Deep Cube invents the future of medical diagnostics with artificial intelligence

A start-up that promises to disrupt the medical sector

A Swiss startup founded in early 2018, Deep Cube is developing artificial intelligence (AI) and embedded AI models for healthcare assistance and diagnostics at breakneck speed: oncology, ophthalmology, neuroscience and biometrics. A true pioneer in its field, the company quickly received the support of the Biopôle, EPFL, Innovaud, HUG and five investors. The team is made up of international prizewinners in artificial intelligence and data science (Top 5 Kaggle, 1st CVPR, 1st MEC) and medical experts. Christophe Patris de Broé, founder & CEO of Deep Cube, opens the doors of their headquarters at the Biopôle Lausanne-Epalinges (VD).

 

Medical artificial intelligence: a real-time diagnostic tool in your pocket

In just six months, Deep Cube has already produced 5 AI models, covering glaucoma diagnosis (in partnership with the Genolier Group, Clinique Montchoisi and the Swiss Glaucoma Research Foundation), breast cancer and mammary tumors, colorectal cancer, diabetic retinopathy (prototype produced for Carl Zeiss Meditech), and soon Alzheimer's disease with HUG and biometric recognition. The technology developed by Deep Cube aims to provide better diagnostic assistance for medical experts, hospitals, clinics and Med/High Tech.

What's more, the startup offers "embedded" artificial intelligence, which merges a computing power model and neural networks on portable chips as small as a USB flash drive. Medical images are stored and analyzed in real time, all in your pocket! These artificial intelligence models are specifically designed to respond to different problems according to customer needs (varied computing power with adapted memory and GPU size) and compatible with all types of portable device.

 

Smarter medical assistance for more reliable, faster and safer diagnosis

RELIABILITY: AI helps medical professionals make the right diagnosis 

Errors in human visual diagnosis are a major concern in medical imaging. The number of misdiagnoses in healthcare is enormous, especially for cancers: doctors and patients want greater precision in medical diagnosis. Artificial intelligence offers improved diagnosis with better sensitivity and specificity (e.g. target tumor delineation). Deep Cube's AI models achieve between +10% and +25% performance in diagnosis, results never achieved before: 97% sensitivity for glaucoma, 99% for breast tumors, 99% for positive diagnosis of diabetic retinopathy, 100% for colorectal tumors and precise localization in less than 1 second.

FAST: AI diagnoses in less than a second

Time is a crucial issue, especially for early detection, as patients don't want to wait 6 or 9 months for Alzheimer's treatment or cancer diagnosis. Deep Cube delivers much better speed with new real-time diagnostics for doctors, hospitals, medical and high-tech equipment manufacturers and patients: cancer is diagnosed in less than 1 second and Alzheimer's disease in 40 milliseconds, instead of 9 months on average, with 20% better results.

SECURITY: Embedded AI is more secure than the cloud

Data confidentiality is a major issue in Europe and the United States with regard to medical confidentiality and the RGPD. Yet the cloud, whether public or private, is now frequently attacked by hackers and is no longer a secure tool. In contrast, with embedded AI, data confidentiality remains at the local level (in the device or can be shared on the local hospital network). No data is transmitted to the cloud, and control of data privacy is returned to the user: doctors and patients.

 

The pioneer in embedded AI technologies announces its A series for Q1 2019

The AI medical diagnostics market is expected to reach $2 billion by 2023, according to Signify Research. Deep Cube is seeking 1.8 million francs by March 2019, for 28% of its open equity capital. The startup aims to invent new technologies (more than 6 new patents and 3 AI methods to patent, doubling AI models per year), such as mobile biometrics or the detection of multiple diseases on the same medical image, which is a revolution in diagnostics.

Medical artificial intelligence is still in its infancy. Its applications are paving the way towards collective intelligence for the benefit of the entire medical community and patients. Thanks again to the Deep Cube team for sharing their projects, vision and expertise with us. The medical sector has a bright future ahead of it with the high performance offered by artificial intelligence.

 

Contact Deep Cube

Website: http: //www.deepcube.ch

Email : chris@deepcube.ch

Tel: 078 631 50 20 

Editor : Sandrine Barrucand

Technical assistance: Marc Bavaud