress of our work in the joint effort to save lives!
Are you interested in keeping up with MetaOptima latest achievements? Click the links below!
- MetaOptima Technology raises $8.6 million CAD
- A Summary Of MetaOptima's Many 2018 Accomplishments
- MetaOptima: A Vision to Reality Awards BC Applicant
- MetaOptima: a Part of BC's Superclusters
-The MetaOptima Team
Are you interested in seeing how artificial intelligence for dermatology can advance your practice and support your clinical decisions? sign up for a demo today!
Topics: Dermoscopy Artificial Intelligence MetaOptima Artificial Intelligence in Dermatology ISIC2018